Citation
Effects of issue involvement and ego-involvement upon acceptance of a counseling message

Material Information

Title:
Effects of issue involvement and ego-involvement upon acceptance of a counseling message
Creator:
Williams, Steven Hugh, 1957-
Publication Date:
Language:
English
Physical Description:
vii, 135 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Assertiveness ( jstor )
Attitude change ( jstor )
Cognitive psychology ( jstor )
College students ( jstor )
Persuasion ( jstor )
Psychological attitudes ( jstor )
Psychological counseling ( jstor )
Psychology ( jstor )
Social psychology ( jstor )
Weak arguments ( jstor )
Counseling -- Psychological aspects ( lcsh )
Dissertations, Academic -- Psychology -- UF
Persuasion (Psychology) ( lcsh )
Psychology thesis Ph. D
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1987.
Bibliography:
Includes bibliographical references (leaves 124-134).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Steven Hugh Williams.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
030387564 ( ALEPH )
16927902 ( OCLC )

Downloads

This item has the following downloads:


Full Text











EFFECTS OF ISSUE INVOLVEMENT AND
EGO-INVOLVEMENT UPON ACCEPTANCE OF A
COUNSELING MESSAGE














By

STEVEN HUGH WILLIAMS



















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY




UNIVERSITY OF FLORIDA 1987
















ACKNOWLEDGEMENTS


I would like to take this opportunity to acknowledge and thank those individuals who have guided and supported me throughout the completion of my dissertation. My first thanks is to the Lord Jesus Christ for the life He has given me during the past six years.

I am grateful to Greg Neimeyer for guiding and encouraging me throughout my graduate studies and especially during the completion of my dissertation. Most of all, I am thankful for his time and his personal interest in me and my professional development. I would also like to thank the members of my committee, Franz Epting, Shae Kosch, Constance Shehan, and Mark Alicke, for their effort and insightful suggestions. It has been my pleasure to work with these outstanding individuals.

I would also like to thank all the people who were an integral part of this dissertation. Dr. Constance Shehan allowed me to utilize her classroom to solicit subjects. A special thanks goes to the undergraduate research assistants who were involved in the technical aspects of this study. Jean Callahan, Lisa Oglander, and John Guy were invaluable in their assistance. April Metzler also gave much time to this project.

ii








I would also like to give a very special

acknowledgment to my wife, Barbara, for her constant love and support and the many sacrifices she has made in her life so that I may pursue my career. Barbara and my son, Adam, have been great sources of support and encouragement throughout this project.















TABLE OF CONTENTS

Page

ACKNOWLEDGMENTS ............... ............ ......... ii

ABSTRACT o. vi


CHAPTERS

I INTRODUCTION ...... ......................

II REVIEW OF THE LITERATURE ... o-.......... 9

The Cognitive Response Approach to Persuasion. 9
The Elaboration Likelihood Model ..... .. 19
Cognitive Processing .............. 25
Objective Processing ....................... 28
Biased Processing .... ... ............... 32
Involvement and Persuasion ... 38
Involvement and Cognitive Processing ........... 44
Involvement Level in Counseling ......... 49
Hypotheses .... ............ .......... 51

III METHODS .................... ..... ... .......54

Research Design ......... .......... 55
Cognitive Responses to Arguments ................ 57
Description of the Sample ..................... 59
Instrumentation ...... ............. ..... ..61
Relationships Grid ......................... 61
Personal Problems Inventory.................. 66
Assertion Survey .......... ..... .... 69
Thought Listing Techniques .. ..... 70
Cognitive Assessment Questionnaire ......... 73
Procedures ................ .............. 73
Analyses of Data .............................. 75
Summary ......................................... 75








iv










IV RESULTS ...................................... 78

Manipulation Checks ..................... 78
Argument Quality .......................... 78
Issue Involvement ..................... 79
Dependent Measures ............................ 80
Cognitive Favorability (CFCR) .............. 81
Attitude Rating ............................ 81
Behavioral Intention Rating ................ 86
Spontaneous Use of Cognitive Restructuring.. 86
Additional Analyses ......... .............. 87

V DISCUSSION ........ .......... .. ........... .... 89

Need for Cognition ............................ 96
Future Considerations ......................... 96


APPENDICES

A RELATIONSHIPS GRID ............................ 101

B PERSONAL PROBLEM INVENTORY .................... 104

C ASSERTION SURVEY .............. 107

D THOUGHT LISTING TECHNIQUE .................... 109

E COVERT REHEARSAL TECHNIQUE ................ 112

F COGNITIVE ASSESSMENT QUESTIONNAIRE- ............ 114

G TRANSCRIPT OF TAPED INSTRUCTIONS ................ 116

REFERENCES. ......................................... .. 124

BIOGRAPHICAL SKETCH ................................... 135

















v
















Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECTS OF ISSUE INVOLVEMENT AND
EGO-INVOLVEMENT UPON ACCEPTANCE OF A COUNSELING MESSAGE

BY

STEVEN HUGH WILLIAMS

May, 1987


Chairman: Dr. Greg J. Neimeyer Major Department: Psychology


Within the general framework of the Elaboration Likelihood Model of persuasion, the present study investigated the influence of issue involvement, egoinvolvement, and argument quality upon the cognitive processing of a counseling message advocating the use of cognitive restructuring in the treatment of assertiveness problems. Issue involvement was measured by the total number of relevant thoughts governed by subjects of the Thought Listing Technique. It was predicted that there would be greater distinction of argument quality under conditions of high issue involvement than under conditions of low issue involvement.

Ego-involvement was measured by the proportion of total variance accounted for by a single construct vi









concerning assertiveness on a modified version of the Role Repertory Grid. It was predicted that there would be greater distinction of argument quality under high egoinvolvement than under low ego-involvement. In addition, it was predicted that under conditions of high issue involvement, high ego-involvement would exert a greater influence on the weak than strong arguments. Under low issue involvement, high ego-involvement would exert an equal influence on both strong and weak arguments.

Results of the 2 x 2 x 2 Analyses of Variance on

outcome expectation attitude ratings, behavioral intention scores', a general cognitive favorability index and covert rehearsal scores did not support the predictions. The analyses revealed significantly higher attitude ratings for the high issue involved group than the low issue involved group. They also revealed significantly greater cognitive favorability for strong than weak argument quality, and a similar trend for attitude rating and argument quality. No significant results were found for behavioral intention and covert rehearsal measures.

These results suggest that message variables strongly determine acceptance of counseling messages. The results of this study were also discussed in relation to the contributions of the Elaboration Likelihood Model to counseling research, and future directions were suggested.




vii















CHAPTER I
INTRODUCTION




Persuasion and resistance to change have been issues of theoretical speculation and empirical investigation throughout the past century (Freud, 1894/1959; Frank, 1961; Ellis, 1985). The extent to which a client yields and/or resists a therapist's persuasive messages can be influenced by a wide range of variables. Since Strong's (1968) initial theoretical conceptualization of counseling as an interpersonal influence process, considerable research has been generated concerning interpersonal influence variables in counseling and counselor power in particular (see Corrigan, Dell, Lewis, & Schmidt, 1980, and Heppner & Dixon, 1981, for reviews). Strong proposed a two-stage model of counseling. In the first stage, counselors attempt to increase their perceived power along the dimensions of expertness, attractiveness, and trustworthiness. In the second stage counselors use their influence to bring about client change. Virtually all of the research conducted so far on counseling as an interpersonal influence process has investigated the three

1










2
above counselor characteristics. Little attention has been paid to the influence of counseling message variables or client characteristics, particularly to those client characteristics which may be associated with resistance to change.

Research within the area of social psychology on attitude change has followed a similar route of development. Early work on attitude change explored communicator variables (such as prestige, expertise, and credibility) message content, and some focus on audience features. However, with the rapid development of a cognitive social psychology in the 1970s, attitude change research expanded to include the consideration of consistency, dissonance, and attributional processes in the communication recipient (Jones, 1985) With this expansion message variables and recipient characteristics were established as major variables in attitude change research (Heppner and Heesacker, 1982) Future research on the interpersonal influence process in counseling needs to more fully integrate the findings from this research in cognitive social psychology.

One social psychological theory that has relevance for counseling research is the Elaboration Likelihood Model (ELM: Petty and Cacioppo, 1981; Petty and Cacioppo, 1986). Petty and Cacioppo refer to elaboration as the extent to which a person thinks about the issue-relevant arguments









3

contained in a persuasion message. Elaboration likelihood would be high if a person is motivated and able to evaluate a message, and it would be low if a person is not motivated and/or not able to evaluate a message. People, of course, vary in their willingness and ability to evaluate messages due to individual and situational factors. The ELM began with Petty and Cacioppols attempts to account for the differential persistence of communication-induced attitude change. After reviewing the literature on attitude persistence, they hypothesized that two distinct routes to persuasion existed (Petty and Cacioppo, 1981). The first route, which they called the central route, involves an individual's thoughtful consideration and personal evaluation of the information presented in the communication. The second route, or peripheral route, involves an individual's selection from the persuasion context of some cue (e.g., counselor trustworthiness) that produces change without the individual having to consider or evaluate the information presented. The central and peripheral routes to persuasion correspond generally to high and low elaboration likelihood, respectively. Petty and Cacioppo found the central route to persuasion to be the more enduring of the two. They further postulate that "variables can affect the amount and direction of attitude change by : (A) serving as persuasive arguments, (B) serving as peripheral cues, and/or (C) affecting the extent









4

or direction of issue and argument elaboration" (Petty and Cacioppo, 1981, p. 132).

From the ELM viewpoint much of the interpersonal

influence research in counseling has focused on peripheral cues such as counselor characteristics. One aim of this study is to expand the focus of interpersonal influence research in counseling to include the study of message and recipient/client variables. In order to provide a context for the study which would approximate to a counseling analogue and provide a framework for the testing of these variables, a life problem that would likely be presented in counseling needed to be identified. The problem chosen concerns social assertiveness. In a needs assessment survey conducted on the Duke University Campus, Talley, Barrow, Fulkerson, and Moore (1983) found that of the 52 needs they asked college students to rate the current importance of the need to assertively stand up for myself was ranked number seven in overall importance. Thus, problems in personal assertion is very likely to be an issue that students will present to college counselors.

In counseling the therapist often presents the client with some task to perform in order to bring about therapeutic change. In the treatment of assertiveness problems, cognitive restructuring has been found to be an effective treatment technique (Kaplan, 1982). If the client acquires a favorable attitude toward the therapeutic








5

task (in this case cognitive restructuring), he/she would more likely actually perform the therapeutic task outside the counseling room. Two variables which could affect the client's attitude toward cognitive restructuring are the quality of the arguments for its use presented by the counselor and the degree of involvement the client possesses toward changing or retaining his/her present level of assertiveness.

In the process of getting a client to try a particular technique, such as cognitive restructuring, the counselor will present any number of arguments for its use. The quality of these arguments will influence the favorability of the client's attitude toward the therapeutic task. In order to provide a means to empirically test the ELM Petty and Cacioppo (1979b) devised a method for developing "strong" and "weak" messages. This method will be used in this study to provide the means to investigate the influence of two different types of subjects involvement upon their cognitive responses toward the therapeutic task of cognitive restructuring.

Because people would have a high level of involvement in a particular life problem before seeking counseling, it would seem that their elaboration likelihood would generally be high. According to the ELM they would thus be more likely to engage in central route processing rather than peripheral route processing. Petty, Cacioppo, and









6

Goldman (1981) provided evidence that under conditions of high involvement a thoughtful evaluation of message content is the most important determinant of attitude change. But is the "involvement" of the laboratory comparable to that brought to counseling or is it more complex? This study will attempt to explore the complexity of personal involvement and its influence upon the cognitive processing of a counseling technique.

In the research on susceptibility to influence,

personal involvement or relevance has been defined as the extent to which the issue under consideration is of personal importance (Petty and Cacioppo, 1979). Variables similar to this definition of personal involvement have been alternately referred to as "issue involvement" (Kiesler, Collins, and Miller, 1969), "personal involvement" (Sherif, Kelly, Rodgers, Sarup, and Tittler, 1973), and "ego-involvement" (Rhine and Severance, 1970; Greenwald, 1980). As mentioned earlier, any person seeking counseling is likely to be highly involved with the problem

or issue presented.

Another factor to consider is that people who are

personally involved with an issue have done some amount of prior thinking on the topic before seeking counseling. It is possible that this information has been organized into a schema or guiding principle for the perception of new information (Fiske and Taylor, 1984). Schema-driven









7
processing tends to be biased toward the maintenance of the guiding schema and may override the objective processing of externally provided communications (Ross, Leper, and Hubbard, 1975; Markus and Sentis, 1982). George A. Kelly's (1955) psychology of personal constructs proposes, as the fundamental postulate of the theory, that "a person's processes are psychologically channelized by the ways in which he anticipates events" (Kelly, 1955, p.46). A person uses his/her personal construct system to interpret and "anticipate" the world. Kelly's organization Collorary suggests that some constructs are more important and have more implications than others. More important or superordinate constructs may serve as schemas for the anticipation of incoming information. They also are more resistant to change (Hinkle, 1966/1965). Thus, if a person has developed a schema or superordinate construct for the dimension of assertiveness, that person's processing of information concerning counseling techniques for treating assertiveness problems may be biased.

Therefore, in this study two types of involvement will be distinguished. The first is issue involvement. This would correspond to the involvement seen in individuals seeking counseling. It concerns the presence or absence of perceived problems in assertiveness. High issue involvement should lead to greater argument elaboration. The second, which will be called ego-involvement, will









8

concern the presence or absence of a well-developed construct for assertiveness. The literature suggests that such a well-developed construct may bias individuals with assertiveness problems against change. Such a process may be one source of resistance in counseling.















CHAPTER II
REVIEW OF THE LITERATURE



There are several areas of literature and empirical research relevant to the current study. First, evidence for a cognitive response approach to attitude change and persuasion will be presented. Second, the Elaboration Likelihood Model, a general cognitive model of persuasion, will be presented. Next, the role of cognitive processing will be further explored. This will include both objective and biased modes of processing. At this point the variable of involvement will be introduced, and its contribution to the attitude change literature will be reviewed. The relationships between involvement and cognitive processing will then be explored, and its applications in a counseling context will be considered. Finally, specific hypotheses will be derived from this discussion.

The Cognitive Response Approach To Persuasion

Whenever people receive a persuasive message or any other communication, they will attempt to place the new information within the context of their existing knowledge of the topic. This is the basic postulate of the cognitive response approach presented by Greenwald (1968). In thinking about the message, the individual may consider

9









10

material not in the message itself. From their prior knowledge and evaluation of the communication the person may generate cognitions that agree, disagree, or are irrelevant to the persuasive message under consideration. To the extent that the message elicits supportive cognitive responses, attitude change in the advocated direction should be facilitated. To the extent that the message elicits negative cognitive responses, attitude change should be inhibited.

The concept that an individual's cognitive responses are an important mediator of attitudes is not a recent development in psychology. Freud's (1900/1939) method of free association was an early attempt at the measurement of cognitive responses and their role as mediators of attitudes in a clinical context. In the early study of attitude change Hovland (1951) suggested that the accurate and complete recording of an audience's thoughts as they listened to a communication would constitute the best method for investigating the internal process of change.

While the cognitive response approach to persuasion was not formally proposed until 1968, researchers have been concerned with cognitive responses in persuasion since the early research of attitude change. In a classic study investigating active versus passive participation in the persuasion process Lewin (1947) compared individual instruction with group discussion. The goal of both











treatments was to increase the consumption of unusual meats (hearts, kidneys, etc.) during World War II. Groups of housewives either listened to a persuasive lecture or participated in a group discussion on the material presented in the lecture. Survey results showed that only

3 percent of the women in the lecture group served one of the meats, while 32 percent of the women in the discussion group served them. While the results of this study are open to different interpretations, it may be that the selfgenerated arguments of the women in the discussion group were more persuasive than the arguments in the lecture.

In later research on the role of active and passive

participation in persuasion, Janis and King (1954) reported that subjects changed their attitudes more in the advocated direction when instructed to give an informal talk on a counterattitudinal topic than did other subjects who passively listened to the talks. In their experiment Janis and King used three different topics. Their results revealed significantly greater attitude change for two of these topics in comparison to the third. From their data and subject interviews they observed that for the two topics with greater change subjects improvised more in their talks. For the third topic subjects stayed close to the prepared outline made available for each topic. In this case improvisation, or self-generated cognition, may have operated as a critical factor in the greater attitude









12

change. In another experiment King and Janis (1956) had subjects either read a persuasive communication to themselves, read it into a tape recorder, or read and then give their own improvised version of the message. Those subjects who improvised changed their attitudes significantly more than subjects in either the oral or silent reading conditions. These results are supportive of the notion that one's own cognitive responses on an issue are the most compelling. In a further test of this hypothesis Greenwald and Albert (1968) had subjects improvise five arguments in response to instructions to advocate either career preparatory or general liberal arts undergraduate education. Subjects also read a set of provided arguments supporting the opposite side. Results indicated that subjects' attitudes tended to be in the direction of the content of their own cognition. moreover, subjects recalled significantly more of their own improvised arguments than provided arguments. They also evaluated their own arguments as more original than those

provided to them.

Another area of research suggesting that a person's own responses are important in mediating persuasion is inoculation theory. McGuire (1964) suggests that resistance to persuasion can be created by providing information and arguments supportive of an individual's original attitude or by "inoculating" the individual by









13

providing weak counterarguments to his or her attitude accompanied by refutations. McGuire states that the inoculation will pose a cognitive threat that will motivate the individual to generate supportive arguments and counterargument refutations for his or her original attitude. This practice should produce greater resistance to subsequent persuasion attempts. The point of interest for the cognitive response approach is, does inoculation lead to self-generated cognitive responses and thus greater resistance to persuasion. McGuire and Papagioris (1961) asked subjects to list thoughts in favor of their initial position one week after being exposed to either supportive defenses or refutational defenses (the inoculation treatment) Subjects who had been exposed to the refutational defenses listed more supportive thoughts than subjects who had been exposed to supportive defenses. In an experimental investigation of the effectiveness of active and passive participation in the defense of one's beliefs McGuire (1964) provided additional evidence for the importance of one's own cognitive responses in mediating attitudinal changes. In the active participation condition he assigned subjects the task of writing either a supportive or refutational defense. In the passive participation condition subjects simply read a defense provided by the investigator. The beliefs used in this study were cultural truisms (e.g., mental illness is not








14

contagious) which people generally have little practice in defending. According to inoculation theory individuals in the active condition would perform more poorly than those in the passive condition when exposed to immediate counterattitudinal attack. However, they should demonstrate increased resistance to attacks which are delayed. His results conformed to this predicted interaction. Active defenses increased in resistance when the attack occurred one week later, while passive defenses declined in resistance except for the passive refutationaldifferent defense. This passive defense, like the active ones, presumably posed a threat that motivated the subjects to generate additional cognitive responses in support of their beliefs.

Further support for the principle that a person's own cognitive responses are important in producing persuasion is offered by studies investigating the relation of attitudes to underlying beliefs and values. In the 1950s and 1960s widespread racial prejudice caused many psychologists to be concerned with changing prevailing attitudes. Two studies designed to assess the impact upon attitudes of changing underlying beliefs that illustrate the importance of cognition were conducted by Carlson (1956) and Stotland, Katz, and Patchen (1959). Carlson (1956) found that in moderately prejudiced college students more favorable attitudes toward racial integration could be









15

produced by making the students aware of how desegregation would help in the attainment of some important goals (e.g., greater American prestige in other countries) Stotland et al. (1959) had college students read a message designed to give insight into the psychodynamics of racial prejudice. They also assigned subjects to various manipulations intended to facilitate the internal restructuring of beliefs (e.g., ordering statements into cause-and-effect sequences). Attitude measures showed no immediate reduction in prejudice, but a significant reduction was found in follow-up measures three to four weeks later. These findings suggest that the restructuring of internal beliefs with subsequent attitude change takes some time to occur.

Thus far early research in support of a cognitive

response approach to persuasion has been presented from three areas: 1) active versus passive participation in persuasion, 2) inoculation theory, and 3) attitudeunderlying beliefs linkages. Cognitive response researchers have also provided empirical evidence for the importance of cognition in producing persuasion. They present four types of evidence for the cognitive response approach based on forewarning manipulations, issue-relevant thinking manipulations, argument quality manipulations, and psychophysiological measures (Cacioppo, Petty, & Stoltenberg 1985).








16

In an experiment utilizing the first approach, Petty and Cacioppo (1977) forewarned some subjects and did not forewarn others that they would be hearing a tape recording prepared by the "Faculty Committee on Academic Affairs" which recommended that seniors be required to pass a comprehensive exam in their major as a prerequisite for graduation. Half of the subjects completed a "thoughtlisting" technique (Cacioppo & Petty, 1981) in the latter

2.5 minutes of the 5 minute interval between the forewarning and the message. Attitude survey results indicated that those subjects who were forewarned, whether or not they listed their thoughts, showed greater resistance to persuasion than those who had not been forewarned. Analyses of the thought-listings showed that the forewarning elicited anticipatory counterargumentation. Forewarned subjects listed twice as many thoughts that were unfavorable than were favorable toward the recommendation. Subjects who had not been forewarned did not generate any issue-relevant thoughts, as they were unaware of the upcoming message's topic.

In the second experiment reported in Petty and Cacioppo (1977), they manipulated issue-relevant thinking. In this study students in an introductory psychology class were told that a guest lecturer, a psychologist from the University Counseling Center, would speak to their class that day, and in return for the visit they would complete a









17

questionnaire he had brought. Half of the students were forewarned that the psychologist would be advocating that all freshmen and sophomores be required to live in campus dorms, a topic which pretesting showed most students to be against. The remaining students were not forewarned of the topic. Half of all of the students were then asked to list their thoughts just prior to the speech. So far this study replicates the conditions of the forewarning study described above. The other half of the students, however, were asked to list all of their thoughts on the topic of requiring students to live in dorms. Thus, half of the subjects were instructed to engage in issue-relevant thinking whether or not they had been forewarned. Attitude measures revealed that unearned subjects who engaged in issue-relevant thinking demonstrated resistance to persuasion equal to that exhibited by the forewarned groups. It appears that just being instructed to think about a topic can produce cognitive responses which support one's beliefs and, subsequently, lead to greater resistance to persuasion.

In the third line of evidence argument quality has been operationally defined such that "strong" arguments elicit more favorable than unfavorable thoughts about a message, whereas "weak" arguments elicit more unfavorable than favorable thoughts concerning the message. In an illustrative study, Cacioppo and Petty (1985) exposed









18

subjects either one or three times to a set of strong or weak arguments in support of the recommendation that senior comprehensive exams be instituted. According to their theory, argument quality should differentiate people's attitudes more after three than after one message presentation. This should occur if it is the nature of people's issue-relevant thoughts rather than the number of different arguments they learn. Results indicate that argument-~quality did differentiate people's attitudes more after three exposures than after one. Subjects also recalled more arguments after three presentations than after one, but the amount of the message learned was unrelated to their attitudes.

The fourth source of evidence comes from the use of psychophysiological measures. Cacioppo and Petty (1979), used integrated electromyographic (IEMG) activity as a measure of subjects' silent language processing. In one experiment, IEMG activity was recorded as subjects sat quietly and as they listened to a tape-recorded message. One minute prior to the message, subjects heard one of three announcements about the message. One group only heard that a message would be presented in one minute; a second group was forewarned that they would hear a message recommending more lenient visitation hours be instituted in dormitories (a proattitudinal message according to pretesting) ; and a third group was forewarned that they









19

would hear a message recommending that stricter visitation hours be instituted (a counterattitudinal message) They found that IEMG activity increased in all conditions during the presentation of the message, which would be expected if IEMG activity indicated the presence of silent language processing. The finding of interest was that IEMG activity had increased following the forewarning of the upcoming counterattitudinal message.

In summary, early research prior to the formal

development of the cognitive response approach to pesuasion supported the importance of people's construing of the information in persuasive messages, rather than their hearing of the material per se, as the most important factor in determining their susceptibility or resistance to persuasion. The methodologies described above have all been used in research on a general framework for understanding persuasion called the Elaboration Likelihood Model (Petty & Cacioppo, 1986), which will be described next.

The Elaboration Likelihood Model

The Elaboration Likelihood Model (ELM) of attitude change is a general response theory, which arose out of Petty and Cacioppo's (1981) attempts to account for the differential persistance of communication-induced attitude change. They hypothesized that two distinct routes to persuasion existed, which they called the central and









20

peripheral routes. The central route involves a person's thoughtful consideration and evaluation of the information presented in the communication; whereas, the peripheral route involves a person's selection from the persuasion context of some cue (e.g., source trustworthiness) that produces attitude change without the person having to thoughtfully consider or evaluate the information presented. Petty and Cacioppo found the central route to be the more enduring of the two routes.

While people are neither universally thoughtful nor universally mindless in evaluating persuasive messages, they do vary in their motivation and ability to consider carefully the information and arguments that comprise a persuasive attempt. Petty and Cacioppo (1981) refer to elaboration likelihood as the degree to which a person thinks about the issue-relevant arguments contained in a persuasion message. If conditions produce or enhance an individual's motivation and ability to engage in issuerelevant thinking, then the elaboration likelihood is considered to be high. Under conditions of high elaboration likelihood a person allocates significant cognitive resources to considering and critically evaluating the presented message. If an individual's motivation or ability to process a message is reduced, then the elaboration likelihood is also decreased. When it is low, a person is likely to forego a careful consideration








21

of issue-relevant information and rely upon positive or negative cues in the persuasion context. A simple but reasonable decision rule could also be used in the persuasion context when elaboration likelihood is low.

In order to provide empirical support for distinct

central and peripheral routes to persuasion and the notion of elaboration likelihood, researchers have used the argument quality manipulation under high and low relevance conditions. In high relevance conditions subjects were led to believe that the issue had direct personal consequences for them, while in low relevance condition subjects were led to believe that the issue had few if any personal consequences. The issues and messages used were easy to understand so all subjects had the ability to think about the information presented. Such a design suggests that subjects in the high relevance conditions should follow the central route to persuasion and subjects in the low relevance conditions should follow the peripheral route.

Petty, Cacioppo, and Goldman (1981) conducted such a study in which college students were exposed to a counterattitudinal appeal (favoring senior comprehensive exams). In addition to personal relevance and argument quality, source expertise was also manipulated in this study. Under conditions of high relevance, students were led to believe that the exam policy would begin next year and thus affect them. Under conditions of low relevance,








22

they were led to believe that the exam policy would begin in 10 years and thus not affect them. For half of the subjects the message was attributed to a Princeton University Professor (high credible source), while for the other half it was attributed to a high school student (low credible source). Source credibility served as a persuasion context cue in this study. Attitude rating results revealed two significant interactions. First, a relevance X message quality interaction showed that argument quality was a more important determinant of persuasion for high than low relevance subjects. Second, a relevance X source credibility interaction suggested that the source cue was a more important determinant for persuasion for low than high relevance subjects. Thus, under conditions of high relevance (high elaboration likelihood), subjects exerted the effort to evaluate the issue-relevant arguments presented. Under condition of low elaboration likelihood, they were persuaded by a context cue and appeared to be unaffected by argument quality.

While in the above study message factors were prepotent under condition of central route processing and source factors were prepotent under conditions of peripheral route processing, the central/peripheral distinction is not one between message and source factors. It refers to whether issue-relevant thinking versus context cues or decision rules leads to attitude change. Different message factors








23

could have differing effects depending upon the level of elaboration likelihood. For example, the quality of the message arguments should have a greater influence when elaboration likelihood is high, while the actual number of arguments could have a greater influence when elaboration likelihood is low. In the latter situation, the person may employ a decision rule such as, "the more arguments the better."

To test these hypotheses Petty and Cacioppo (1984)

conducted two studies. In the first experiment, college students received a message on the issue of instituting senior comprehensive exams. Personal relevance was manipulated as outlined in the Petty, Cacioppo, and Goldman (1981) study described previously. Subjects received one of four messages in support of the exam proposal: 1) three strong arguments, 2) three weak arguments, 3) nine strong arguments, and 4) nine weak arguments. Attitude rating results revealed that number of arguments was a more important determinant of persuasion under low than high relevance but that quality of the arguments was more important under high than low relevance condition. In the second experiment, the message concerned a proposal to increase tuition, but relevance was manipulated by either stating that the proposal was for the student's own university (high relevance) or for a distant university (low relevance). In this study the message contained








24

either three strong argument, three weak argument, or six arguments (three strong and three weak). Results indicate that under high relevance conditions three strong arguments elicited more attitude agreement with the proposal than three weak arguments. The agreement for six arguments was greater than that for the three weak arguments but less than that for the three strong arguments. Under low relevance conditions, six arguments produced the most agreement followed by three strong and finally three weak. Thus, in these two studies argument quantity served as a cue under low relevance conditions, but quality of the arguments was more important under high relevance conditions.

This research suggests that two distinct routes to persuasion do exist and differ in their level of elaboration likelihood. The ELM provides a organizing framework for understanding these major cognitive processes underlying persuasion and how variables relate to these processes. Petty and Cacioppo (1986) present the ELM in postulate form as follows:

1. People are motivated to hold correct attitudes
(p. 127).
2. Although people want to hold correct attitudes, the amount and nature of issue-relevant elaboration
in which people are willing or able to engage to
evaluate a message vary with individual and
situational factors (p. 128).
3. Variables can affect the amount and direction of attitude change by: (A) serving as persuasive arguments, (B) serving as peripheral cues, and/or
(C) affecting the extent or direction of issue and
argument elaboration (p. 132).










25

4. Affecting motivation and/or ability to process
a message in a relatively objective manner can do
so by either enhancing or reducing argument
scrutiny (p. 138).
5. As motivation and/or ability to process
arguments is decreased, peripheral cues become
relatively more important determinants of
persuasion. Conversely, as argument scrutiny is
increased, peripheral cues become relatively less
important determinants of persuasion (p. 152).
6. Variables affecting message processing in a
relatively biased manner can produce either a positive (favorable) or negative (unfavorable) motivational and/or ability bias to the issuerelevant thoughts attempted (p. 163).
7. Attitude changes that result mostly from
processing issue-relevant arguments (central route)
will show greater temporal persistence, greater
prediction of behavior, and greater resistance to
counterpersuasion than attitude changes that result
mostly from peripheral cues (p. 175).

While reviewing the accumulated empirical evidence for

the ELM is beyond the scope of this chapter, relevant

research will be presented in the discussions of cognitive

processing and the variable of involvement.

Cognitive Processing

In the ELM Petty and Cacioppo (1981) propose that two

major cognitive routes, the central and the peripheral, are

used by people to process persuasive messages. The

understanding of these cognitive processes involves the

analysis of the structures and mechanisms that comprise

mental activity as well as the impact of various source,

message, and recipient variables upon mental activity.

Petty and Cacioppo (1986) further propose that variables

can affect the cognitive processing of message arguments so

that argument processing proceeds in a relatively objective









26

or a relatively biased manner. In objective processing, some variable "either motivates or enables subjects to see the strengths of cogent arguments and the flaws in specious ones, or inhibits them from doing do" (p. 136). In biased processing some variable "either motivates or enables subjects to generate a particular kind of thought in response to a message, or inhibits a particular kind of thought" (p. 136). objective processing appears to have much in common with "bottom-up" or "data-driven" processing, since the elaboration is primarily impartial and data focused. On the other hand, biased processing appears to share similarities with "top-down" or "theorydriven" processing, as the elaboration may be guided by prior knowledge, such as a relevant attitude schema (Landman & Manis, 1983; Fiske & Taylor, 1984).

Argument processing lies at the heart of the ELM

(postulate 3) along with two other critical constructs: argument quality and peripheral cues. While peripheral cues are not investigated in this study, argument quality is used as a manipulation to assess the degree to which variables of involvement affect cognitive processing in either a relatively objective or biased manner. In accordance with the ELM, a persuasive message with strong arguments should elicit more agreement when it is considered carefully (high elaboration) than when consideration is low (low elaboration), but a persuasive









27

message consisting of weak arguments should produce less agreement when consideration is high rather than low. By manipulating argument quality along with another variable, it is possible to assess whether that variable enhances or reduces argument processing in a relatively objective or biased manner. In relatively objective processing if the variable increases processing, subjects' attitudes and thoughts should be more clearly distinguished when the variable is present rather than absent. In the case of biased processing a variable will produce varying levels of effects depending upon the direction of the bias and argument strength. A variable that biases thinking in a positive direction should generally have a greater influence on a strong than a weak message, because it will be more difficult for an individual to generate favorable thoughts to weak than strong arguments. On the other hand, a variable that produces a negative bias should have a greater effect on a weak than a strong message. This should occur because it will generally be more difficult for an individual to generate counterarguments to strong than weak arguments. Research utilizing the argument quality manipulation has provided empirical evidence for the effect of numerous variables upon the extent of cognitive processing in both relatively objective and biased manners. A sampling of this research will be presented next.









28

objective Processing

In order to provide support for objective cognitive processing ELM researchers have studied the effects of a number of situational and motivational variables upon the extent of argument processing. These have included 1) distraction, 2) repetition, 3) personal responsibility, 4) need for cognition, and 5) personal involvement. Illustrative studies of the first four variables will be presented next, but research on personal involvement will be presented in a later section.

From the ELM viewpoint distraction is seen as affecting cognitive processing in a relatively objective way. Specifically, the ELM predicts a "thought disruption" hypothesis concerning the effects of distraction upon persuasion. The reasoning underlying this hypothesis consists of the following arguments. Distraction from information processing should enhance persuasion for a message consisting of weak arguments. This should occur because without distraction, weak arguments should generally elicit predominantly unfavorable thoughts. However, distraction should disrupt these unfavorable thoughts and thus enhance agreement. For a message consisting of strong arguments favorable thoughts would generally be predominant, and distraction would disrupt these thoughts leading to reduced agreement.









29

To test this hypothesis Petty, Wells, and Brock (1976) presented college students with either strong or weak arguments for the proposal that their tuition be raised by 20 percent accompanied by a distraction task. Their distraction task consisted of having subjects record the quadrant in which Xs flashed on a screen in front of them. Distraction varied according to the rate of presentation of the Xs with four levels: 1)no distraction, 2) low distraction (15 second intervals), 3) medium distraction (5 second intervals), and 4) high distraction (3 second intervals). After hearing the messages subjects completed attitude measures and listed their thoughts during the message. Attitude rating results revealed a significant argument quality X distraction interaction which is consistent with the thought disruption hypothesis. Under conditions of weak argument quality increasing distraction was associated with more favorable attitudes; however, under conditions of strong argument quality increasing distraction was associated with less favorable attitudes. Analyses of the thought listings showed that high distraction decreased counterargument generation for the weak message but not for the strong one. Moreover, high distraction tended to decrease the amount of favorable thoughts elicited by the strong message but not the weak one. For similar results see Lammers and Becker (1980).









30

Repetition is another variable which, in moderate

amounts, can increase cognitive processing of information in a relatively objective way. once again, Cacioppo and Petty (1985) used strong and weak arguments for the institution of a senior comprehensive exam; this time, however, half of their college student subjects heard the message once, and half beard the message three times in succession. Students showed greater attitudinal differentiation of strong from weak messages when the message was repeated three times rather than just presented once. Similar results on the impact of repetition on persuasion are reported by Petty and Cacioppo (1984).

Personal responsibility has also been found to enhance objective processing of issue relevant arguments. Social psychological research on the "social loafing" effect (Latane & Darley, 1970; Latane, Williams, & Harkins, 1979) suggests the classical finding of reduced personal responsibility in group performance in comparison to individual performance may result from a loss of motivation in the group setting. For the persuasion context this implies that if personal responsibility is increased, then motivation to engage in significant cognitive processing should also increase. In a test of this hypothesis Petty, Harkins, and Williams (1980) asked college students to provide peer feedback on editorial messages ostensibly written by journalism students. Subjects were led to









31

believe that they were either one out of a group of ten people who were evaluating the editorials or the only person responsible for the evaluation. Editorials contained either strong, weak, or a mixture of arguments for the institution of comprehensive senior exams. Attitude results revealed that individual evaluators were significantly more favorable toward the strong message and were significantly less favorable toward the weak message than group evaluators. Responsibility did not affect evaluations of the mixed message. Thus, as personal responsibility increased, argument quality became a more important determinant of the evaluation.

Finally, need for cognition, an individual difference variable, refers to the extent to which people need to structure information in reasonable, integrated ways (Cohen, Stotland, & Wolfe, 1955). Cacioppo, Petty, and Morris (1983) exposed high and low need for cognition college students to either strong or weak arguments for a proposal which would raise their tuition. After hearing the message, subjects were asked to evaluate the message and to give their personal opinion about the issue. Results indicate that high need for cognition students produced more polarized evaluations and attitudes for the strong and weak messages. Furthermore, a significantly higher correlation between argument evaluation and personal opinion was found for the high (r = .70) than the low









32

(r .22) need for cognition group. This would be expected since high need for cognition subjects should be more likely to deduce their attitudes from a careful consideration of the central arguments in the proposal.

it is now apparent that numerous variables can affect people's motivation and/or ability to process persuasive arguments in a relatively objective manner. As noted earlier, however, variables can also affect cognitive processing in a relatively biased way. Biased Processing

Studies of the impact of forewarning and prior

knowledge have provided evidence within the ELM framework for the relatively biased cognitive processing of persuasive messages. Forewarning has been seen to affect an individual's motivation to process arguments in a biased manner, whereas prior knowledge generally affects an individual's ability to engage in biased processing. Forewarning can be further distinguished as either warning of message content or warning of persuasive intent.

In a study of the forewarning of message content, Petty and Cacioppo (1977) manipulated argument quality, forewarning, and also issue-relevant thinking. Details and results of this study were presented earlier in the section on ELM research in support of the cognitive response approach. Briefly, if subjects engage in issue-relevant thinking, they resisted the message whether they were









33

forewarned or not. These results support the notion that it is not so much the warning but rather the accessing of attitude-supportive cognitions and the subsequent processing of the persuasive message in light of these cognitions which facilitates resistance. It appears that the forewarning served to motivate subjects to begin thinking about their beliefs and the information and thoughts that they already had to support those beliefs. In line with this reasoning, a content forewarning is more effective when there is some time delay between the warning and message to allow thinking (Hass & Grady, 1975; Petty &

Cacioppo, 1977) .

Hass and Grady (1975) also report that a forewarning of persuasive intent is just as effective when it immediately precedes a message as when it comes several minutes before a message. This suggests that an intent warning functions differently from a content warning. They propose that an announcement of an intent to persuade may arouse a psychological state of "reactance" that motivates a person to defend their freedom to hold a particular attitude (cf. Brehm, 1972). Such a state could produce biased processing. In a study designed to explore whether warnings of persuasive intent produced biased processing Petty and Cacioppo (1979b) told students that they would be evaluating radio editorials. Some students were given a warning that the editorials were intended to persuade









34

college students, while others were just told that the editorials were a journalism class project. Personal relevance was also manipulated in this study by telling the students that certain college regulations would change next year (high relevance) or ten years from now (low relevance). The editorial message used with all subjects consisted of five strong arguments for the institution of senior comprehensive exams as a requirement for graduation. Attitude measure results revealed a main effect for warning and a warning X relevance interaction. The interaction showed that warning significantly decreased attitude agreement only under the high relevance condition. Since the warning reduced agreement even though the arguments used were strong, this suggests that it induced biased rather than objective processing. As this effect was stronger for high than low relevance conditions the warning did not appear to function as a simple rejection cue but rather as a motivator for counterargument and resistance to change.

The common thread that runs through the operation of both types of forewarning effects is the presence of some organized structure of prior knowledge within the recipient of the persuasive attempt. In the social psychological literature a person's organized structure of knowledge is referred to as a schema (Landman & Manis, 1983; Markus & Zajonc, 1985). Markus and Zajonc (1985) state that the









35

processing of persuasive messages "may be seen as consisting of schema formation or activation, of the integration of input with these schemes, and of the updating or revision of these schema to accommodate new input" (p. 150). While it is possible that schemes would enable a person to process messages more objectively, the literature suggests that when processing is guided by a schema, it tends to be biased toward the perseverance of the existing schema (Ross, Lepper, & Hubbard, 1975; Crocker, Fiske, & Taylor, 1984; Fiske & Taylor, 1984). Thus, if a person has a great deal of prior knowledge, or a well-developed schema, he or she would tend to be able to more effectively counterargue messages opposing their initial attitudes and to cognitively argue for messages supporting their initial attitudes (Lord, Ross, & Lepper, 1979).

It should be noted that a schema as currently defined within the literature is conceptually similar to George A. Kelly's (1955) theorizing on personal constructs. Like the schema theorists he sees people as information processors interested in the organization of knowledge and prediction of future events. His Fundamental Postulate states that "a person's processes are psychologically channelized by the ways in which he anticipate events" (p. 46). In addition, the Organization Corollary states that, "each person characteristically evolves, for his convenience in









36

anticipating events, a construction system embracing ordinal relationships between constructs" (p. 56). Constructs are an individual's basic interpretations of the world and are bipolar in that they describe the way in which certain things are perceived as alike and different from others (Bannister and Mair, 1968) Constructs which are superordinate in a person's system subsume other constructs, have relatively more implications, and are generally based on more information. Thus, they function similarly to schemas.

In an investigation of the effects of prior knowledge

and schemas upon cognitive processing, Cacioppo, Petty, and Sidera (1982) provide evidence that prior knowledge does influence the processing of persuasive messages. Subjects who characterized themselves using trait adjectives as either "religious" or "legalistic" people were exposed to strong or weak proattitudinal messages, which were either schema-congruent or schema-incongruent. After presentation of the message subjects rated its persuasiveness and listed their thoughts. Results suggested that with proattitudinal messages subjects who received schema-relevant information were more positive about the quality of the communication's arguments and in their listed thoughts. Thus, it appears that self-schemas influence cognitive responses in a biased or top-down fashion.









37

With counterattitudinal message it would be expected that prior knowledge would increase a person's ability to counterargue the message. To test this hypothesis Wood (1982) divided subjects into high and low prior knowledge groups for the topic of environmental preservation. One to two weeks later subjects read a counterattitudinal message consisting of four arguments against environmental preservation. Attitude measure and thought listing revealed the subjects who had high prior knowledge changed less in the advocated direction than did subjects with low prior knowledge. In addition, high prior knowledge subjects generated more counterarguments and fewer favorable thoughts toward the message.

Wood, Kallgren, and Priesler (1985) extended this finding by adding argument quality and message length manipulations. They divided subjects into high, medium, and low prior knowledge about environmental preservation groups and exposed them to one of four persuasive messages. Two messages consisted of three strong arguments against preservation, and two messages consisted of three weak arguments against preservation. In the message length manipulation two versions of each type of argument were developed. one version contained short concise statements, while the other contained more wordy versions of the same arguments. Both versions were equal in argument strength and ease of comprehension. Attitude measure results reveal









38

that high prior knowledge subjects were overall more resistant to the different messages than low prior knowledge subjects and that this effect was stronger for weak than strong arguments. This result is consistent with the notion that prior knowledge would enhance a person's ability to counterargue an incongruent message. With weak arguments it would be easier to counterargue than with strong arguments. In addition, low prior knowledge subjects' attitudes were affected by argument length, but high prior knowledge subjects' attitudes were not.

In sum, research on forewarning and prior knowledge has provided support for the view that when a person has a well-developed level of organized knowledge on a topic, message processing will be biased. This will occur, because prior knowledge enables the counterarguing of incongruent messages and the strengthening of congruent ones. The presence of schemes or superordinate constructs are indicators of such well-developed levels of prior knowledge.

Involvement and Persuasion

In the preceding sections numerous variables (e.g.,

need for cognition, forewarning) which may affect cognitive processing in either a relatively objective or relatively biased manner have been discussed. one of the most important variables which can affect the extent and type of cognitive processing employed is personal involvement. In









39

past psychological research similar variables have been alternately referred to as "issue involvement" (Kiesler, Collins, and Miller, 1969), "personal involvement" (Sherif, Kelly, Rodgers, Sarup, and Tittler, 1973), and "egoinvolvement" (Greenwald, 1980). This has occurred because involvement can be judged in a variety of ways, such as the degree to which an issue has "intrinsic importance" (Sherif and Hovland, 1961) the centrality of the issue to a person's self, and the number of personal consequences of the issue.

Involvement became a central variable in research on persuasion with the advent of social judgment theory (Sherif and Hovland, 1961; Sherif, Sherif, and Nebergall, 1965). Social judgement theory views persuasion as a twostage process. First, one makes a judgment about the position of a persuasive message in relation to one's own position. Second, attitude change or no change occurs depending upon the assimilation or contrast effects evoked by the judged discrepancy between the message and one's own position. Central to the theory are two assumptions: (1) an individual's own stand on an issue serves as an internal anchor for judging messages, and (2) the more "involved" an individual is in the issue, the stronger the anchoring effects of the initial opinion. Greater involvement thus leads to more resistance to persuasion, because people were postulated to hold broader "latitudes of rejection" as









40

involvement increased. For Sherif and Hovland, involvement seemed to refer to both the intensity with which an attitude is held and the importance of that attitude for the self-identity.

Sheriff and Hovland (1961) described two field studies in support of social judgment theory. For illustrative purposes the "prohibition study" (Hovland, Harvey, and Sheriff, 1957) will be presented here. The prohibition study was conducted in Oklahoma shortly after a close referendum which favored prohibition in the final outcome. To insure that subjects were deeply involved with the issue, dry-stand subjects were recruited from Women's Christian Temperance Union groups, the Salvation Army, and strict denominational colleges. Wet-side subjects were selected from acquaintances of the experimenters. A group of moderate-stand subjects were also included in the study. Subjects were exposed to a 15 minute tape recorded message, which presented either a wet, moderately wet, or dry stand on prohibition. Dependent measures were obtained on the following: (1) the communication's estimated position, (2) subject's reactions to the communication in terms of fairness and impartiality, and (3) subject's preferred position and latitudes of acceptance and rejection.

In support of social judgment theory, Hovland, Harvey, and Sherif (1957) found that for the majority of the subjects with initially extreme stands, their latitudes of









41

rejection were broader than their latitudes of acceptance. Furthermore, for the majority of the moderate subjects, the width of their latitudes of acceptance exceeded the width of their latitudes of rejection. Thus, it appears that the two groups of subjects use very different reference scales as internal anchors.

Results for the dependent measures reflect the functioning of these different internal anchors. In judging the speaker's position, extremely dry subjects saw the message as advocating a much wetter position than it did, while extremely wet subjects judged it as advocating a drier position than it did. Moderate subjects were more accurate in judging it as advocating a slightly wet position. Moreover, the closer the communication was to the respondent's own position, the more likely that subject was to favorably evaluate the message in terms of fairness and impartiality. Attitude change results indicated that in comparison to moderate subjects, approximately twice as many extreme subjects remained unchanged by the message. Greater involvement thus leads to more resistance to persuasion.

Other early research has also found that increasing involvement was associated with resistance to persuasion. Miller (1965) found that high issue involvement consistently decreased the persuasive effect of a discrepant communication on attitude measures, but









42

involvement level did not affect subjects' latitudes of acceptance. According to social judgment theory, high involvement should reduce latitudes of acceptance. Eagly and Manis (1966) also report that highly involved subjects react more negatively toward persuasive messages that

contradict their beliefs.

other researchers, however, have not found results consistent with the notion that higher levels of involvement increase resistance to persuasion. Zimbardo (1960) defined involvement in terms of an individual's concern with the social consequences of his or her response in a given situation rather than the intrinsic importance of an issue for the individual. He found greater attitude change with highly involved subjects than with less involved subjects. Freedman (1964) also operationalized involvement in terms of concern about a response. In addition, he exposed subjects to messages which were slightly, moderately, or extremely discrepant from their initial positions. His results indicated that under low involvement, there was more change with greater discrepancy; but under high involvement, their relationship was nonmonotonic with maximum change occurring at moderate discrepancy. Moreover, under moderate discrepancy, both high and low involved subjects changed by a similar amount. Similar results were reported by Rhine and Sev erance (1970).









43

Contradictory results have also been reported by

researchers who have manipulated forewarning along with involvement. Apsler and Sears (1968) either forewarned or did not forewarn subjects that they would read a proposal calling for the replacement of professors by supervised teaching assistants. Involvement was manipulated by informing half of the subjects that the proposal would go into effect ten years from now. Results indicated a significant interaction with forewarning inhibiting attitude change under high personal involvement but facilitating change under low personal involvement. Dean, Austin, and Watts (1971) found, however, that forewarning inhibited attitude change for both high and low involvement issues and its effect was even more pronounced for the low involvement issue. On the other hand, Petty and Cacioppo (1979b) found the inhibiting effect of forewarning to be greater under high than low involvement conditions. In these latter two studies the forewarning given was one of persuasive intent not content, but all three studies manipulated involvement in similar ways.

one interesting finding in the Petty and Cacioppo

(1979b) study was that when no forewarning was given, high involved subjects tended to show more attitude change than low involved subjects. Similar findings were also reported by Apsler and Sears (1968) and Eagly (1967). Eagly (1967) gave subjects favorable or unfavorable discrepant









44

information about either themselves (high involvement) or another person (low involvement). She found that when favorable information was provided, high involvement subjects showed more attitude change than low involvement subjects. When unfavorable information was provided, the reverse was true.

In sum, research has produced contradictory findings on the effects of level of involvement. To some degree this has resulted from differing definitions of involvement, but the contrary results have persisted even when involvement was similarly defined. In hopes of reconciling these opposing results, involvement will now be examined in relation to cognitive processing.

Involvement and Cognitive Processing

From the ELM perspective Petty and Cacioppo (1979a)

propose that increasing involvement with an issue increases one's motivation to cognitively process issue-relevant information and can lead to either increased or decreased persuasion. The research presented in the previous section clearly demonstrates that high involvement can lead to either enhanced or inhibited persuasion. Petty and Cacioppo further postulate that whether persuasion is enhanced or inhibited depends upon the individuals initial opinion. If the persuasive message is contradictory to the individual's initial attitudes, it is likely that the individual is motivated and able to generate









45

counterarguments to the message. Therefore, as issuerelevant thinking increases, counterargumentation and resistance to persuasion also increases. On the other hand, if the persuasive message is congruent with the individual's initial attitude, it is likely that the individual is motivated and able to generate favorable cognitions. As involvement and issue-relevant thinking increases in this situation, more favorable thoughts might be generated, and increased persuasion would result.

In order to test these hypotheses Petty and Cacioppo

(1979a) conducted two experiments. In the first experiment subjects were exposed to either a proattitudinal or counterattitudinal message concerning coed visitation hours. Issue involvement was manipulated by stating that the changes in visitation hours would go into effect at their own university (high involvement) or at a distant university (low involvement). Dependent measures consisted of attitude scales and thought listings. Results on the attitude index revealed a significant interaction between involvement and type of message. Analysis of the interaction revealed that increased involvement increased agreement with the proattitudinal message but decreased agreement with the counterattitudinal message. The thought listing technique revealed similar results. Under high involvement conditions subjects produced more positive cognitions and fewer counterarguments to the proattitudinal









46

communication than to the counterattitudinal one. Message direction showed no significant effects on cognitive responses under low involvement conditions.

In the second experiment Petty and Cacioppo (1979a) again varied involvement, but they presented only a counterattitudinal message with either strong or weak arguments. The message in this experiment concerned the institution of senior comprehensive exams. The ELM predicts that increased involvement will lead to more issue-relevant thinking about the arguments presented in the message. For the message with strong arguments, which are difficult to counterargue, increased involvement should be associated with more persuasion. For the message with weak and easy-to-counterargue arguments, increased involvement will be associated with decreased persuasion. While high involvement may initially motivate an individual to reject a counterattitudinal message, the enhanced cognitive processing of the message should enable the virtues and flaws of the arguments to be objectively recognized. Results of the cognitive response measures supported these hypotheses. Under high involvement conditions subjects produced more favorable cognitions and fewer counterarguments to the strong than to the weak arguments. Argument quality produced no significant effects on cognitive responses under low involvement conditions. Furthermore, high involvement increased the










47

amount of counterarguments generated to weak arguments and increased the amount of favorable thoughts generated to strong arguments. These results show that increased involvement increases the distinction of strong and weak arguments.

This interaction of personal involvement and argument quality has been replicated several times (Petty, Cacioppo, and Goldman, 1981; Petty, Cacioppo, and Heesacker, 1981; Petty and Cacioppo, 1984) For example, in the context of examining the effects of source expertise under conditions of high and low involvement, Petty, Cacioppo, and Goldman (1981) found that under high conditions objective cognitive processing increased distinction of argument quality and that argument quality was the primary determinant of persuasion. under low involvement conditions, the peripheral cue of source expertise exerted a greater effect than argument quality.

Further support for the notion that increasing

involvement heightens message scrutiny was provided by studies in which involvement was manipulated along with source and message cues. Results of Chaiken's (1980) work indicate that high issue involvement subjects tend to use a systematic information processing strategy, whereas low involvement subjects tend to use a heuristic processing strategy. Subjects who use a systematic strategy focus on









48

the content of persuasive messages and carefully evaluate the content in a relatively objective manner.

Although the research cited above suggests that people become more likely to evaluate carefully and objectively the issue-relevant arguments in a persuasive message as personal involvement increases, circumstances may occur in which cognitive processing becomes biased as involvement increases. As suggested in the section on biased processing, the presence of a well-organized structure of prior knowledge may lead to biased processing. An individual who is involved with an issue has done some amount of prior thinking about the pool of issue-relevant arguments. Some individuals may have done a great deal of prior thinking and have a well-developed construct system or schema for the issue in question. Such individuals may have a greater ability to counterargue persuasive messages, a greater store of supportive arguments for their own beliefs, and/or little motivation to consider yet another persuasive appeal.

Ostrom and Brock (1968) suggest that when an issue is intimately connected with an individual's central values, its personal involvement may be intense enough to generate biased processing. Greenwald (1980) proposes that the ego is characterized by a cognitive bias of conservatism or resistance to change that serves to protect the self's organization of knowledge. If the ego is threatened, it









49

will respond with cognitions aimed at preserving the status quo. Thus, extremely high levels of involvement, or personal importance for a person's central aspects of identity, may also lead to biased processing. Biased processing may then consist of either a negative bias for counterattitudinal messages or a positive bias for proattitudinal messages.

Involvement Level in Counseling

Clear cases of high or low involvement probably do not exist in the therapy room. People generally seek help with issues in which they are involved to some significant degree. Prior to pursuing therapy, they will have given thought to how to deal with the issue and perhaps will have tried many self-generated change attempts. They may even have sought the advice of friends, family, and others.

Strong (1971) suggested that different types of

counseling may elicit different levels of involvement from clients, but it seems clear that clients may also bring different levels of involvement with them to counseling. Stoltenberg and McNeill (1984) showed that college students who were undecided about a career regarded the issue of career exploration as more personally involving than students who had already chosen a career. They also found undecided students agreed more with a message advocating a career exploration course than did decided students.









50

This finding suggests that more involved clients may see counseling as more beneficial and may attend more to the counseling process. Thus, clients with a high level of involvement should adhere more to central route cognitive processing of therapeutic messages, whereas clients with a low level of involvement may depend upon the peripheral route. Moreover, as central route processing appears to lead to more enduring attitude change (Petty and Cacioppo, 1981), it behooves the therapist to motivate clients to evaluate carefully the pros and cons of their present behaviors and to weigh the costs and benefits of change and various means of change.

An example of a highly involved client perhaps would

be the articulate, well-educated, and young business person seeking counseling concerning improving his or her relationships with co-workers and superiors. One would expect this person to consider carefully the issues and suggestions in counseling. The low involved client may be represented by the person forced to attend an alcohol therapy group due to a DWI conviction. He or she may rather be at home or even just spend the weekend in jail than attend therapy weekly for 3 months. Such an individual may elaborate very little on the issues raised in counseling but may be influenced by persuasion cues (e.g., counselor social attractiveness).









51

On the other hand, client factors may also serve to influence cognitive processing in a biased manner. An individual may be so extremely involved in an issue that he or she avoids thinking about it altogether (e.g., the denial of a person with alcohol dependence). Processing may also become biased in the service of one's own ego or for self-protection (e.g., Greenwald, 1980). If a person organizes his or her identity around being a certain type of individual (e.g., an easy to get along with, always ready to please, nonassertive person), he or she would probably be less objective in considering therapeutic messages designed to change that behavior. This would especially be the case if this identity had provided the individual with rewarding secondary gains in the past. Such a client may be well prepared to generate cognition in support of his or her present attitudes and behaviors and actively counterargue any therapeutic communications aimed at changing that basic identity. Thus, this biased processing would lead to greater resistance to change.

Hypotheses

As outlined in the preceding sections of this review, cognitive processing of persuasive messages may follow either a central information-processing route (a careful consideration of the message content) or a peripheral information-processing route (a cue or decision rule basis for persuasion). Within the more central route, cognitive









52

processing may be relatively objective or relatively biased. As a result of this, individuals with different levels of involvement will engage in different types of cognitive processing.

The present study will screen for four groups of

subjects: (1) high ego and high issue involved, (2) high ego and low issue involved, (3) low ego and high issue involved, and (4) low ego and low issue involved. Subjects will be asked to consider a message advocating the use of the cognitive restructuring counseling technique for the treatment of assertiveness problems. The topic of assertiveness was chosen for this investigation because of its relevance for a college population. Using measures of CFCR, outcome expectation, behavioral intention, and spontaneous use of cognitive restructuring to assess the subjects' cognitive responses to strong and weak arguments advocating the use of the technique, the following predictions are made:

1. Replicating the findings of Petty and Cacioppo (1979b), there will be a significant interaction
between issue involvement and argument quality.
This effect will be such that there will be greater distinction between strong and weak arguments under
conditions of high issue involvement than under
conditions of low issue involvement.

2. A significant interaction is also predicted
between ego-involvement and argument quality with a
greater distinction between strong and weak messages occurring under high than low egoinvolvement.

3. A biasing effect is also predicted to be
reflected in a significant interaction between








53
issue involvement and ego-involvement. Under conditions of high issue involvement, high ego-involvement will exert a greater influence on the weak than the strong messages. Under low involvement, high ego-involvement will exert an equal influence on both strong and weak messages.














CHAPTER III
METHODS


This study was designed to investigate the effects of ego-involvement and issue involvement upon the cognitive processing of strong and weak arguments for the use of cognitive restructuring as a treatment technique for assertiveness. The sample for the study consisted of 120 students drawn from the University of Florida introductory psychology subject pool and an undergraduate sociology class, Marriage and Family. Cognitive responses and acceptance of arguments was assessed across four groups: 1) high ego and high issue involvement, 2) high ego and low issue involvement, 3) low ego and high issue involvement, and 4) low ego and low issue involvement. Half of each group was exposed to strong arguments and half to weak arguments in favor of using cognitive restructuring. Levels of ego-involvement were assessed by using the Relationships Grid (see Appendix A), a form of the dyad grid (Ryle and Lunghi, 1970). Levels of issue involvement were measured through the use of the Simple Rathus Assertiveness Schedule (McCormick, 1984). The preceding instruments were administered to one General Psychology class and one Marriage and Family class.

54









55

Cognitive responses to strong and weak arguments for cognitive restructuring were assessed through the use of the Thought-Listing Technique (Cacioppo and Petty, 1981). Several seven-point Likert-type scales on which subjects rated their attitudes towa rd cognitive restructuring and their intentions to use it were also used to measure their acceptance or rejection of the arguments. An additional Thought-Listing Technique was used to assess the spontaneous use of cognitive restructuring by subjects during a covert rehearsal technique.

This chapter is divided into several sections. In the first section, the research design and variables of interest are discussed. In the second section, the population of interest and the study's sample and sample selection and recruitment procedures are discussed. In the third section, the instruments used in this study are presented. In the fourth section, the experimental procedure and data collection is described. In the fifth section, the method of data analyses are presented. Methodological limitations of the study and a summary of the methodology are presented in the final two sections.

Research Design

This study involved a 2 x 2 x 2 (Ego-involvement: high or low X Issue involvement: high or low X Intervention Quality: strong or weak arguments) factorial design. Subjects were crossed on the first two factors according to








56

a median split for issue involvement and upper and lower divisions of a trichotomization of ego-involvement and then randomly assigned to either strong or weak intervention quality. Ego-involvement was operationalized in terms of a summed relationship score for the construct, "Assertive-Nonassertive" on the Relationships Grid. The summed relationship score (Bannister, 1965) is a measure of the total variance within a construct system accounted for by one construct. The greater the total variance for a construct the greater the implications and importance it holds for a person. The variable, ego-involvement, had two levels: high and low. The high ego-involvement level refered to students who scored in the uppermost division of a trichotomization of the sample's summed relationship scores for the construct, Assertive--Nonassertive. The low ego-involvement level refered to students who scored in the lowermost division for that score.

The second factor, issue involvement, was

operationalized by the use of an objective inventory assessing personal assertiveness, the Simple Rathus Assertiveness Schedule (McCormick, 1984). Subjects who scored above the median for the pool of subjects on the objective inventory (high trait assertiveness) were considered to have low issue involvement. As scores below the median on the objective inventory would indicate above









57

average concern with assertion, subjects who scored in the below the median range comprised the high issue involvement condition.

The third factor, argument quality, consisted of two levels, strong arguments and weak arguments. Strong and weak arguments for the use and effectiveness of the cognitive restructuring technique were developed by Greg Neimeyer and his research team (Unpublished data, 1986). Following the procedure outlined by Petty and Cacioppo (1981) for developing arguments for a topic, they generated strong and weak arguments for the use of cognitive restructuring in the treatment of eating disorders. In the present study these arguments have been adopted to the use of cognitive restructuring in the treatment of assertiveness.

Cognitive Responses to Arguments

The variables of interest in the present study

consisted of the subjects' cognitive responses to arguments for cognitive restructuring. Cognitive responses were operationalized in terms of cognitive favorability toward cognitive restructuring, outcome expectation and behavioral intention ratings, and judges' ratings of spontaneous use of cognitive restructuring by subjects during a covert rehearsal condition.

Cognitive favorability toward cognitive restructuring (CFCR) was derived from the Thought-Listing Technique









58

(TLT). The TLT is used to identify a person's subjective thoughts or reactions to a topic. In the TLT subjects were instructed to list the thoughts they had while listening to a taped message on cognitive restructuring (CR). Subsequently, they also rated whether their thoughts were in favor of using CR, opposed to using CR, or neither in favor of nor opposed to it. A general index of CFCR was calculated for each subject by subtracting the number of unfavorable thoughts from the number of favorable thoughts listed and then dividing by the total number of relevant thoughts yielding a ratio score (see Cacioppo and Petty, 1981). Higher CFCR scores indicated more positive or favorable thoughts toward CR, while lower or negative CFCR scores indicated less favorable thoughts or opposition to the use of the CR technique. It was expected that under conditions of high issue involvement individuals with high ego-involvement would have lower CFCR than individuals with low ego-involvement. It was also expected that under conditions of low issue involvement individuals with high ego-involvement would have greater CFCR than individuals with low ego-involvement.

Cognitive responses toward the CR technique were also operationalized in terms of outcome expectations and behavioral intention ratings. The outcome expectation asked subjects to indicate, "To what extent do you think the cognitive restructuring technique would be beneficial









59
to you, personally?" As a message of behavioral intention, subjects were asked to indicate "How likely are you to use the cognitive restructuring technique in the future in situations requiring assertion"?7 Subjects responded to those items as well as a number of manipulation check and ancillary items on a seven point Likert-type scale. It was expected that analysis of outcome expectation ratings would parallel the predictions for CFCR. In addition, under conditions of high issue involvement it was expected that subjects with low ego-involvement would have greater intentions to use the CR technique than subjects with high ego-involvement.

Finally, the spontaneous use of cognitive

restructuring during a covert rehearsal condition (see Appendix E) was also used to operationalize subjects' cognitive responses to the CR arguments. It was reasoned that subjects with greater favorability toward cognitive restructuring and greater behavioral intention to use it will generate more CR thoughts during a covert rehearsal of a situation requiring assertion.

Description of the Sample

The population of interest in this study included all undergraduate students attending the University of Florida, Gainesville, Florida.

The sample in this study was drawn from the University of Florida introductory psychology subject pool and from an









60

undergraduate sociology course on marriage and family at the University of Florida. During the first week of spring semester 1987, students in one section of Psychology 2013 (General Psychology) and Sociology 2430 (Marriage and Family) were asked to complete a pretest consisting of identifying information, the Relationships Grid, and the Personal Problems Inventory (see Appendices A & B).

on the basis of the scores from these two pretests, subjects were divided into four separate groups: 1) high ego and high issue involvement, 2) high ego and low issue involvement, 3) low ego and high issue involvement, and 4) low ego and low issue involvement. In order to assign students to these four groups the following procedure was followed. First, all students will be rank ordered according to their summed relationship scores for the construct, "Assertive--Nonassertive" on the Relationships Grid, and a trichotomization was performed on the basis of these scores. The uppermost division constituted the high ego-involvement group; the lowermost division constituted the low ego-involvement group. Second, all students were rank ordered according to their scores on the Simple Rathus Assertiveness Schedule, which is entitled the Personal Problem Inventory in this study. On the basis of these scores a median split was performed, and the upper half of the split consisted of the low involvement group, while the lower half consisted of a high issue involvement group.









61

Third, students were assigned to the appropriate group according to their high and/or low classifications on ego and issue involvement. This procedure resulted in the inclusion of 104 subjects (N=40, F=64). The percentage of males within each group is presented in Table 1. Mean egoinvolvement and issue involvement scores and standard deviations for all groups in this study are also listed in Table 1.

Instrumentation

Six instruments were used in this study: 1) The

Relationships Grid, 2) Personal Problems Inventory (Simple Rathus Assertiveness Schedule), 3) Assertion Survey, 4) Thought Listing for Message, 5) Covert Rehearsal Technique, and 6) Cognitive Assessment Questionnaire. Relationships Grid

The Relationships Grid (RG; see Appendix A) is a form of the dyad grid (Ryle & Lunghi, 1970). The RG is an adaptation of the original repertory grid technique developed by G.A. Kelly (1955) to elicit and measure personal construct systems. "A grid may be defined as any form of sorting task which allows for the assessment of relationships between constructs and which yields these primary data in matrix form" (Bannister and Mair, 1968, p. 136). The basic components of a grid are elements and constructs. In the RG, like*other dyad grids, the elements used consist of relationships (e.g. my relationship with my












62





0
u IV CN r-I LO
m co Ln


4-4 u 4 r-i C4 r-i
0
ON 0 r-4
W v 0 co m


> r-4 C14 r--4 CN
w r--l r-4


0
44 > Lf)
.,.f 0
ul 4-) u
W 04 En

w U) 00 CN ON C)
u :3 C) r- C14 co
4-) a
4.) C C;
co r- r-4 1-4
4 4J r--4 r-4
(a

E 1w 00 00 N
0) O)p C14 m Ln m
>
r-4
U) 0
c > 0 CN lw co
0 c cn CN CN
.H 1-4 En
4-J I a)
(a 0 r-4 4J
.,q m Q C
> W m (1)
a) -4

> 3: al 3:
> r-I 0 0
0
41 >


4.) 0
> 0)
-4 41 PQ
0
4-) >

4J


E
Q)
a) (n >
r-4
0
0 0






lu l









63

spouse, my relationship with my father) instead of individuals as in the original repertory grid. Dyad grids use relationships as elements in order to gain information about a person's interpersonal behavior across a range of important relationships in his/her life. Dyad grids have been used in the past in clinical work with married couples (Ryle and Breen, 1972a and 1972b; Ryle and Lipshitz, 1975 and 1976) and in individual therapy to provide hypotheses for the conduct of therapy (Ryle and Lunghi, 1971; Ryle, 1979a; Ryle 1981) and to assess change (Ryle, 1979b; Ryle, 1980) Elements and constructs for this study's RG were adapted from Ryle's Relationships Grid (Ryle, 1985) and modified for the purposes of this study. For a complete listing of elements and constructs see Appendix A. The elements of interest consisted of the subjects' relationships with several important figures in their lives and how they saw themselves in those relationships. Constructs described interpersonal behavior and were presented in the classical bipolar form.

The RG, a ten by ten grid, involves having subjects rate how they see themselves in their relationships with ten interpersonal figures (e.g., mother, sibling, roommate) along ten provided constructs using a seven-point Likerttype scale. For example, a subject will be asked to rate his/her relationship with mother along the construct, Assertive +3 +2 +1 0 -1 -2 -3 Nonassertive. The subject









64

may believe him/herself to be moderately nonassertive in that relationship and place a -2 in the appropriate square in the grid. The left pole or side of each construct will be assigned a positive valence, while the right pole will be assigned a negative valence. The Relationships Grid and complete instructions are provided in Appendix A. Estimated time for completion of the RG is fifteen minutes.

Numerous measures can be derived from a grid, but the measure of interest for this study is the summed relationship score for the construct, "AssertiveNonassertive." The summed relationship score (Bannister, 1965) is a measure of the total variance within a construct system accounted for by a single construct. The greater the total variance accounted for by one construct; the more closely it is related to all other constructs. A construct with greater variance and more implications for the rest of an individual's construct system would be more superordinate and hold more importance for that individual than a construct accounting for less variance in the system. In this study subjects with higher summed relationship scores for whom the construct, AssertiveNonassertive, accounted for a greater amount of variance in their construct systems were designated as highly involved. Subjects for whom the assertiveness dimension accounted for less variance in the system were designated as low ego involved.









65

The calculation of variance scores is described by

Bannister (1965). Pearson product-moment correlations are first calculated between all constructs. Second, each correlation is squared, multiplied by 100, and the original sign retained. These scores are called relationship scores. The relationship scores for any one construct may be summed to give the total variance for that construct. In this study the relationship scores for the construct, "Assertive-Nonassertive," will be summed to give the total variance accounted for by that construct. If this score is then divided by the total variance in the system across all constructs, then it yields an index of the proportion of variance accounted for by the single construct AssertiveNon Assertive. A trichotomization was performed along the distribution of these scores. The uppermost division was designated as high ego-involvement and the lowermost as low ego-involvement.

Investigations into the psychometric properties of the Rep Grid are complicated by the varieties of grids and scorings in use. While the reliability and validity of summed relationship scores is not reported in the literature, psychometric properties of total variance scores (Bannister, 1960) are reported. Bannister (1962) reports an immediate retest reliability of 0.35 in a sample of 30 normal subjects. Honess (1977) reports the same reliability correlation in a rank order grid but a









66

correlation of 0.62 for a bi-polar implications grid. While reliability for total variance scores is generally low, this may be due to its sensitivity to construct system change as the individual and/or situation changes. Total variance scores have been shown to effectively discriminate thought-disordered schizophrenics from normals and other psychiatric groups (Bannister, 1962; Bannister and

Fransella, 1966).

Rather than using the total variance score, this study uses the proportion of variance accounted for by the single relevant construct. The psychometric properties of this score are not yet established. It seems reasonable to say that despite the instability of the overall variance in the system, the proportion of variance accounted for by any one construct might remain relatively constant. To test this assertion 30 students from an undergraduate psychology course were administered the RG twice with a 7 day interval between administrations. Proportion scores for the relevant construct were used to calculate a Pearson product-moment correlation. This test-retest correlation was .66, (p <.0001).

Personal Problems Inventory

The Personal Problems Inventory (PPI; see Appendix B) was used to assess issue involvement concerning assertion. It consisted of the Simple Rathus Assertiveness Schedule (SRAS) (McCormick, 1984). The SRAS is an easier-to-read,









67

revised version of the Rathus Assertiveness Schedule (Rathus, 1973). The SRAS is a 30-item self-report assertiveness inventory, which provides a global rating of perceived trait assertiveness. Respondents mark each item in terms of how characteristic the behavior is of the individual from 5 (very much like me) to 1 (very unlike me) A total score is obtained by summing item scores after correcting for reversed scoring weights.

In an investigation of the test-retest reliability of the SRAS, thirty undergraduate psychology students were administered the SRAS twice with a 7 day interval between adminstrations. A Pearson product-moment correlation yielded a coefficient of .91 (p <.0001).

McCormick (1984) provides evidence for a satisfactory degree of equivalence between the Simple Rathus Assertiveness Schedule (SRAS) and the Rathus Assertiveness Schedule (RAS) He reports a mean interim correlation of

0.79 between the two versions of the test with all correlations reaching statistical significance. The correlation between the total scores of the two tests was 0.94. He also reports a correlation of 0.90 between total odd and even item scores for both versions. Finally, in his sample of 116 undergraduate students he found an almost identical distribution of scores on both tests. Mean scores and standard deviations for the two versions were M=96.12, SD=23.85 (SRAS) and M=96.29, SD=24.94 (RAS).








68

Psychometric research on the RAS suggests that it has moderately high stability (test-retest reliability) and moderate to high homogeneity. Rathus (1973) presents a five-week test-retest reliability of 0.78 and a split-half Pearson product moment correlation of 0.77 between total odd and total even item scores. Vaal (1975) presents similar correlations of 0.76 for test-retest reliability over an eight-week period and 0.77 for split-half internal consistency. With college students seeking assertion training Heimberg and Harrison (1980) obtained over an eleven to fifteen day period a test-retest reliability correlation of 0.80 and a split-half internal consistency correlation also of 0.80. Other reports of internal consistency correlations have ranged from 0.69 to 0.86 (Mann and Flowers, 1978; Quillan, Besing, and Dinning, 1977; Futch, Scheirer, and Lisman, 1982).

A base of validity information has also been

established for the RAS. When comparing RAS scores with peer ratings on a seventeen-item semantic differential Rathus (1973) found that total RAS scores correlated positively with each of the items comprising the peerrating scale's assertiveness factor. Several studies have tested Wolpe's (1973) argument of an inverse relationship between trait anxiety and assertiveness; conformation of Wolpe's argument would provide indirect support for the construct validity of the RAS. Orenstein, Orenstein, and








69

Carr (1975) found RAS scores to be significantly correlated (range -0.60 to -0.75) with interpersonal fears as measured by the social factors from the Wolpe-Lang (1964) Fear Survey. Other studies by Morgan (1974) and Hollandsworth (1976) have found lower correlations between RAS scores and a social fear survey. In Morgan's study of college students the correlations ranged from 0.172 to -0.239 and in Hollandsworth's study a correlation of 0.436 was found. The combined data from these studies does appear to provide some support for the construct validity of the RAS.

Concurrent validity of the RAS has also been

established in several studies. Rathus (1973) reports a correlation of 0.705 between RAS scores and observer ratings of verbal behavior in response to five questions asking for assertive behavior. Frankel (1977) reports significant correlations between the RAS and the Conflict Resolution Inventory and between the RAS and the Assertion Inventory. MacDonald (1975/1974) also supports the concurrent validity of the RAS in her findings of a moderate relationship between RAS scores and behavioral measures of assertiveness. In addition, the RAS has been used in treatment outcome studies and has been shown to be a sensitive index of pretreatment to post treatment change (Rathus, 1973; Blanchard, Turner, Eschette, and Coury, 1977).








70

Assertion Survey

The Assertion Survey (AS; see Appendix C) is a

questionnaire developed by the researcher to assess outcome expectations. It consisted of eight 7-point Likert-type items. Outcome expectation was assessed by the item, "To what extent do you think the cognitive restructuring technique would be beneficial to you, personally?" Behavioral intention was assessed by the item, "How likely are you to use the cognitive restructuring technique in the future in situations requiring assertion?" As a check on the argument quality manipulation, subjects answered a seven-point Likert item asking them to rate the quality of the arguments in the intervention: very poor arguments (1) to very good arguments (7). In addition, as a general check on the issue involvement manipulation, subjects answered another Likert item asking them the extent to which the message has implications for them personally: not at all relevant to me (1) to very relevant to me (7). Ancillary items assessed other aspects of the message including voice quality, speaker qualification, and ratio of delivery.

Thought Listing Techniques

The Thought Listing Techniques used in this study were variations of the "thought-listing procedure" developed by Brock (1967) and Greenwald (1968). The procedure has been refined to become a suitable self-report technique for









71

obtaining a written listing of an individual's thoughts on a topic and a general index of CFCR toward the topic (Cacioppo and Petty, 1981). In the present study Thought Listing was employed under two conditions: 1) after listening to the taped arguments for cognitive restructuring and completing the Assertion Inventory and 2) as part of the covert rehearsal of a personal assertion situation. Under the first condition the Thought Listing Technique (see Appendix D) was used to compute a CFCR index, and under the second condition (see Appendix E) it was used to tally the spontaneous use of cognitive restructuring. Under the first condition subjects listed their thoughts concerning the arguments for CR according to directions similar to those of Petty and Cacioppo (1977). In addition, they rated their thoughts along the dimension of favorableness toward cognitive restructuring. The cognitive favorability index was calculated for each subject by subtracting the number of unfavorable thoughts from the number of favorable thoughts listed and then dividing by the total number of relevant thoughts (Cacioppo and Petty, 1981). Petty et al. (1976) report correlations between subjects' own rating and independent judges ratings of .82 for favorable thoughts and .79 for counterarguments. Under the second condition, two undergraduate psychology major students trained by the researcher and unaware of the hypotheses of the study and the involvement group








72

classifications of the subjects judged the thoughts listed for frequency of spontaneous use of cognitive restructuring. This consisted of the post-coding of each subject's listed thoughts into one of five categories: 1) self-instructional statement, 2) self-assuring statement, 3) self-doubting statement, 4) presentation content statement, and 5) other statement. Inter-judge reliability Pearson product-moment correlations were .81 for selfinstructional statements, .85 for self-assuring statements, .92 for self-doubting statements, .72 for presentation content, and .34 for other statements.

Support for the Thought Listing Technique (TLT) as a reliable measure can be found in a study by Cullen (1969/1968). She compared the split-half reliability and test-retest reliability of the TEIT with Likert attitude and Thurstone attitude scales on two topics, birth control and segregation. ,She found average split-half reliabilities of

0.78 for Thought-Listing, 0.83 for Likert scales, and 0.55 for Thurstone scales. The average test-retest reliabilities were 0.64 for the TEIT, 0.83 for Likert scales, and 0.53 for Thurstone scales.

In indirect support for the construct validity of the TLT, Petty, Harkins, and Williams (1980) found that the implied or real presence of others working on the same cognitive task decreased the thoughts listed by an individual in that implied or real presence of others when









73

compared to the individual working alone. This finding corresponds to the well-known "social loafing" argument (Latane and Dat ley, 1970).

Cognitive Assessment Questionnaire

The Cognitive Assessment Questionnaire (CAQ; see

Appendix F) was used to assess subjects' tendency to engage in and enjoy effortful cognitive endeavors. It consisted of the short form of the Need for Cognition Scale (NCS) (Cacioppo, Petty, and Kao, 1984) The short NCS is an 18item self-report inventory which provides a global assessment of need for cognition. Cacioppo, Petty, and Kao (1984) report a significant correlation (r = .95, p <,001) between subjects' scores on original and 18-item versions of the NCS. The NCS shows strong internal consistency (r.76, p<.001) and some evidence exists for its content and predictive validity (Cacioppo & Petty, 1982).

Procedures

Procedures were set up to insure standardization of

administration. Instruments were presented to each subject by cassette tape (see Appendix G for a complete transcript of the tapes for each experimental condition).

Subjects were run in group administrations in a language laboratory classroom. They were seated and provided with a package of written materials, the measures of interest. They were then instructed by the researcher to place their headphones on and follow the instructions








74

that they heard on the tape carefully. All further instructions to the subject were on the taped message.

The taped message consisted of a general introduction to the study and the cognitive restructuring technique for all subjects. Half of the subjects in each involvement level group were randomly assigned to one of two experimental conditions, either strong arguments in favor of the effectiveness of cognitive restructuring or weak arguments favoring cognitive restructuring. (See Petty & Cacioppo, 1984, for a discussion of the development of these messages). The next section of the tape that the subjects heard presented either strong or weak arguments for cognitive restructuring depending upon the experimental condition being conducted. This was followed by instructions for the completion of the Assertion Survey followed by three minutes of silence to enable the completion of the questions. Next, instructions for the thought listing for the message were presented and followed by three minutes of silence during which subjects were instructed to "List all of the thoughts that came to mind as you listened to the taped message." Instructions for the Cogntive Assessment Questionnaire were presented next and followed by five minutes of silence during which the subjects completed the questionnaire. The tape resumed with instructions for covert rehearsal of a personal assertion situation and its associated thought listing,









75

again allowing for three minutes to complete the task. Finally, subjects were instructed to remove their headphones, turn in their materials to the researcher, and receive their experimental credit from the researcher.

All subjects participating in the study will be given a subject number code which will identify their responses while maintaining their anonymity. All subjects also received either experimental credit or extra class credit for their participation in the study.

Analyses of Data

The analyses compared the impact of high and low

levels of ego-involvement, issue involvement, and message quality on attitudes toward cognitive restructuring. Dependent variables included an index of cognitive favorability in response to the taped messages, outcome expectation attitude ratings, behavioral intention ratings, and post-coding of covert rehearsal thoughts. A series of 2 x 2 x 2 analyses of variance were performed to determine the effects of ego-involvement, issue involvement, argument quality, and interactions on these variables. Duncan's Multiple Comparison Procedure was used to evaluate the effects predicted in the research hypotheses.

Summary

This study was designed to test and compare the

varying levels of ego-involvement, issue involvement, and argument quality upon a subject's cognitive responses to









76

the cognitive restructuring technique presented in a message advocating its use in treating assertiveness problems.

The design was a 2 x 2 x 2 between subjects factorial design. The factors were ego-involvement (high and low), issue involvement (high and low), and argument quality (strong or weak). The dependent variables consisted of the following: 1) cognitive favorability index, 2) outcome expectations for the use of the CR technique, 3) behavioral intention toward cognitive restructuring, and 4) spontaneous use of CR during covert rehearsal.

The sample was drawn from the University of Florida introductory psychology subject pool and an undergraduate sociology class. It consisted of students pretested to form four distinct groups: 1) high ego involved, high issue involved; 2) high ego, low issue involved; 3) low ego, high issue involved; and 4) low ego, low issue involved.

Instruments used in this study included the

Relationships Grid, Personal Problems Inventory (which consisted of the Simple Rathus Assertiveness Schedule), Assertion Survey, Cognitive Assessment Questionnaire (Need for Cognition Scale), and Thought Listing Techniques for messages about cognitive restructuring and covert








77

rehearsal. Procedures for collection and evaluation of data were standardized in order to prevent bias between groups.

Analysis of data compared the four involvement groups and argument quality. A 2 x 2 x 2 ANOVA for each cognitive response variable determined the effects of egoinvolvement, issue involvement, argument quality, and interactions.














CHAPTER IV
RESULTS



Manipulation Checks

Before investigating whether the manipulations produced the desired effects on the dependent measures, the effectiveness of the manipulations should be ascertained. Manipulation checks were planned for argument quality and issue involvement and analyzed using a series of 2 x 2 x 2 analyses of variance.

Argument quality

To see whether subjects differentiated the strength of the arguments used in the message quality manipulation, subjects answered a seven-point Likert item asking them to rate the quality of the arguments in the intervention: very poor arguments (1) to very good arguments (7). Results of the Analyses of Variance indicate that the manipulation successfully influenced how strongly the subjects rated the quality of the two groups of arguments, F(1,96)=30.43, p<.0001: strong arguments M=5.0, weak arguments M=3.5. The involvement variables (issue and ego) did not significantly affect this manipulation check, either alone or in interaction with the argument quality manipulation.


78








79

In addition, as predicted by the ELM subjects who heard the strong argument quality message generated a significantly greater proportion of positive thoughts (M=.76) than did subjects who heard the weak message (M=.55), F(1,90) = 12.21, P<.0007. Furthermore, subjects who heard the message with weak arguments generated a significantly greater proportion of negative thoughts (M=.45) than did subjects who heard the strong message (M=.23), F(1,90) = 12.21, p<.0007. Issue Involvement

The other category of manipulati on checks investigated

the degree of issue involvement that subjects believed that they had with the use of the cognitive restructuring technique for treating personal problems in assertion. As an initial check on issue involvement a seven-point Likert item asking subjects to indicate the extent to which the message had implications for them personally was' included. An analysis of variance revealed a trend toward a main effect of issue involvement, F(1,96)=2.8, p=.10; high issue M=4.25, low issue M=3.72. All other variables (ego and argument quality) also did not significantly affect this manipulation check. As these results cast doubt on the effectiveness of the issue involvement manipulation, an alternative operationalization of involvement was pursued. Petty and Cacioppo .(1986) suggest that as personal involvement increases, people become more motivated to








80

process the issue-relevant arguments presented. One way in which argument processing could be measured would be by the number of relevant cognitions that a person generates. In order to assign subjects to high and low levels of argument processing (high and low issue involvement)-a median split was performed on scores of total number of relevant thoughts generated by subjects on the thought listing technique. The median score was deleted and subjects in the upper half of the distribution constituted the high issue involvement group, while subjects in the lower half constituted the low issue involvement group. Analysis of variance of the involvement manipulation check revealed that high issue involvement subjects believed that the message had significantly more implications for them than did the low issue involvement group, F(1,96)=6.17, p<.Ol: high issue involvement M=4.4, low issue involvement M=3.6. The other variables (ego-involvement and argument quality) did not significantly affect this manipulation check.



Dependent Measures

Data for each of the dependent measures was analyzed by a 2(High and Low Issue Involvement) x2(High and Low EgoInvolvement) x2(Strong and Weak Argument Quality) Analyses of Variance. Analyses for each dependent measure are presented separately.









81

Cognitive Favorability (CFCR)

The analysis of variance revealed a significant main effect of argument quality for the CFCR score, F(1,90)=12.21, p<.001, such that subjects who heard the strong arguments had a greater proportion score for overall cognitive favorability toward cognitive restructuring (M=0.54) than did subjects who heard the weak arguments (M=0.10) (see Table 2) Means and standard deviations for CFCR for all levels of the design are presented in Table 3. Attitude Rating

The analysis of variance revealed a significant main effect of issue involvement for the outcome expectation attitude rating, F(1,96)=9.23, p<.003, such that subjects in the high issue involvement condition believed that the cognitive restructuring technique would be significantly more beneficial to them (M=4.58) than did subjects in the low issue involvement condition (M=3.76) (see Table 4). Means and standard deviations for attitude rating are presented in Table 5.

The results also revealed a trend toward a main effect of argument quality for the attitude rating, F(1,96)=3.54, p=.06. These results suggest that the group that heard the strong arguments had a tendency to believe that the cognitive restructuring technique would be more beneficial to them (M=4.4) than did the group that heard the weak arguments (M=3.9).











82




N ON r-q r- Ln Ln CN
"W r-i 0 0 lw %D v
w 0 0 t.0 co V en


C14 r-I r-q r- v qw r-4
C14 0 C14 CN 0 ull a,%
DW 0 0 a 0 0 0
C) 0 C14 0 0 0 0


(n IV i a% o r- lw 0
00 0 r-4 0 rq N r- r-4
ON C> 0 r-4 0 CN m v
En 0 0 0 0
(D Ln 0 0 0 0 CD

fa w X


44 r-4 -4 r--i r--4 r-4 r-I 0
ra






0 00 0 4 C) r-I C14 r- C14
u W aN 0 0 r-4 0 C14 m M
44 0 Ln 0 0 C) 0 w
0 M

rT4


0
44

w
u
>1
-4

> 0
44 w Cl >4
0 4J
x x .,q
4J 4J 4J

Of
as (D w a) >1 x
c > > > 41
r-4 r-4 r--i -4 0
0 0 0 0)
> > >
x
u (D (1) x Q) w
w :3 :3 :3 Z 0
:3 En 0 (a U) ri) 0 ul ).4
0 U) 0) 0) rA W
E-4 M C9 w w











83








0i m* 04 m aO~O
Ur~0 w 0 V$ LA %D co IV co



U t' 0 LA LA CN .4 U 0 0 0 0D 0 0;
0: >
44 > ci -4 r4 r
0i) v 4a
>C
W ci)

(D
U41
4-j'

m cw C: Nd C
04 04-4 0 0 i 0 Mi 0
r_> -W J 4 .J 4 -' 3 4 -'
r-4 ::$ /)C) )U


m4J co)



4-i



C

4J W
E-i 4' 4J 0 '
04 r4a 0


> 0.JJ 04'9


ci) 0 .
0 0 0i





ai) ci)

U)
E-4U












84





04 m Ln m 10 LO r-A 1w
0 v %D Ln C14 m




w m r- lw 00 CN
N r-4 Ln Ln w C%4 r--4

0) Clq 0 C>



r- 00 00 00 tn r-4 Ln
LO 0 LO r- m r- Nr 00
Ln m M-4 m Ln C14 C>

ON -W r- r-4 r-I a CN
r-4


0
0

ro r-I r-4 w






qw r- 00 00 co LO
ra ul C> LO r- m r- "W co
1-4 C14 Ln m m Ln N r-4
-W 44
0 r-i
r-q
4J 04

cn

0
44







> 0 (a
m 3
44 Of
0
x x
4J -W -W
C r_ C:
w a) CR
E
Q) >1 x
> > > 4J
r--q r--4 -4 -4 0
0 0 0 -q (n
> > >

>1 0, x
4J
u x
W :3 0
:3 U) 0 (D Ul 0 U)
0 En w 0) (1)
La Ol -q W -q











85







r- Ln ko Ln C14 Ln "0
Ln C) ko "W r- 110 0 0)
ru CN r-4

%0 Ln .o
w W CN 10 cn "W as a% m
0
4J
m 44
C4 0 41

ra 0 ra
:3 -H r-W 4J Q)
-4 a 04
,u W
>
>1 4J 0) 00 co co Ln a% co f-41 m r-I r- v Ln "W
4j -,q
0

ra r-4 m C14 0
U) >11 Ol n k.0 00 0
4J
4

to

W E-4 4 Ln r-I N
m 4 a r-4 r-A r-A r-i
a4 W
c ro 4-1
4J

w 4-1 -14 Sd C: -14
01 0 0 m 0 co 0 ra 0 (13
C: 44 4-1 w 0) w a) w (1) w a)
0 al 4J :3: 4J 4J 3: 41 .11:
w U) U)
W 0) 4 Ol
fu w E
0 0)
> u > 4-1
a) to c
o >
ro 0 r. E
4 -4 >
m 41
ro a 0 V
0 0) 3: 3: ol 3:
> 0 0 0 0
4-) W a) r_ $-a --a -l
4-)

4J ra 0
4J (0

3: 0 Q)
-1 > 4J
> r-4 c
(o 0
>
Q)

0

0 0



U)








86

Behavioral Intention Rating

The analysis of variance revealed no significant

findings for the behavioral intention score. means and standard deviations for the behavioral intention score are presented in Table 5.

Spontaneous Use of Cognitive Restructuring

Only the post-coding categories of self-instructional

(SI), self-assuring (SA), and self-doubting (SD) statements were deemed to have adequate inter-judge reliability (range .81 to .92). These three categories were analyzed by separate 2 x 2 x 2 Analyses by Variance.

Self-instructional statements. The analysis of variance revealed no significant findings for the SI score.

Self-assuring statements. The analysis of variance revealed no significant findings for the SA score. However, the results did reveal a trend toward an interaction effect of ego-involvement x argument quality for the SA score, F(1,89)=2.8, p=.10. Examination of the means suggests that under the high ego-involvement condition argument quality did not influence the number of self-assuring statements generated, but under the low egoinvolvement condition subjects who heard the message with strong arguments generated significantly less self assuring statements than subjects who heard the message with weak arguments.








87

Self-doubting statements. The analysis of variance revealed no significant findings for the SD score.

Additional Analyses

Correlations computed between eight of the design's

variables using Pearson product-moment correlations yielded several significant relationships. The correlations most germane to the present study included those between the need for cognition variable and the other variables in the study. These analyses yielded only one significant correlation between the need for cognition score and the score on the Simple Rathus Assertiveness Scale (SRAS), r .372, p<.0001. This result suggests that as subjects increase in need for cognition, they see themselves as more assertive. Perhaps even more interesting is the nonsignificant correlation between the need for cognition variable and the total number of relevant thoughts generated by the subjects. This nonsignificant correlation does not support Cacioppo and Petty's (1984) notion that individuals high in need for cognition would tend to think more extensively about messages presented to them. Correlations are presented in Table 6.










88
4J r-I r-I a 0) 00 CN cn m aN r--4 Ln 0 v 00
C 4J qw 0 00 r-I r-I Ln IV r-I a r- Ln a% -4 co
04 000 0 lw 000 CN 0 r-4 0 Ln 000
> 0) CD a 0 0 0 0 0 a 0
:3 v v v v v v
0 v 04 04 04 04 04 04
4 QA
E-#
r-q
(n En v CD m r--i 0) C-4 Ln M CN 00 CN N 0
m (n CN m r-q lw co co v 110 Ln r-i r- C)
0 r- r-I q 0 CD 0 m 0 o 0110 cn C>
0 0 0 0 0 0 0 0
u I v v v I v I 1 1 1; 0 %,;
ul 04 04 04 04 04

C
w 0 co
0-4 w lw r- v r-4 CN C14 N -4 w qw r4-4 4J 0 m 14 m C%4 CN r-A 0 m CO m m
-4 0 r-I C14 -4 N 0 m l r-4 0
ro 0 0 0 0 0 0 0 0 *
w 0) 04 v v v
w 0 04 04 04 04 04
z u
co r-q
Ln v) rl- k.0 m C> O)o IV C>
%D N r-4 kO m C> Ln c CN 0
4 u a tn 0 OD (Y) C) CY) CD qw 0
lu V 0 0
> v 1 1; 1; 1;
4J 04 04 04 04
r.
a) r-I
ra (a a 1-1
r. w 0 r-4 Ln w r- 00 C) 0
(D 0 M 00 Ln r- m C14 m 0
q 0 Lo m r- 0
> 0 0
M w v I v
,a 41 04 04
ra a) .
r. m 0-4


> ro a)
.,1 :3 C qw tn N a) C14
4J 4j -4 r- 0 r-4 Lf) 00 w
04 -ri 4.) C-4 0 r-4 CN 4 0
.r.4 4J rO 0 0 0 .
w 4J v v I v
u
U)
a)
4J
44 (3) 41 r-4 OD r-4 00
0 E 0 00 C) 00
0 a)
C: .
0 4 I v v
.H < Of 04 04
4.)
M

4-J

0 W r-4 r4
w v
a) 0 > 04
4-J 0) r-q
0 0
> 4J 4J u
C c 1-1
Q) (1) M rE E 41 (13 Q) 4 0 w 0
Q) a) r_ >1 ra 0 -,q 0 -r-I
> > 0) -W :3 CP -4 4J 44 4.) 4
E 41 > C:
0 1 0 :1 Ix ra >
> 0 > m m 4-) -W a 41 0) al a)
U) r_ U) r_ w ::3 4.) CO Q) a) 0 %
J- W -4 F:4 Ol 4 % M -4 u z u (Tj















CHAPTER V
DISCUSSION



The primary purpose of the present investigation was to examine the effects of issue involvement and egoinvolvement upon the cognitive responses that individuals generate to arguments advocating the use of cognitive restructuring to treat problems in assertiveness. Hypotheses for these effects were developed from a review of the social psychological literature on the Elaboration Likelihood Model of Attitude Change. The present study attempted to extend this model to include messages that may be used in a counseling context.

In discussing variables that can affect a person's ability to process message arguments Petty and Cacioppo (1986) cite personal relevance/involvement as "perhaps the most important variable in this regard" (p. 144). Their ELM suggests that as issue involvement increases, people become more motivated to engage in the cognitive work necessary to evaluate the issue-relevant arguments presented in a message. Petty and Cacioppo (1979b) provide evidence consistent with this view. The first hypothesis of the present study was also based upon this view. more specifically, it predicted a significant interaction 89








90

between issue involvement and argument quality such that there would be greater distinction between strong and weak arguments under conditions of high issue involvement than under conditions of low issue involvement. The predicted issue involvement X argument quality interaction was not found to be significant for any of the dependent measures.

The second hypothesis made a similar prediction of a significant interaction between ego-involvement and argument quality such that there would be a greater distinction between strong and weak arguments under condition of high ego-involvement than under conditions of low ego-involvement. Once again, the predicted interaction was not found for any of the dependent measures.

The results do indicate that the strength of the

argument quality significantly differentiated scores on the CFCR dependent measure and demonstrated a strong tendency to differentiate scores on the attitude rating dependent measure. These results, coupled with the significant argument quality manipulation check, suggest that overall the content (arguments) of a counseling message made a major difference in its effectiveness in advocating cognitive restructuring for treating assertiveness problems. Also consistent with the ELM Model was the finding that strong messages produced more positive cognitive responses, whereas weak messages generated more negative thoughts. For the counseling situation this








91

implies that it may be advantageous for the therapist to screen out distractions, to repeat important messages over a number of sessions, and even to provide the client with a written summary of central points made during therapy, all manipulations that may also have the effect of enhancing client involvement.

This effect of the message variable also has other important implications for counseling. According to the ELM, clients who carefully evaluate message content are more likely to have enduring, behavior-related, centralroute attitude change than are those who attend only to peripheral cues, such as counselor personal characteristics. Message variables (such as argument quality) may play crucial roles in eliciting behavior change and in the maintenance of therapy gains. Future research should investigate message factors to determine those that are most helpful to counselors in bringing about client change.

Results also show that issue involvement produced a significant main effect for attitude rating such that subjects who generated more thoughts about the message believed that cognitive restructuring had more personal benefits than did subjects who generated fewer thoughts. Theoretically, an explanation of this finding could be found in Zajoncls (1980) theory of mere exposure, which describes changes in evaluation of objects as a result of









92

relatively primitive affective and associational processes. It may be that those subjects who thought more about the message associated it more with counseling and the common notion that counseling is beneficial. Because a main effect for issue involvement was not significant for the CFCR measure, it does not appear that high issue involvement subjects saw the cognitive restructuring technique itself as being better than did the low issue involvement subjects. In other words, greater favorability toward the technique was related to more thinking in general, but not to more favorable thinking per se.

This theoretical explanation further suggests some

methodological factors that may have affected the results of the study. It may be that the high and low levels of issue involvement in this study do not correspond to the high and low levels of issue involvement of previous ELM research reported in the social psychological literature. In previous research involvement manipulations were designed to induce relatively pure forms of the central and peripheral routes to persuasion. For example, in the 44

Petty, Cacioppo, and Goldman (1981) study high involvement subjects were told that the message they were to evaluate had relatively important consequences for them (i.e., if they did not pass the senior comprehensive exam that was to be instituted the next year, they would not graduate). In contrast to this, low involvement subjects were led to









93

believe that the proposal would have no personal consequences for them. These two groups of subjects represent relatively extreme levels of involvement. In the present study, and in the counseling situation in general, the issues of concern are applicable to almost everyone. That is to say, most topics discussed in counseling (for example, assertion, problems in personal relationships) are by definition highly involving since such issues are only presented if they are personally problematic concerns. Therefore, the levels of issue involvement in the present study might best be described as moderately high and moderately low. Further support for a moderate level of involvement came from the nonsignificant difference on Rathus scores between high and low issue involvement conditions. This could explain the current failure to duplicate the interaction between issue involvement and argument quality found in previous ELM research. This same argument could also be extended to the ego-involvement variable.

Another methodology question concerns the

operationalization of issue involvement in counselingrelated research. Past investigations have attempted to manipulate issue involvement in a number of ways. These attempts have included manipulation of clients' "perceived need" or requests for help (Dixon & Claiborn, 1981; Heppner & Dixon, 1978) and motivations for counseling (Heppner &




Full Text
ACKNOWLEDGEMENTS
I would like to take this opportunity to acknowledge
and thank those individuals who have guided and supported
me throughout the completion of my dissertation. My first
thanks is to the Lord Jesus Christ for the life He has
given me during the past six years.
I am grateful to Greg Neimeyer for guiding and
encouraging me throughout my graduate studies and
especially during the completion of my dissertation. Most
of all, I am thankful for his time and his personal
interest in me and my professional development. I would
also like to thank the members of my committee, Franz
Epting, Shae Kosch, Constance Shehan, and Mark Alicke, for
their effort and insightful suggestions. It has been my
pleasure to work with these outstanding individuals.
I would also like to thank all the people who were an
integral part of this dissertation. Dr. Constance Shehan
allowed me to utilize her classroom to solicit subjects. A
special thanks goes to the undergraduate research
assistants who were involved in the technical aspects of
this study. Jean Callahan, Lisa Oglander, and John Guy
were invaluable in their assistance. April Metzler also
gave much time to this project.


63
spouse, my relationship with my father) instead of
individuals as in the original repertory grid. Dyad grids
use relationships as elements in order to gain information
about a person's interpersonal behavior across a range of
important relationships in his/her life. Dyad grids have
been used in the past in clinical work with married couples
(Ryle and Breen, 1972a and 1972b; Ryle and Lipshitz, 1975
and 1976) and in individual therapy to provide hypotheses
for the conduct of therapy (Ryle and Lunghi, 1971; Ryle,
1979a; Ryle 1981) and to assess change (Ryle, 1979b; Ryle,
1980). Elements and constructs for this study's RG were
adapted from Ryle's Relationships Grid (Ryle, 1985) and
modified for the purposes of this study. For a complete
listing of elements and constructs see Appendix A. The
elements of interest consisted of the subjects'
relationships with several important figures in their lives
and how they saw themselves in those relationships.
Constructs described interpersonal behavior and were
presented in the classical bipolar form.
The RG, a ten by ten grid, involves having subjects
rate how they see themselves in their relationships with
ten interpersonal figures (e.g., mother, sibling, roommate)
along ten provided constructs using a seven-point Likert-
type scale. For example, a subject will be asked to rate
his/her relationship with mother along the construct,
Assertive +3 +2 +1 O -1 -2 -3 Nonassertive. The subject


CHAPTER I
INTRODUCTION
Persuasion and resistance to change have been issues of
theoretical speculation and empirical investigation
throughout the past century (Freud, 1894/1959; Frank, 1961;
Ellis, 1985). The extent to which a client yields and/or
resists a therapist's persuasive messages can be influenced
by a wide range of variables. Since Strong's (1968)
initial theoretical conceptualization of counseling as an
interpersonal influence process, considerable research has
been generated concerning interpersonal influence variables
in counseling and counselor power in particular (see
Corrigan, Dell, Lewis, & Schmidt, 1980, and Heppner &
Dixon, 1981, for reviews). Strong proposed a two-stage
model of counseling. In the first stage, counselors
attempt to increase their perceived power along the
dimensions of expertness, attractiveness, and
trustworthiness. In the second stage counselors use their
influence to bring about client change. Virtually all of
the research conducted so far on counseling as an
interpersonal influence process has investigated the three
1


IV RESULTS 78
Manipulation Checks 78
Argument Quality 78
Issue Involvement 79
Dependent Measures 80
Cognitive Favorability (CFCR) 81
Attitude Rating 81
Behavioral Intention Rating 86
Spontaneous Use of Cognitive Restructuring.. 86
Additional Analyses 87
V DISCUSSION 89
Need for Cognition 96
Future Considerations 96
APPENDICES
A RELATIONSHIPS GRID 101
B PERSONAL PROBLEM INVENTORY 104
C ASSERTION SURVEY 107
D THOUGHT LISTING TECHNIQUE 109
E COVERT REHEARSAL TECHNIQUE 112
F COGNITIVE ASSESSMENT QUESTIONNAIRE 114
G TRANSCRIPT OF TAPED INSTRUCTIONS 116
REFERENCES 124
BIOGRAPHICAL SKETCH 135
v


31
believe that they were either one out of a group of ten
people who were evaluating the editorials or the only
person responsible for the evaluation. Editorials
contained either strong, weak, or a mixture of arguments
for the institution of comprehensive senior exams.
Attitude results revealed that individual evaluators were
significantly more favorable toward the strong message and
were significantly less favorable toward the weak message
than group evaluators. Responsibility did not affect
evaluations of the mixed message. Thus, as personal
responsibility increased, argument quality became a more
important determinant of the evaluation.
Finally, need for cognition, an individual difference
variable, refers to the extent to which people need to
structure information in reasonable, integrated ways
(Cohen, Stotland, & Wolfe, 1955). Cacioppo, Petty, and
Morris (1983) exposed high and low need for cognition
college students to either strong or weak arguments for a
proposal which would raise their tuition. After hearing
the message, subjects were asked to evaluate the message
and to give their personal opinion about the issue.
Results indicate that high need for cognition students
produced more polarized evaluations and attitudes for the
strong and weak messages. Furthermore, a significantly
higher correlation between argument evaluation and personal
opinion was found for the high (r = .70) than the low


94
Heesacker, 1982). Such manipulations have resulted in no
differential effects. More recent studies (Stoltenberg &
McNeill, 1984; Heesacker, 1986) have utilized approaches
that focused on one specific problem that may be presented
in counseling. For example, in the Stoltenberg and McNeill
(1984) study conditions of high and low issue involvement
were designated by a median split on the Decisiveness scale
of the Career Maturity Inventory. The manipulations of
issue involvement used in the latter two studies have
resulted in mixed findings. As a result, all of the
authors cited above have called for better operational
definition and further research in regard to client
involvement.
The present study originally defined issue involvement
in a similar manner to the latter two studies cited above.
However, because the manipulation check did not yield
significant results, issue involvement was redefined in an
empirical manner based upon previous ELM research. Based
on a median split on scores of total number of relevant
thoughts, subjects were assigned to high and low issue
involvement conditions. According to the ELM, those two
groups by definition represent high and low levels of
elaboration likelihood within the sample. Yet significant
interactions effects between issue involvement and argument
quality did not appeal in the results. Once again, while
these two groups were high and low groups for the sample,


I would also like to give a very special
acknowledgment to my wife, Barbara, for her constant love
and support and the many sacrifices she has made in her
life so that I may pursue my career. Barbara and my son,
Adam, have been great sources of support and encouragement
throughout this project.
in


132
Rhine, R.J. & Severance, L. J. (1970). Ego-involvement,
discrepancy, source credibility, and attitude change.
Journal of Personality and Social Psychology, 16, 17 5-
190.
Ross, L., Lepper, M. R., & Hubbard, M. (1975).
Perseverance in self-perception and social perception:
Biased attributional processes in the debriefing
paradigm. Journal of Personality and Social
Psychology, 32, 880-892.
Ryle, A. (1979a). The focus in brief interpretive
psychotherapy: Dilemmas, traps and snaps as target
problems. British Journal of Psychiatry, 134, 46-54.
Ryle, A. (1979b). Defining goals and assessing change in
brief psychotherapy: A pilot study using target
ratings and the dyad grid. British Journal of Medical
Psychology, 52, 223-233.
Ryle, A. (1980). Some measures of goal attainment in
focussed integrated active psychotherapy: A study of
fifteen cases. British Journal of Psychiatry, 137,
475-486.
Ryle, A. (1981). Dyad grid dilemmas in patient and control
subjects. British Journal and Medical Psychology, 54,
353-358.
Ryle, A. (1985). The dyad grid and psychotherapy research.
In N. Beail (Ed.), Repertory grid technique and
personal constructs: Applications in clinical and
educational settings (pp~I 190-206) Cambridge, MA:
Bookline Books.
Ryle, A., & Breen, D. (1972a). The use of the double dyad
in the clinical setting. British Journal of Medical
Psychology, 45, 383-389.
Ryle, A., & Breen D. (1972b). A comparison of adjusted
and maladjusted couples using the double dyad grid.
British Journal of Medial Psychology, 45, 375-382.
Ryle, A., & Lipshitz, S. (1975). Recording change in
marital therapy with the reconstruction grid. British
Journal of Medical Psychology, 48, 39-48.
Ryle, A., & Lipshitz, S. (1976). Repretory grid
elucidation of a difficult conjoint therapy. British
Journal of Medical Psychology, 49, 281-285.


20
peripheral routes. The central route involves a person's
thoughtful consideration and evaluation of the information
presented in the communication; whereas, the peripheral
route involves a person's selection from the persuasion
context of some cue (e.g., source trustworthiness) that
produces attitude change without the person having to
thoughtfully consider or evaluate the information
presented. Petty and Cacioppo found the central route to
be the more enduring of the two routes.
While people are neither universally thoughtful nor
universally mindless in evaluating persuasive messages,
they do vary in their motivation and ability to consider
carefully the information and arguments that comprise a
persuasive attempt. Petty and Cacioppo (1981) refer to
elaboration likelihood as the degree to which a person
thinks about the issue-relevant arguments contained in a
persuasion message. If conditions produce or enhance an
individual's motivation and ability to engage in issue
relevant thinking, then the elaboration likelihood is
considered to be high. Under conditions of high
elaboration likelihood a person allocates significant
cognitive resources to considering and critically
evaluating the presented message. If an individual's
motivation or ability to process a message is reduced, then
the elaboration likelihood is also decreased. When it is
low, a person is likely to forego a careful consideration


Table 3
Means with Standard Deviations in Parentheses for CFCR Score
for the Three-Way Interaction of Issue Involvement, Ego-
Involvement, and Argument Quality
Variables
Independent
Dependent
Issue Involvement Ego-
Involvement
Argument Quality
n
CFCR Score
High
High
Strong
11
0.47(0.64)
High
High
Weak
12
0.01(0.64)
High
Low
Strong
13
0.52(0.40)
High
Low
Weak
14
0.12(0.57)
Low
High
Strong
14
0.50(0.61)
Low
High
Weak
14
0.24(0.81)
Low
Low
Strong
11
0.67(0.49)
Low
Low
Weak
9
-0.04(0.89)


126
Chaiken, S. (1980). Heuristic versus systematic
information processing and the use of source versus
message cues in persuasion. Journal of Personality and
Social Psychology, 39, 752-766.
Cohen, A., Stotland, E., & Wolfe, D. (1955). An
experimental investigation of need for cognition.
Journal of Abnormal and Social Psychology, 51, 291-294.
Corrigan, J. D., Dell, D. M., Lewis, K. N., & Schmidt, L.
D. (1980). Counseling as a social influence process:
A review [Monograph]. Journal of Counseling
Psychology, 27, 395-441"!
Crocker, J., Fiske, S. T. & Taylor, S. E. (1984).
Schematic bases of belief change. In J.R. Eiser (Ed.).
Attitudinal judgment (pp. 197-226). New York:
Springer-Ver lag.
Cullen, D. M. (1969). Attitude measurement by cognitive
sampling. (Doctoral dissertation, The Ohio State
University, 1968). Dissertation Abstracts, 29, 1597A.
Dean, R., Austin, J., & Watts, W. (1971). Forewarning
effects in persuasion: Field and classroom
experiments. Journal of Personality and Social
Psychology, 18, 210-221.
Dixon, D. N., & Claiborn, C. D. (1981). Effects of need
and commitment on career exploration behaviors.
Journal of Counseling Psychology, 28, 411-415.
Eagly, A.H. (1967). Involvement as a determinant of
response to favorable and unfavorable information.
Journal of Personality and Social Psychology, 7, (3,
pt. 2), 1-15.
Eagly, A. H., & Mannis, M. (1966). Evaluation of message
and communication as a function of involvement.
Journal of Personality and Social Psychology, 3, 482-
485.
Ellis, A. (1985). Overcoming resistance: Rational-
emotive therapy with difficult clients. New York:
Springer.
Fiske, S. T., & Taylor, S. E. (1984). Social cognition.
Reading, MA: Addison-Wesley.
Frank, J. D. (1961). Persuasion and healing. Baltimore:
John Hopkins.


35
processing of persuasive messages "may be seen as
consisting of schema formation or activation, of the
integration of input with these schemas, and of the
updating or revision of these schema to accommodate new
input" (p. 150). While it is possible that schemas would
enable a person to process messages more objectively, the
literature suggests that when processing is guided by a
schema, it tends to be biased toward the perseverance of
the existing schema (Ross, Lepper, & Hubbard, 1975;
Crocker, Fiske, & Taylor, 1984; Fiske & Taylor, 1984).
Thus, if a person has a great deal of prior knowledge, or a
well-developed schema, he or she would tend to be able to
more effectively counterargue messages opposing their
initial attitudes and to cognitively argue for messages
supporting their initial attitudes (Lord, Ross, & Lepper,
1979) .
It should be noted that a schema as currently defined
within the literature is conceptually similar to George A.
Kelly's (1955) theorizing on personal constructs. Like the
schema theorists he sees people as information processors
interested in the organization of knowledge and prediction
of future events. His Fundamental Postulate states that "a
person's processes are psychologically channelized by the
ways in which he anticipate events" (p. 46). In addition,
the Organization Corollary states that, "each person
characteristically evolves, for his convenience in


60
undergraduate sociology course on marriage and family at
the University of Florida. During the first week of spring
semester 1987, students in one section of Psychology 2013
(General Psychology) and Sociology 2430 (Marriage and
Family) were asked to complete a pretest consisting of
identifying information, the Relationships Grid, and the
Personal Problems Inventory (see Appendices A & B).
On the basis of the scores from these two pretests,
subjects were divided into four separate groups: 1) high
ego and high issue involvement, 2) high ego and low issue
involvement, 3) low ego and high issue involvement, and 4)
low ego and low issue involvement. In order to assign
students to these four groups the following procedure was
followed. First, all students will be rank ordered
according to their summed relationship scores for the
construct, "AssertiveNonassertive" on the Relationships
Grid, and a trichotomization was performed on the basis of
these scores. The uppermost division constituted the high
ego-involvement group; the lowermost division constituted
the low ego-involvement group. Second, all students were
rank ordered according to their scores on the Simple Rathus
Assertiveness Schedule, which is entitled the Personal
Problem Inventory in this study. On the basis of these
scores a median split was performed, and the upper half of
the split consisted of the low involvement group, while the
lower half consisted of a high issue involvement group.


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Assistant Professor of Sociology
This dissertation was submitted to the Graduate Faculty of
the Department of Psychology in the College of Liberal Arts
and Sciences and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of
Doctor of Philosophy.
May, 1987
Dean, Graduate School


76
the cognitive restructuring technique presented in a
message advocating its use in treating assertiveness
problems.
The design was a 2 x 2 x 2 between subjects factorial
design. The factors were ego-involvement (high and low),
issue involvement (high and low), and argument quality
(strong or weak). The dependent variables consisted of the
following: 1) cognitive favorability index, 2) outcome
expectations for the use of the CR technique, 3) behavioral
intention toward cognitive restructuring, and 4)
spontaneous use of CR during covert rehearsal.
The sample was drawn from the University of Florida
introductory psychology subject pool and an undergraduate
sociology class. It consisted of students pretested to
form four distinct groups: 1) high ego involved, high
issue involved; 2) high ego, low issue involved; 3) low
ego, high issue involved; and 4) low ego, low issue
involved.
Instruments used in this study included the
Relationships Grid, Personal Problems Inventory (which
consisted of the Simple Rathus Assertiveness Schedule),
Assertion Survey, Cognitive Assessment Questionnaire (Need
for Cognition Scale), and Thought Listing Techniques for
messages about cognitive restructuring and covert


52
processing may be relatively objective or relatively
biased. As a result of this, individuals with different
levels of involvement will engage in different types of
cognitive processing.
The present study will screen for four groups of
subjects: (1) high ego and high issue involved, (2) high
ego and low issue involved, (3) low ego and high issue
involved, and (4) low ego and low issue involved. Subjects
will be asked to consider a message advocating the use of
the cognitive restructuring counseling technique for the
treatment of assertiveness problems. The topic of
assertiveness was chosen for this investigation because of
its relevance for a college population. Using measures of
CFCR, outcome expectation, behavioral intention, and
spontaneous use of cognitive restructuring to assess the
subjects' cognitive responses to strong and weak arguments
advocating the use of the technique, the following
predictions are made:
1. Replicating the findings of Petty and Cacioppo
(1979b), there will be a significant interaction
between issue involvement and argument quality.
This effect will be such that there will be greater
distinction between strong and weak arguments under
conditions of high issue involvement than under
conditions of low issue involvement.
2. A significant interaction is also predicted
between ego-involvement and argument quality with a
greater distinction between strong and weak
messages occurring under high than low ego-
involvement.
3. A biasing effect is also predicted to be
reflected in a significant interaction between


APPENDIX G
TRANSCRIPT OF TAPED INSTRUCTIONS
Cognitive restructuring is a therapeutic technique
which is based on a simple observation. The things we
think about affect the way we feel. If you were asked to
make yourself feel anxious, for example, you could probably
do this by thinking about an upcoming exam, picturing
yourself giving a talk in front of a large class, or
imagining yourself asking someone new out for a date. The
cognitive restructuring technique relies on the fact that
just as you can make yourself feel upset, you can also make
yourself feel better by controlling the things you think
about and tell yourself. The basic procedure consists of
three steps. First, identifying the things you say to
yourself. Second, assessing how realistic those things
are. And third, changing your unrealistic thoughts to more
realistic or reasonable ones. For example, if you are
thinking of asking someone out for a date, and you are
telling yourself. "He or she will never go out with me.
They will just laugh at me." You could ask yourself how
realistic this thought really is, and then change it to
something more probable like, "They may or may not go out
116


86
Behavioral Intention Rating
The analysis of variance revealed no significant
findings for the behavioral intention score. Means and
standard deviations for the behavioral intention score are
presented in Table 5.
Spontaneous Use of Cognitive Restructuring
Only the post-coding categories of self-instructional
(SI), self-assuring (SA), and self-doubting (SD) statements
were deemed to have adequate inter-judge reliability (range
.81 to .92). These three categories were analyzed by
separate 2x2x2 Analyses by Variance.
Self-instructional statements. The analysis of variance
revealed no significant findings for the SI score.
Self-assuring statements. The analysis of variance
revealed no significant findings for the SA score.
However, the results did reveal a trend toward an
interaction effect of ego-involvement x argument quality
for the SA score, F(l,89)=2.8, p=.10. Examination of the
means suggests that under the high ego-involvement
condition argument quality did not influence the number of
self-assuring statements generated, but under the low ego-
involvement condition subjects who heard the message with
strong agruments generated significantly less self assuring
statements than subjects who heard the message with weak
arguments.


8
concern the presence or absence of a well-developed
construct for assertiveness. The literature suggests that
such a well-developed construct may bias individuals with
assertiveness problems against change. Such a process may
be one source of resistance in counseling.


55
Cognitive responses to strong and weak arguments for
cognitive restructuring were assessed through the use of
the Thought-Listing Technique (Cacioppo and Petty, 1981).
Several seven-point Likert-type scales on which subjects
rated their attitudes toward cognitive restructuring and
their intentions to use it were also used to measure their
acceptance or rejection of the arguments. An additional
Thought-Listing Technique was used to assess the
spontaneous use of cognitive restructuring by subjects
during a covert rehearsal technique.
This chapter is divided into several sections. In the
first section, the research design and variables of
interest are discussed. In the second section, the
population of interest and the study's sample and sample
selection and recruitment procedures are discussed. In the
third section, the instruments used in this study are
presented. In the fourth section, the experimental
procedure and data collection is described. In the fifth
section, the method of data analyses are presented.
Methodological limitations of the study and a summary of
the methodology are presented in the final two sections.
Research Design
This study involved a 2 x 2 x 2 (Ego-involvement: high
or low X Issue involvement: high or low X Intervention
Quality: strong or weak arguments) factorial design.
Subjects were crossed on the first two factors according to


45
counterarguments to the message. Therefore, as issue
relevant thinking increases, counterargumentation and
resistance to persuasion also increases. On the other
hand, if the persuasive message is congruent with the
individual's initial attitude, it is likely that the
individual is motivated and able to generate favorable
cognitions. As involvement and issue-relevant thinking
increases in this situation, more favorable thoughts might
be generated, and increased persuasion would result.
In order to test these hypotheses Petty and Cacioppo
(1979a) conducted two experiments. In the first experiment
subjects were exposed to either a proattitudinal or
counterattitudinal message concerning coed visitation
hours. Issue involvement was manipulated by stating that
the changes in visitation hours would go into effect at
their own university (high involvement) or at a distant
university (low involvement). Dependent measures consisted
of attitude scales and thought listings. Results on the
attitude index revealed a significant interaction between
involvement and type of message. Analysis of the
interaction revealed that increased involvement increased
agreement with the proattitudinal message but decreased
agreement with the counterattitudinal message. The thought
listing technique revealed similar results. Under high
involvement conditions subjects produced more positive
cognitions and fewer counterarguments to the proattitudinal


49
will respond with cognitions aimed at preserving the status
quo. Thus, extremely high levels of involvement, or
personal importance for a person's central aspects of
identity, may also lead to biased processing. Biased
processing may then consist of either a negative bias for
counterattitudinal messages or a positive bias for
proattitudinal messages.
Involvement Level in Counseling
Clear cases of high or low involvement probably do not
exist in the therapy room. People generally seek help with
issues in which they are involved to some significant
degree. Prior to pursuing therapy, they will have given
thought to how to deal with the issue and perhaps will have
tried many self-generated change attempts. They may even
have sought the advice of friends, family, and others.
Strong (1971) suggested that different types of
counseling may elicit different levels of involvement from
clients, but it seems clear that clients may also bring
different levels of involvement with them to counseling.
Stoltenberg and McNeill (1984) showed that college students
who were undecided about a career regarded the issue of
career exploration as more personally involving than
students who had already chosen a career. They also found
undecided students agreed more with a message advocating a
career exploration course than did decided students.


73
compared to the individual working alone. This finding
corresponds to the well-known "social loafing" argument
(Latane and Darley, 1970).
Cognitive Assessment Questionnaire
The Cognitive Assessment Questionnaire (CAQ; see
Appendix F) was used to assess subjects' tendency to engage
in and enjoy effortful cognitive endeavors. It consisted
of the short form of the Need for Cognition Scale (NCS)
(Cacioppo, Petty, and Kao, 1984). The short NCS is an 18-
item self-report inventory which provides a global
assessment of need for cognition. Cacioppo, Petty, and Kao
(1984) report a significant correlation (r = .95, p <,001)
between subjects' scores on original and 18-item versions
of the NCS. The NCS shows strong internal consistency (r =
.76, p<.001), and some evidence exists for its content and
predictive validity (Cacioppo & Petty, 1982).
Procedures
Procedures were set up to insure standardization of
administration. Instruments were presented to each subject
by cassette tape (see Appendix G for a complete transcript
of the tapes for each experimental condition).
Subjects were run in group administrations in a
language laboratory classroom. They were seated and
provided with a package of written materials, the measures
of interest. They were then instructed by the researcher
to place their headphones on and follow the instructions


CHAPTER IV
RESULTS
Manipulation Checks
Before investigating whether the manipulations produced
the desired effects on the dependent measures, the
effectiveness of the manipulations should be ascertained.
Manipulation checks were planned for argument quality and
issue involvement and analyzed using a series of 2 x 2 x 2
analyses of variance.
Argument quality
To see whether subjects differentiated the strength of
the arguments used in the message quality manipulation,
subjects answered a seven-point Likert item asking them to
rate the quality of the arguments in the intervention:
very poor arguments (1) to very good arguments (7).
Results of the Analyses of Variance indicate that the
manipulation successfully influenced how strongly the
subjects rated the quality of the two groups of arguments,
F(1,96)=30.43, p<.0001: strong arguments M=5.0, weak
arguments M=3.5. The involvement variables (issue and ego)
did not significantly affect this manipulation check,
either alone or in interaction with the argument quality
manipulation.
78


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EFFECTS OF ISSUE INVOLVEMENT AND
EGO-INVOLVEMENT UPON ACCEPTANCE OF A
COUNSELING MESSAGE
BY
STEVEN HUGH WILLIAMS
May, 1987
Chairman: Dr. Greg J. Neimeyer
Major Department: Psychology
Within the general framework of the Elaboration
Likelihood Model of persuasion, the present study
investigated the influence of issue involvement, ego-
involvement, and argument quality upon the cognitive
processing of a counseling message advocating the use of
cognitive restructuring in the treatment of assertiveness
problems. Issue involvement was measured by the total
number of relevant thoughts governed by subjects of the
Thought Listing Technique. It was predicted that there
would be greater distinction of argument quality under
conditions of high issue involvement than under conditions
of low issue involvement.
Ego-invo1vement was measured by the proportion of
total variance accounted for by a single construct
vi


59
to you, personally?" As a message of behavioral intention,
subjects were asked to indicate "How likely are you to use
the cognitive restructuring technique in the future in
situations requiring assertion"? Subjects responded to
those items as well as a number of manipulation check and
ancillary items on a seven point Likert-type scale. It was
expected that analysis of outcome expectation ratings would
parallel the predictions for CFCR. In addition, under
conditions of high issue involvement it was expected that
subjects with low ego-involvement would have greater
intentions to use the CR technique than subjects with high
ego-involvement.
Finally, the spontaneous use of cognitive
restructuring during a covert rehearsal condition (see
Appendix E) was also used to operationalize subjects'
cognitive responses to the CR arguments. It was reasoned
that subjects with greater favorability toward cognitive
restructuring and greater behavioral intention to use it
will generate more CR thoughts during a covert rehearsal of
a situation requiring assertion.
Description of the Sample
The population of interest in this study included all
undergraduate students attending the University of Florida,
Gainesville, Florida.
The sample in this study was drawn from the University
of Florida introductory psychology subject pool and from an


13
providing weak counterarguments to his or her attitude
accompanied by refutations. McGuire states that the
inoculation will pose a cognitive threat that will motivate
the individual to generate supportive arguments and
counterargument refutations for his or her original
attitude. This practice should produce greater resistance
to subsequent persuasion attempts. The point of interest
for the cognitive response approach is, does inoculation
lead to self-generated cognitive responses and thus greater
resistance to persuasion. McGuire and Papagioris (1961)
asked subjects to list thoughts in favor of their initial
position one week after being exposed to either supportive
defenses or refutational defenses (the inoculation
treatment). Subjects who had been exposed to the
refutational defenses listed more supportive thoughts than
subjects who had been exposed to supportive defenses. In
an experimental investigation of the effectiveness of
active and passive participation in the defense of one's
beliefs McGuire (1964) provided additional evidence for the
importance of one's own cognitive responses in mediating
attitudinal changes. In the active participation condition
he assigned subjects the task of writing either a
supportive or refutational defense. In the passive
participation condition subjects simply read a defense
provided by the investigator. The beliefs used in this
study were cultural truisms (e.g.,
mental illness is not


38
that high prior knowledge subjects were overall more
resistant to the different messages than low prior
knowledge subjects and that this effect was stronger for
weak than strong arguments. This result is consistent with
the notion that prior knowledge would enhance a person's
ability to counterargue an incongruent message. With weak
arguments it would be easier to counterargue than with
strong arguments. In addition, low prior knowledge
subjects' attitudes were affected by argument length, but
high prior knowledge subjects' attitudes were not.
In sum, research on forewarning and prior knowledge has
provided support for the view that when a person has a
well-developed level of organized knowledge on a topic,
message processing will be biased. This will occur,
because prior knowledge enables the counterarguing of
incongruent messages and the strengthening of congruent
ones. The presence of schemas or superordinate constructs
are indicators of such well-developed levels of prior
knowledge.
Involvement and Persuasion
In the preceding sections numerous variables (e.g.,
need for cognition, forewarning) which may affect cognitive
processing in either a relatively objective or relatively
biased manner have been discussed. One of the most
important variables which can affect the extent and type of
cognitive processing employed is personal involvement. In


26
or a relatively biased manner. In objective processing,
some variable "either motivates or enables subjects to see
the strengths of cogent arguments and the flaws in specious
ones, or inhibits them from doing do" (p. 136). In biased
processing some variable "either motivates or enables
subjects to generate a particular kind of thought in
response to a message, or inhibits a particular kind of
thought" (p. 136). Objective processing appears to have
much in common with "bottom-up" or "data-driven"
processing, since the elaboration is primarily impartial
and data focused. On the other hand, biased processing
appears to share similarities with "top-down" or "theory-
driven" processing, as the elaboration may be guided by
prior knowledge, such as a relevant attitude schema
(Landman & Manis, 1983; Fiske & Taylor, 1984).
Argument processing lies at the heart of the ELM
(postulate 3) along with two other critical constructs:
argument quality and peripheral cues. While peripheral
cues are not investigated in this study, argument quality
is used as a manipulation to assess the degree to which
variables of involvement affect cognitive processing in
either a relatively objective or biased manner. In
accordance with the ELM, a persuasive message with strong
arguments should elicit more agreement when it is
considered carefully (high elaboration) than when
consideration is low (low elaboration), but a persuasive


Table 2. Analysis of Variance for CFCR Scores
Source
Sum of squares
df
Mean Square
F
P
Issue Involvement
0.989
1
0.989
0.22
.642
Ego
0.004
1
0.004
0.01
.919
Quality
5.011
1
5.011
12.21
.001
Issue Involvement x Ego
0.109
1
0.109
0.27
.607
Issue Involvement x Quality
0.016
1
0.016
0.04
.845
Ego x Quality
0.221
1
0.221
0.54
.465
Issue x Ego x Quality
0.374
1
0.374
0.91
. 342
Error
36.929
90
0.410
03


79
In addition, as predicted by the ELM subjects who heard
the strong argument quality message generated a
significantly greater proportion of positive thoughts
(M=.76) than did subjects who heard the weak message
(M=.55), F(l,90) = 12.21, p<.0007. Furthermore, subjects
who heard the message with weak arguments generated a
significantly greater proportion of negative thoughts
(M=.45) than did subjects who heard the strong message
(M=.23) F(1,90) = 12.21, p<.0007.
Issue Involvement
The other category of manipulation checks investigated
the degree of issue involvement that subjects believed that
they had with the use of the cognitive restructuring
technique for treating personal problems in assertion. As
an initial check on issue involvement a seven-point Likert
item asking subjects to indicate the extent to which the
message had implications for them personally was included.
An analysis of variance revealed a trend toward a main
effect of issue involvement, F(l,96)=2.8, p=.10; high issue
M=4.25, low issue M=3.72. All other variables (ego and
argument quality) also did not significantly affect this
manipulation check. As these results cast doubt on the
effectiveness of the issue involvement manipulation, an
alternative operationalization of involvement was pursued.
Petty and Cacioppo (1986) suggest that as personal
involvement increases, people become more motivated to


110
Please list the thoughts you had while listening to the
tape, listing one thought per line.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18 .
19.
20.
21.
22.
23.
24


30
Repetition is another variable which, in moderate
amounts, can increase cognitive processing of information
in a relatively objective way. Once again, Cacioppo and
Petty (1985) used strong and weak arguments for the
institution of a senior comprehensive exam; this time,
however, half of their college student subjects heard the
message once, and half heard the message three times in
succession. Students showed greater attitudinal
differentiation of strong from weak messages when the
message was repeated three times rather than just presented
once. Similar results on the impact of repetition on
persuasion are reported by Petty and Cacioppo (1984).
Personal responsibility has also been found to enhance
objective processing of issue relevant arguments. Social
psychological research on the "social loafing" effect
(Latane & Darley, 1970; Latane, Williams, & Harkins, 1979)
suggests the classical finding of reduced personal
responsibility in group performance in comparison to
individual performance may result from a loss of motivation
in the group setting. For the persuasion context this
implies that if personal responsibility is increased, then
motivation to engage in significant cognitive processing
should also increase. In a test of this hypothesis Petty,
Harkins, and Williams (1980) asked college students to
provide peer feedback on editorial messages ostensibly
written by journalism students. Subjects were led to


CHAPTER V
DISCUSSION
The primary purpose of the present investigation was
to examine the effects of issue involvement and ego-
involvement upon the cognitive responses that individuals
generate to arguments advocating the use of cognitive
restructuring to treat problems in assertiveness.
Hypotheses for these effects were developed from a review
of the social psychological literature on the Elaboration
Likelihood Model of Attitude Change. The present study
attempted to extend this model to include messages that may
be used in a counseling context.
In discussing variables that can affect a person's
ability to process message arguments Petty and Cacioppo
(1986) cite personal relevance/involvement as "perhaps the
most important variable in this regard" (p. 144). Their
ELM suggests that as issue involvement increases, people
become more motivated to engage in the cognitive work
necessary to evaluate the issue-relevant arguments
presented in a message. Petty and Cacioppo (1979b) provide
evidence consistent with this view. The first hypothesis
of the present study was also based upon this view. More
specifically, it predicted a significant interaction
89


81
Cognitive Favorability (CFCR)
The analysis of variance revealed a significant main
effect of argument quality for the CFCR score,
F(l,90)=12.21, p<.001, such that subjects who heard the
strong arguments had a greater proportion score for overall
cognitive favorability toward cognitive restructuring
(M=0.54) than did subjects who heard the weak arguments
(M=0.10) (see Table 2). Means and standard deviations for
CFCR for all levels of the design are presented in Table 3.
Attitude Rating
The analysis of variance revealed a significant main
effect of issue involvement for the outcome expectation
attitude rating, F(1,96)=9.23, p<.003, such that subjects
in the high issue involvement condition believed that the
cognitive restructuring technique would be significantly
more beneficial to them (M=4.58) than did subjects in the
low issue involvement condition (M=3.76) (see Table 4).
Means and standard deviations for attitude rating are
presented in Table 5.
The results also revealed a trend toward a main effect
of argument quality for the attitude rating, F(1,96)=3.54,
p=.06. These results suggest that the group that heard the
strong arguments had a tendency to believe that the
cognitive restructuring technique would be more beneficial
to them (M=4.4) than did the group that heard the weak
arguments (M=3.9).


122
thoughts that you had putting only one thought per space.
Please turn the page now and begin.
(Silence lasting three minutes)
Please stop and turn to the next page and follow the
instructions at the top of the page. You will have two
minutes to complete those instructions. Then please listen
for further instructions.
(Silence lasting two minutes)
Please stop and turn to the next page and follow the
instructions at the top of the page. You will have five
minutes to complete those instructions. Then please listen
for further instructions.
(Silence lasting five minutes)
Now that you have completed your ratings please turn to
the next page and read the instructions with me.
As a final part of this study, we would like for you to
take a minute to imagine a situation that requires that you
take a risk and express yourself. The situation that we
would like for you to imagine is that of you giving a ten
minute oral class presentation on the thing that you value
most in life. What we would like for you to do is to take
the next thirty seconds to conjure up the clearest possible
image that you can of that situation. Imagine a picture of
yourself, the room, the other students, your professor,
etc. Do this now.


127
Frankel, P. B. (1977). A factor analytic study of measures
of assertiveness. (Doctoral dissertation, California
School of Professional Psychology, 1976) Dissertation
Abstracts International, 37, 4676B-4677B.
Freedman, J. D. (1964). Involvement, discrepancy, and
change. Journal of Abnormal and Social Psychology, 69,
290-295.
Freud, S. (1939). The interpretation of dreams. In A. A.
Brill (Ed. and Trans.), The Basic Writings of Sigmund
Freud. (pp.179-549). New York: Random House.
(Original work published 1900).
Freud, S. (1959). The defence neuro-psychoses. In E.
Jones (Ed.) and J. Richman (Trans.), Sigmund Freud:
Collected Pages (Vol. 1, pp. 59-75). New York:
Basic Books. (Original work published 1894).
Futch, E. J., Scheirer, C. J., & Lisman, S. A. (1982).
Factor analyzing a scale of assertiveness: A critique
and demonstration. Behavior Modification, 6_, 23-43.
Greenwald, A. G. (1968). Cognitive learning, cognitive
responses to persuasion, and attitude change. In A.
G. Greenwald, T.C. Brock, & T.M. Ostrom (Eds.),
Psychological Foundations of Attitudes, (pp. 147-170).
New York: Academic Press.
Greenwald, A. G. (1980). The totalitarian ego:
Fabrication and revision of personal history. American
Psychologist, 35, 603-618.
Greenwald, A. G., & Albert, R. D. (1968). Acceptance and
recall of improvised arguments. Journal of Personality
and Social Psychology, 8_, 31-34.
Hass, R. G., & Grady, K. (1975). Temporal delay, type of
forewarning, and resistance to influence. Journal of
Experimental Social Psychology, 11, 459-469.
Heesacker, M. (1986). Counseling pretreatment and the
elaboration likelihood model of attitude change.
Journal of Counseling Psychology, 33, 107-114.
Heimberg, R. G., & Harrison, D. F. (1980). Use of the
Rathus Assertiveness Schedule with offenders: A
question of questions. Behavior Therapy, 11, 278-281.


68
Psychometric research on the RAS suggests that it has
moderately high stability (test-retest reliability) and
moderate to high homogeneity. Rathus (1973) presents a
five-week test-retest reliability of 0.78 and a split-half
Pearson product moment correlation of 0.77 between total
odd and total even item scores. Vaal (1975) presents
similar correlations of 0.76 for test-retest reliability
over an eight-week period and 0.77 for split-half internal
consistency. With college students seeking assertion
training Heimberg and Harrison (1980) obtained over an
eleven to fifteen day period a test-retest reliability
correlation of 0.80 and a split-half internal consistency
correlation also of 0.80. Other reports of internal
consistency correlations have ranged from 0.69 to 0.86
(Mann and Flowers, 1978; Quillan, Besing, and Dinning,
1977; Futch, Scheirer, and Lisman, 1982).
A base of validity information has also been
established for the RAS. When comparing RAS scores with
peer ratings on a seventeen-item semantic differential
Rathus (1973) found that total RAS scores correlated
positively with each of the items comprising the peer
rating scale's assertiveness factor. Several studies have
tested Wolpe's (1973) argument of an inverse relationship
between trait anxiety and assertiveness; conformation of
Wolpe's argument would provide indirect support for the
construct validity of the RAS. Orenstein, Orenstein, and


128
Heppner, P. P., & Dixon, D. N. (1978). The effects of
client perceived need and counselor role on client's
behavior. Journal of Counseling Psychology, 25, 514-
519.
Heppner, P. P., & Dixon, D. N. (1981). A review of the
interpersonal influence process in counseling.
Personnel and Guidance Journal, 59, 542-550.
Heppner, P., & Heesacker, M. (1982). Interpersonal
influence process in real-life counseling:
Investigating client perceptions, counselor experience
level, and counselor power over time. Journal of
Counseling Psychology, 29, 215-223.
Hinkle, D. (1966). The change of personal constructs from
the viewpoint of a theory of construct implications.
(Doctoral dissertation, The Ohio State University,
1965). Dissertation Abstracts, 26, 6837.
Hollandsworth, J. G. (1976). Further investigation of the
relationship between expressed social fear and
assertiveness. Behavior Research and Therapy, 14, 85-
87.
Honess, T. (1977). A comparison of the implication and
repretory grid techniques. Unpublished manuscript,
University of Exeter, England.
Hovland, C. I. (1951). Changes in attitude through
communication. Journal of Abnormal and Social
Psychology, 46, 424-437.
Hovland, C. I., Harvey, 0. J., & Sherif, M. (1957).
Assimilation and contrast effects in reactions to
communication and attitude change. Journal of Abnormal
and Social Psychology, 55, 244-252.
Janis, I. L., & King, B. (1954). The influence of role-
playing on opinion change. Journal of Abnormal and
Social Psychology, 49, 211-218.
Jones, E. E. (1985). Major developments in social
psychology during the past five decades. In G. Lindzey
& E. Aronson (Eds.), Handbook of Social Psychology:
Vol. I. Theory and Method (pp. 47-107). New York:
Random House.
Kaplan, D. A. (1982). Behavioral, cognitive, and
behavioral-cognitive approaches to group assertion
training therapy. Cognitive Therapy and Research, 6,
301-314.


11
treatments was to increase the consumption of unusual meats
(hearts, kidneys, etc.) during World War II. Groups of
housewives either listened to a persuasive lecture or
participated in a group discussion on the material
presented in the lecture. Survey results showed that only
3 percent of the women in the lecture group served one of
the meats, while 32 percent of the women in the discussion
group served them. While the results of this study are
open to different interpretations, it may be that the self
generated arguments of the women in the discussion group
were more persuasive than the arguments in the lecture.
In later research on the role of active and passive
participation in persuasion, Janis and King (1954) reported
that subjects changed their attitudes more in the advocated
direction when instructed to give an informal talk on a
counterattitudinal topic than did other subjects who
passively listened to the talks. In their experiment Janis
and King used three different topics. Their results
revealed significantly greater attitude change for two of
these topics in comparison to the third. From their data
and subject interviews they observed that for the two
topics with greater change subjects improvised more in
their talks. For the third topic subjects stayed close to
the prepared outline made available for each topic. In
this case improvisation, or self-generated cognition, may
have operated as a critical factor in the greater attitude


74
that they heard on the tape carefully. All further
instructions to the subject were on the taped message.
The taped message consisted of a general introduction
to the study and the cognitive restructuring technique for
all subjects. Half of the subjects in each involvement
level group were randomly assigned to one of two
experimental conditions, either strong arguments in favor
of the effectiveness of cognitive restructuring or weak
arguments favoring cognitive restructuring. (See Petty &
Cacioppo, 1984, for a discussion of the development of
these messages). The next section of the tape that the
subjects heard presented either strong or weak arguments
for cognitive restructuring depending upon the experimental
condition being conducted. This was followed by
instructions for the completion of the Assertion Survey
followed by three minutes of silence to enable the
completion of the questions. Next, instructions for the
thought listing for the message were presented and followed
by three minutes of silence during which subjects were
instructed to "List all of the thoughts that came to mind
as you listened to the taped message." Instructions for
the Cogntive Assessment Questionnaire were presented next
and followed by five minutes of silence during which the
subjects completed the questionnaire. The tape resumed
with instructions for covert rehearsal of a personal
assertion situation and its associated thought listing,


71
obtaining a written listing of an individual's thoughts on
a topic and a general index of CFCR toward the topic
(Cacioppo and Petty, 1981). In the present study Thought
Listing was employed under two conditions: 1) after
listening to the taped arguments for cognitive
restructuring and completing the Assertion Inventory and 2)
as part of the covert rehearsal of a personal assertion
situation. Under the first condition the Thought Listing
Technique (see Appendix D) was used to compute a CFCR
index, and under the second condition (see Appendix E) it
was used to tally the spontaneous use of cognitive
restructuring. Under the first condition subjects listed
their thoughts concerning the arguments for CR according to
directions similar to those of Petty and Cacioppo (1977).
In addition, they rated their thoughts along the dimension
of favorableness toward cognitive restructuring. The
cognitive favorability index was calculated for each
subject by subtracting the number of unfavorable thoughts
from the number of favorable thoughts listed and then
dividing by the total number of relevant thoughts (Cacioppo
and Petty, 1981). Petty et al. (1976) report correlations
between subjects' own rating and independent judges ratings
of .82 for favorable thoughts and .79 for counterarguments.
Under the second condition, two undergraduate psychology
major students trained by the researcher and unaware of the
hypotheses of the study and the involvement group


CHAPTER II
REVIEW OF THE LITERATURE
There are several areas of literature and empirical
research relevant to the current study. First, evidence
for a cognitive response approach to attitude change and
persuasion will be presented. Second, the Elaboration
Likelihood Model, a general cognitive model of persuasion,
will be presented. Next, the role of cognitive processing
will be further explored. This will include both objective
and biased modes of processing. At this point the variable
of involvement will be introduced, and its contribution to
the attitude change literature will be reviewed. The
relationships between involvement and cognitive processing
will then be explored, and its applications in a counseling
context will be considered. Finally, specific hypotheses
will be derived from this discussion.
The Cognitive Response Approach To Persuasion
Whenever people receive a persuasive message or any
other communication, they will attempt to place the new
information within the context of their existing knowledge
of the topic. This is the basic postulate of the
cognitive response approach presented by Greenwald (1968).
In thinking about the message, the individual may consider
9


37
With counterattitudinal message it would be expected
that prior knowledge would increase a person's ability to
counterargue the message. To test this hypothesis Wood
(1982) divided subjects into high and low prior knowledge
groups for the topic of environmental preservation. One to
two weeks later subjects read a counterattitudinal message
consisting of four arguments against environmental
preservation. Attitude measure and thought listing
revealed the subjects who had high prior knowledge changed
less in the advocated direction than did subjects with low
prior knowledge. In addition, high prior knowledge
subjects generated more counterarguments and fewer
favorable thoughts toward the message.
Wood, Kallgren, and Priesler (1985) extended this
finding by adding argument quality and message length
manipulations. They divided subjects into high, medium,
and low prior knowledge about environmental preservation
groups and exposed them to one of four persuasive messages.
Two messages consisted of three strong arguments against
preservation, and two messages consisted of three weak
arguments against preservation. In the message length
manipulation two versions of each type of argument were
developed. One version contained short concise statements,
while the other contained more wordy versions of the same
arguments. Both versions were equal in argument strength
and ease of comprehension. Attitude measure results reveal


115
11. I really enjoy a task that involves coining up with
new solutions to problems.
12. Learning new ways to think doesn't excite me very
much.
13. I prefer my life to be filled with puzzles that I
must solve.
14. The notion of thinking abstractly is appealing to
me.
15. I would prefer a task that is intellectual,
difficult, and important to one that is somewhat
important but does not require much thought.
16. I feel relief rather than satisfaction after
completing a task that required a lot of mental
effort.
17. It's enough for me that something gets the job
done; I don't care how or why it works.
18. I usually end up deliberating about issues even
when they do not effect me personally.


APPENDIX D
THOUGHT LISTING TECHNIQUE
Thoughts Regarding The Message
We would now like to get an idea of the thoughts that
crossed your mind while listening to the tape. The next
page contains the form we have prepared for you to use to
record your thoughts and ideas. Simply write down the
first idea that comes to mind on the first line, the second
idea on the second line, etc. Please put only one idea or
thought on a line. You should try to record those thoughts
and ideas as concisely as possiblea phrase is sufficient.
Ignore spelling, grammar, and punctuation.
You will have three minutes to write down your
thoughts. We have deliberately provided more space than we
think most people will need to insure that everyone would
have plenty of room to write his/her ideas and thoughts.
So don't worry if you don't fill every space. Just write
down whatever your thoughts were while listening to the
tape. Please be completely honest and list all of the
thoughts that you hadputting only one thought per space.
Do not turn the page until instructed to do so.
109


BIOGRAPHICAL SKETCH
Steven Hugh Williams was born on June 10 1957 in
Clarksburg, West Virginia. In 1975, he graduated from
Bridgeport Senior High School in Bridgeport, West Virginia.
He received his Bachelor of Arts degree in psychology in
1979 from West Virginia University in Morgantown, West
Virginia.
In August, 1981, Steven began graduate studies in
counseling psychology at the University of Florida. He
received his Master of Science degree in 1984. In August,
1986, Steven completed his predoctoral internship at the
Veterans Administration Medical Center in Bay Pines,
Florida. He expects to graduate with the Doctor of
Philosophy degree in May, 1987.
On May 7, 1983, Steven married Barbara Jessie Heier.
Barbara is currently employed as a pediatric intensive care
nurse at Shands Teaching Hospital at the University of
Florida. They have one son, Adam Benjamin.
135


102
10. Yourself in your relationship with a former friend
of either sex whom you once thought was a close
friend but in whom you were later disappointed.
Each relationship is identified by a single word in the
columns along the top of the grid. They are presented from
left to right in the same order as above.
What you are to do is to think about your relationships
and rate them along the ten descriptors which appear along
the right side of the grid. You will use the rating scale
at the top of the descriptors. (+3 +2 +1 0 -1 -2 -3). For
example, in the first case you will be considering your
relationship with your mother along the descriptor, "calm
anxious." If you see yourself as very calm in your
relationship with her, place a +3 in the first box on the
grid. Use a +2 if you see yourself as fairly calm; use a
+1 if you see yourself as slightly calm with your mother.
A "0" rating is a neutral, whereas -1, -2, and -3 indicate
that you see yourself as correspondingly more anxious in
the relationship. When you have finished this rating move
down to the next descriptor and indicate how "respectful--
not respectful" you consider yourself to be with your
mother. Continue in this way until you have rated your
relationship with your mother along all ten descriptors.
Then move over to the next column and consider your
relationship with your roommate and rate yourself in that
relationship along each of the ten descriptors. Continue
your ratings in this way until you have rated each
relationship along all of the descriptors and every box in
the grid is filled in.
Please turn this page over and begin your ratings.


29
To test this hypothesis Petty, Wells, and Brock (1976)
presented college students with either strong or weak
arguments for the proposal that their tuition be raised by
20 percent accompanied by a distraction task. Their
distraction task consisted of having subjects record the
quadrant in which Xs flashed on a screen in front of them.
Distraction varied according to the rate of presentation of
the Xs with four levels: l)no distraction, 2) low
distraction (15 second intervals), 3) medium distraction (5
second intervals), and 4) high distraction (3 second
intervals). After hearing the messages subjects completed
attitude measures and listed their thoughts during the
message. Attitude rating results revealed a significant
argument quality X distraction interaction which is
consistent with the thought disruption hypothesis. Under
conditions of weak argument quality increasing distraction
was associated with more favorable attitudes; however,
under conditions of strong argument quality increasing
distraction was associated with less favorable attitudes.
Analyses of the thought listings showed that high
distraction decreased counterargument generation for the
weak message but not for the strong one. Moreover, high
distraction tended to decrease the amount of favorable
thoughts elicited by the strong message but not the weak
one. For similar results see Lammers and Becker (1980).


19
would hear a message recommending that stricter visitation
hours be instituted (a counterattitudinal message). They
found that IEMG activity increased in all conditions during
the presentation of the message, which would be expected if
IEMG activity indicated the presence of silent language
processing. The finding of interest was that IEMG activity
had increased following the forewarning of the upcoming
counterattitudinal message.
In summary, early research prior to the formal
development of the cognitive response approach to pesuasion
supported the importance of people's construing of the
information in persuasive messages, rather than their
hearing of the material per se, as the most important
factor in determining their susceptibility or resistance to
persuasion. The methodologies described above have all
been used in research on a general framework for
understanding persuasion called the Elaboration Likelihood
Model (Petty & Cacioppo, 1986), which will be described
next.
The Elaboration Likelihood Model
The Elaboration Likelihood Model (ELM) of attitude
change is a general response theory, which arose out of
Petty and Cacioppo's (1981) attempts to account for the
differential persistance of communication-induced attitude
change. They hypothesized that two distinct routes to
persuasion existed, which they called the central and


APPENDIX C
ASSERTION SURVEY
Please circle the number that best represents your answer
to each question below:
1. How effective do you think this technique would be in
helping most people to become more assertive?
(Not at all effective) 1234567 (very effective)
2. To what extent do you think the cognitive restructuring
technique would be beneficial to you, personally?
(Not at all beneficial) 1234567 (Very beneficial)
3. How likely are you to use the cognitive restructuring
technique in the future in situations requiring
assertion?
(Not at all likely) 1234567 (Very likely)
4. Please rate the following aspects of the tape:
a. Voice Quality
Good 1234567 Bad
b. Rate of Delivery
Too Fast 1234567 Too Slow
c. Enthusiasm for Topic
Good 1234567 Bad
5. To what extent did the tape hold your attention?
Not at all 1234567 Very much
6. To what extent did the taped message have implications
for you personally?
Not at all relevant to me 1234567 Extremely relevant
to me
107


25
4. Affecting motivation and/or ability to process
a message in a relatively objective manner can do
so by either enhancing or reducing argument
scrutiny (p. 138) .
5. As motivation and/or ability to process
arguments is decreased, peripheral cues become
relatively more important determinants of
persuasion. Conversely, as argument scrutiny is
increased, peripheral cues become relatively less
important determinants of persuasion (p. 152).
6. Variables affecting message processing in a
relatively biased manner can produce either a
positive (favorable) or negative (unfavorable)
motivational and/or ability bias to the issue
relevant thoughts attempted (p. 163).
7. Attitude changes that result mostly from
processing issue-relevant arguments (central route)
will show greater temporal persistence, greater
prediction of behavior, and greater resistance to
counterpersuasion than attitude changes that result
mostly from peripheral cues (p. 175).
While reviewing the accumulated empirical evidence for
the ELM is beyond the scope of this chapter, relevant
research will be presented in the discussions of cognitive
processing and the variable of involvement.
Cognitive Processing
In the ELM Petty and Cacioppo (1981) propose that two
major cognitive routes, the central and the peripheral, are
used by people to process persuasive messages. The
understanding of these cognitive processes involves the
analysis of the structures and mechanisms that comprise
mental activity as well as the impact of various source,
message, and recipient variables upon mental activity.
Petty and Cacioppo (1986) further propose that variables
can affect the cognitive processing of message arguments so
that argument processing proceeds in a relatively objective


Table 1. Means with Standard Deviations in Parentheses for Levels of
Issue Involvement and Ego-Involvement
Variables
Independent
Descriptive
Issue Involvement
Ego-Involvement
n
%M
Rathus Score
Variance Score
High
High
30
24
77.08(10.60)
115.46(16.14)
High
Low
22
38
78.12 (6.11)
27.09(13.32)
Low
High
24
58
110.293(8.15)
119.80(24.81)
Low
Low
28
32
111.80 (8.50)
21.31(11.55)
N)


3
contained in a persuasion message. Elaboration likelihood
would be high if a person is motivated and able to evaluate
a message, and it would be low if a person is not motivated
and/or not able to evaluate a message. People, of course,
vary in their willingness and ability to evaluate messages
due to individual and situational factors. The ELM began
with Petty and Cacioppo's attempts to account for the
differential persistence of communication-induced attitude
change. .After reviewing the literature on attitude
persistence, they hypothesized that two distinct routes to
persuasion existed (Petty and Cacioppo, 1981). The first
route, which they called the central route, involves an
individual's thoughtful consideration and personal
evaluation of the information presented in the
communication. The second route, or peripheral route,
involves an individual's selection from the persuasion
context of some cue (e.g., counselor trustworthiness) that
produces change without the individual having to consider
or evaluate the information presented. The central and
peripheral routes to persuasion correspond generally to
high and low elaboration likelihood, respectively. Petty
and Cacioppo found the central route to persuasion to be
the more enduring of the two. They further postulate that
"variables can affect the amount and direction of attitude
change by : (A) serving as persuasive arguments, (B)
serving as peripheral cues, and/or (C) affecting the extent


16
In an experiment utilizing the first approach, Petty
and Cacioppo (1977) forewarned some subjects and did not
forewarn others that they would be hearing a tape recording
prepared by the "Faculty Committee on Academic Affairs"
which recommended that seniors be required to pass a
comprehensive exam in their major as a prerequisite for
graduation. Half of the subjects completed a "thought
listing" technique (Cacioppo & Petty, 1981) in the latter
2.5 minutes of the 5 minute interval between the
forewarning and the message. Attitude survey results
indicated that those subjects who were forewarned, whether
or not they listed their thoughts, showed greater
resistance to persuasion than those who had not been
forewarned. Analyses of the thought-listings showed that
the forewarning elicited anticipatory counterargumentation.
Forewarned subjects listed twice as many thoughts that were
unfavorable than were favorable toward the recommendation.
Subjects who had not been forewarned did not generate any
issue-relevant thoughts, as they were unaware of the
upcoming message's topic.
In the second experiment reported in Petty and Cacioppo
(1977), they manipulated issue-relevant thinking. In this
study students in an introductory psychology class were
told that a guest lecturer, a psychologist from the
University Counseling Center, would speak to their class
that day, and in return for the visit they would complete a


125
Cacioppo, J. T., & Petty, R. E. (1981). Social
psychological procedures for cognitive response
assessment: The thought listing technique. In T.
Merluzzi C. Glass, & M. Genest (Eds.), Cognitive
assessment. New York: Guilford.
Cacioppo, J. T., & Petty, R. E. (1982). The need for
cognition. Journal of Personality and Social
Psychology, 42, 116-131.
Cacioppo, J. T., & Petty, R. E. (1984). The need for
cognition: Relationship to attitudinal processes. In
R. P. McGlynn, J. E. Maddux, C. D. Stoltenberg, & J. H.
Harvey (Eds.), Social perception in clinical and
counseling psychology (pp. 113-139). Lubbock, TX:
Texas Tech Press.
Cacioppo, J. T., & Petty, R. E. (1985). Central and
peripheral routes to persuasion: The role of message
repetition. In L. Alwitt & A. Mitchell (Eds.),
Psychological processes and advertising effects:
Theory, research, and applications (pp. 91-111).
Hillsdale, NJ: Erlbaum.
Cacioppo, J. T., Petty, R. E., & Kao, C. (1984). The
efficient assessment of need for cognition. Journal of
Personality Assessment, 48, 306-307.
Cacioppo, J. T., Petty, R. E., & Morris K. (1983). Effects
of need for cognition of message evaluation, recall,
and persuasion. Journal of Personality and Social
Psychology, 45, 805-818.
Cacioppo, J. T., Petty, R. E., & Sidera, J. A. (1982). The
effects of a salient self-schema on the evaluation of
proattitudinal editorials: Top-down versus bottom-up
message processing. Journal of Experimental Social
Psychology, 18, 324-338.
Cacioppo, F. T., Petty, R. E., & Stoltenberg, C. D. (1985).
Processes of social influence: The elaboration
likelihood model of persuasion. in P. Kendall (Ed.),
Advances in Cognitive-Behavioral Research and Therapy,
(Vol. pp. 215-274) New York: Academic Press .
Carlson, E. R. (1956). Attitude change through
modification of attitude structure. Journal of
Abnormal and Social Psychology, 52, 256-261.


28
Objective Processing
In order to provide support for objective cognitive
processing ELM researchers have studied the effects of a
number of situational and motivational variables upon the
extent of argument processing. These have included 1)
distraction, 2) repetition, 3) personal responsibility, 4)
need for cognition, and 5) personal involvement.
Illustrative studies of the first four variables will be
presented next, but research on personal involvement will
be presented in a later section.
From the ELM viewpoint distraction is seen as affecting
cognitive processing in a relatively objective way.
Specifically, the ELM predicts a "thought disruption"
hypothesis concerning the effects of distraction upon
persuasion. The reasoning underlying this hypothesis
consists of the following arguments. Distraction from
information processing should enhance persuasion for a
message consisting of weak arguments. This should occur
because without distraction, weak arguments should
generally elicit predominantly unfavorable thoughts.
However, distraction should disrupt these unfavorable
thoughts and thus enhance agreement. For a message
consisting of strong arguments favorable thoughts would
generally be predominant, and distraction would disrupt
these thoughts leading to reduced agreement.


41
rejection were broader than their latitudes of acceptance.
Furthermore, for the majority of the moderate subjects, the
width of their latitudes of acceptance exceeded the width
of their latitudes of rejection. Thus, it appears that the
two groups of subjects use very different reference scales
as internal anchors.
Results for the dependent measures reflect the
functioning of these different internal anchors. In
judging the speaker's position, extremely dry subjects saw
the message as advocating a much wetter position than it
did, while extremely wet subjects judged it as advocating a
drier position than it did. Moderate subjects were more
accurate in judging it as advocating a slightly wet
position. Moreover, the closer the communication was to
the respondent's own position, the more likely that subject
was to favorably evaluate the message in terms of fairness
and impartiality. Attitude change results indicated that
in comparison to moderate subjects, approximately twice as
many extreme subjects remained unchanged by the message.
Greater involvement thus leads to more resistance to
persuasion.
Other early research has also found that increasing
involvement was associated with resistance to persuasion.
Miller (1965) found that high issue involvement
consistently decreased the persuasive effect of a
discrepant communication on attitude measures, but


123
(Silence lasting thirty seconds)
Now that you have the situation in mind we would like
to know how clear it is for you. People often differ in
how sharp their mental pictures are, and we would like for
you to rate how clear your mental image is on the scale
below. Do this now.
(Silence lasting twenty seconds)
Now that you have indicated how clear the image is for
you, we would like to have you conjure up the image again.
This time concentrate on the things that you are saying to
yourself or thinking in the situation. In other words, as
you imagine yourself in this situation, what kind of
thoughts run through your head? During the next three
minutes we would like for you to write down every thought
that runs through your mind in this situation. As you did
before ignore spelling, grammar, and punctuation. Just
write down one thought per line on the following page.
Please turn the page now and begin.
(Silence lasting three minutes)
You have now finished the study. Thank you. Please
turn your materials in to the researcher to receive your
credit, and thank you again.


EFFECTS OF ISSUE INVOLVEMENT AND
EGO-INVOLVEMENT UPON ACCEPTANCE OF A
COUNSELING MESSAGE
By
STEVEN HUGH WILLIAMS
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1987

ACKNOWLEDGEMENTS
I would like to take this opportunity to acknowledge
and thank those individuals who have guided and supported
me throughout the completion of my dissertation. My first
thanks is to the Lord Jesus Christ for the life He has
given me during the past six years.
I am grateful to Greg Neimeyer for guiding and
encouraging me throughout my graduate studies and
especially during the completion of my dissertation. Most
of all, I am thankful for his time and his personal
interest in me and my professional development. I would
also like to thank the members of my committee, Franz
Epting, Shae Kosch, Constance Shehan, and Mark Alicke, for
their effort and insightful suggestions. It has been my
pleasure to work with these outstanding individuals.
I would also like to thank all the people who were an
integral part of this dissertation. Dr. Constance Shehan
allowed me to utilize her classroom to solicit subjects. A
special thanks goes to the undergraduate research
assistants who were involved in the technical aspects of
this study. Jean Callahan, Lisa Oglander, and John Guy
were invaluable in their assistance. April Metzler also
gave much time to this project.

I would also like to give a very special
acknowledgment to my wife, Barbara, for her constant love
and support and the many sacrifices she has made in her
life so that I may pursue my career. Barbara and my son,
Adam, have been great sources of support and encouragement
throughout this project.
in

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ii
ABSTRACT vi
CHAPTERS
IINTRODUCTION 1
IIREVIEW OF THE LITERATURE 9
The Cognitive Response Approach to Persuasion. 9
The Elaboration Likelihood Model 19
Cognitive Processing 25
Objective Processing 28
Biased Processing 32
Involvement and Persuasion 38
Involvement and Cognitive Processing 44
Involvement Level in Counseling 49
Hypotheses 51
IIIMETHODS 54
Research Design 55
Cognitive Responses to Arguments 57
Description of the Sample 59
Instrumentation 61
Relationships Grid 61
Personal Problems Inventory 66
Assertion Survey 69
Thought Listing Techniques 70
Cognitive Assessment Questionnaire 73
Procedures 73
Analyses of Data 75
Summary 75
IV

IV RESULTS 78
Manipulation Checks 78
Argument Quality 78
Issue Involvement 79
Dependent Measures 80
Cognitive Favorability (CFCR) 81
Attitude Rating 81
Behavioral Intention Rating 86
Spontaneous Use of Cognitive Restructuring.. 86
Additional Analyses 87
V DISCUSSION 89
Need for Cognition 96
Future Considerations 96
APPENDICES
A RELATIONSHIPS GRID 101
B PERSONAL PROBLEM INVENTORY 104
C ASSERTION SURVEY 107
D THOUGHT LISTING TECHNIQUE 109
E COVERT REHEARSAL TECHNIQUE 112
F COGNITIVE ASSESSMENT QUESTIONNAIRE 114
G TRANSCRIPT OF TAPED INSTRUCTIONS 116
REFERENCES 124
BIOGRAPHICAL SKETCH 135
v

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
EFFECTS OF ISSUE INVOLVEMENT AND
EGO-INVOLVEMENT UPON ACCEPTANCE OF A
COUNSELING MESSAGE
BY
STEVEN HUGH WILLIAMS
May, 1987
Chairman: Dr. Greg J. Neimeyer
Major Department: Psychology
Within the general framework of the Elaboration
Likelihood Model of persuasion, the present study
investigated the influence of issue involvement, ego-
involvement, and argument quality upon the cognitive
processing of a counseling message advocating the use of
cognitive restructuring in the treatment of assertiveness
problems. Issue involvement was measured by the total
number of relevant thoughts governed by subjects of the
Thought Listing Technique. It was predicted that there
would be greater distinction of argument quality under
conditions of high issue involvement than under conditions
of low issue involvement.
Ego-invo1vement was measured by the proportion of
total variance accounted for by a single construct
vi

concerning assertiveness on a modified version of the Role
Repertory Grid. It was predicted that there would be
greater distinction of argument quality under high ego-
involvement than under low ego-involvement. In addition,
it was predicted that under conditions of high issue
involvement, high ego-involvement would exert a greater
influence on the weak than strong arguments. Under low
issue involvement, high ego-involvement would exert an
equal influence on both strong and weak arguments.
Results of the 2x2x2 Analyses of Variance on
outcome expectation attitude ratings, behavioral intention
scores, a general cognitive favorability index and covert
rehearsal scores did not support the predictions. The
analyses revealed significantly higher attitude ratings for
the high issue involved group than the low issue involved
group. They also revealed significantly greater cognitive
favorability for strong than weak argument quality, and a
similar trend for attitude rating and argument quality. No
significant results were found for behavioral intention and
covert rehearsal measures.
These results suggest that message variables strongly
determine acceptance of counseling messages. The results
of this study were also discussed in relation to the
contributions of the Elaboration Likelihood Model to
counseling research, and future directions were suggested.
vi 1

CHAPTER I
INTRODUCTION
Persuasion and resistance to change have been issues of
theoretical speculation and empirical investigation
throughout the past century (Freud, 1894/1959; Frank, 1961;
Ellis, 1985). The extent to which a client yields and/or
resists a therapist's persuasive messages can be influenced
by a wide range of variables. Since Strong's (1968)
initial theoretical conceptualization of counseling as an
interpersonal influence process, considerable research has
been generated concerning interpersonal influence variables
in counseling and counselor power in particular (see
Corrigan, Dell, Lewis, & Schmidt, 1980, and Heppner &
Dixon, 1981, for reviews). Strong proposed a two-stage
model of counseling. In the first stage, counselors
attempt to increase their perceived power along the
dimensions of expertness, attractiveness, and
trustworthiness. In the second stage counselors use their
influence to bring about client change. Virtually all of
the research conducted so far on counseling as an
interpersonal influence process has investigated the three
1

2
above counselor characteristics. Little attention has been
paid to the influence of counseling message variables or
client characteristics, particularly to those client
characteristics which may be associated with resistance to
change.
Research within the area of social psychology on
attitude change has followed a similar route of
development. Early work on attitude change explored
communicator variables (such as prestige, expertise, and
credibility), message content, and some focus on audience
features. However, with the rapid development of a
cognitive social psychology in the 1970s, attitude change
research expanded to include the consideration of
consistency, dissonance, and attributional processes in the
communication recipient (Jones, 1985). With this expansion
message variables and recipient characteristics were
established as major variables in attitude change research
(Heppner and Heesacker, 1982). Future research on the
interpersonal influence process in counseling needs to more
fully integrate the findings from this research in
cognitive social psychology.
One social psychological theory that has relevance for
counseling research is the Elaboration Likelihood Model
(ELM: Petty and Cacioppo, 1981; Petty and Cacioppo, 1986).
Petty and Cacioppo refer to elaboration as the extent to
which a person thinks about the issue-relevant arguments

3
contained in a persuasion message. Elaboration likelihood
would be high if a person is motivated and able to evaluate
a message, and it would be low if a person is not motivated
and/or not able to evaluate a message. People, of course,
vary in their willingness and ability to evaluate messages
due to individual and situational factors. The ELM began
with Petty and Cacioppo's attempts to account for the
differential persistence of communication-induced attitude
change. .After reviewing the literature on attitude
persistence, they hypothesized that two distinct routes to
persuasion existed (Petty and Cacioppo, 1981). The first
route, which they called the central route, involves an
individual's thoughtful consideration and personal
evaluation of the information presented in the
communication. The second route, or peripheral route,
involves an individual's selection from the persuasion
context of some cue (e.g., counselor trustworthiness) that
produces change without the individual having to consider
or evaluate the information presented. The central and
peripheral routes to persuasion correspond generally to
high and low elaboration likelihood, respectively. Petty
and Cacioppo found the central route to persuasion to be
the more enduring of the two. They further postulate that
"variables can affect the amount and direction of attitude
change by : (A) serving as persuasive arguments, (B)
serving as peripheral cues, and/or (C) affecting the extent

4
or direction of issue and argument elaboration" (Petty and
Cacioppo, 1981, p. 132).
From the ELM viewpoint much of the interpersonal
influence research in counseling has focused on peripheral
cues such as counselor characteristics. One aim of this
study is to expand the focus of interpersonal influence
research in counseling to include the study of message and
recipient/client variables. In order to provide a context
for the study which would approximate to a counseling
analogue and provide a framework for the testing of these
variables, a life problem that would likely be presented in
counseling needed to be identified. The problem chosen
concerns social assertiveness. In a needs assessment
survey conducted on the Duke University Campus, Talley,
Barrow, Fulkerson, and Moore (1983) found that of the 52
needs they asked college students to rate the current
importance of the need to assertively stand up for myself
was ranked number seven in overall importance. Thus,
problems in personal assertion is very likely to be an
issue that students will present to college counselors.
In counseling the therapist often presents the client
with some task to perform in order to bring about
therapeutic change. In the treatment of assertiveness
problems, cognitive restructuring has been found to be an
effective treatment technique (Kaplan, 1982). If the
client acquires a favorable attitude toward the therapeutic

5
task (in this case cognitive restructuring), he/she would
more likely actually perform the therapeutic task outside
the counseling room. Two variables which could affect the
client's attitude toward cognitive restructuring are the
quality of the arguments for its use presented by the
counselor and the degree of involvement the client
possesses toward changing or retaining his/her present
level of assertiveness.
In the process of getting a client to try a particular
technique, such as cognitive restructuring, the counselor
will present any number of arguments for its use. The
quality of these arguments will influence the favorability
of the client's attitude toward the therapeutic task. In
order to provide a means to empirically test the ELM Petty
and Cacioppo (1979b) devised a method for developing
"strong" and "weak" messages. This method will be used in
this study to provide the means to investigate the
influence of two different types of subjects involvement
upon their cognitive responses toward the therapeutic task
of cognitive restructuring.
Because people would have a high level of involvement
in a particular life problem before seeking counseling, it
would seem that their elaboration likelihood would
generally be high. According to the ELM they would thus be
more likely to engage in central route processing rather
than peripheral route processing. Petty, Cacioppo, and

6
Goldman (1981) provided evidence that under conditions of
high involvement a thoughtful evaluation of message content
is the most important determinant of attitude change. But
is the "involvement" of the laboratory comparable to that
brought to counseling or is it more complex? This study
will attempt to explore the complexity of personal
involvement and its influence upon the cognitive processing
of a counseling technique.
In the research on susceptibility to influence,
personal involvement or relevance has been defined as the
extent to which the issue under consideration is of
personal importance (Petty and Cacioppo, 1979). Variables
similar to this definition of personal involvement have
been alternately referred to as "issue involvement"
(Kiesler, Collins, and Miller, 1969), "personal
involvement" (Sherif, Kelly, Rodgers, Sarup, and Tittler,
1973), and "ego-involvement" (Rhine and Severance, 1970;
Greenwald, 1980). As mentioned earlier, any person seeking
counseling is likely to be highly involved with the problem
or issue presented.
Another factor to consider is that people who are
personally involved with an issue have done some amount of
prior thinking on the topic before seeking counseling. It
is possible that this information has been organized into a
schema or guiding principle for the perception of new
information (Fiske and Taylor, 1984). Schema-driven

7
processing tends to be biased toward the maintenance of the
guiding schema and may override the objective processing of
externally provided communications (Ross, Leper, and
Hubbard, 1975; Markus and Sentis, 1982). George A. Kelly's
(1955) psychology of personal constructs proposes, as the
fundamental postulate of the theory, that "a person's
processes are psychologically channelized by the ways in
which he anticipates events" (Kelly, 1955, p.46). A person
uses his/her personal construct system to interpret and
"anticipate" the world. Kelly's Organization Collorary
suggests that some constructs are more important and have
more implications than others. More important or
superordinate constructs may serve as schemas for the
anticipation of incoming information. They also are more
resistant to change (Hinkle, 1966/1965). Thus, if a person
has developed a schema or superordinate construct for the
dimension of assertiveness, that person's processing of
information concerning counseling techniques for treating
assertiveness problems may be biased.
Therefore, in this study two types of involvement will
be distinguished. The first is issue involvement. This
would correspond to the involvement seen in individuals
seeking counseling. It concerns the presence or absence of
perceived problems in assertiveness. High issue
involvement should lead to greater argument elaboration.
The second, which will be called ego-involvement, will

8
concern the presence or absence of a well-developed
construct for assertiveness. The literature suggests that
such a well-developed construct may bias individuals with
assertiveness problems against change. Such a process may
be one source of resistance in counseling.

CHAPTER II
REVIEW OF THE LITERATURE
There are several areas of literature and empirical
research relevant to the current study. First, evidence
for a cognitive response approach to attitude change and
persuasion will be presented. Second, the Elaboration
Likelihood Model, a general cognitive model of persuasion,
will be presented. Next, the role of cognitive processing
will be further explored. This will include both objective
and biased modes of processing. At this point the variable
of involvement will be introduced, and its contribution to
the attitude change literature will be reviewed. The
relationships between involvement and cognitive processing
will then be explored, and its applications in a counseling
context will be considered. Finally, specific hypotheses
will be derived from this discussion.
The Cognitive Response Approach To Persuasion
Whenever people receive a persuasive message or any
other communication, they will attempt to place the new
information within the context of their existing knowledge
of the topic. This is the basic postulate of the
cognitive response approach presented by Greenwald (1968).
In thinking about the message, the individual may consider
9

10
material not in the message itself. From their prior
knowledge and evaluation of the communication the person
may generate cognitions that agree, disagree, or are
irrelevant to the persuasive message under consideration.
To the extent that the message elicits supportive cognitive
responses, attitude change in the advocated direction
should be facilitated. To the extent that the message
elicits negative cognitive responses, attitude change
should be inhibited.
The concept that an individual's cognitive responses
are an important mediator of attitudes is not a recent
development in psychology. Freud's (1900/1939) method of
free association was an early attempt at the measurement of
cognitive responses and their role as mediators of
attitudes in a clinical context. In the early study of
attitude change Hovland (1951) suggested that the accurate
and complete recording of an audience's thoughts as they
listened to a communication would constitute the best
method for investigating the internal process of change.
While the cognitive response approach to persuasion was
not formally proposed until 1968, researchers have been
concerned with cognitive responses in persuasion since the
early research of attitude change. In a classic study
investigating active versus passive participation in the
persuasion process Lewin (1947) compared individual
instruction with group discussion. The goal of both

11
treatments was to increase the consumption of unusual meats
(hearts, kidneys, etc.) during World War II. Groups of
housewives either listened to a persuasive lecture or
participated in a group discussion on the material
presented in the lecture. Survey results showed that only
3 percent of the women in the lecture group served one of
the meats, while 32 percent of the women in the discussion
group served them. While the results of this study are
open to different interpretations, it may be that the self
generated arguments of the women in the discussion group
were more persuasive than the arguments in the lecture.
In later research on the role of active and passive
participation in persuasion, Janis and King (1954) reported
that subjects changed their attitudes more in the advocated
direction when instructed to give an informal talk on a
counterattitudinal topic than did other subjects who
passively listened to the talks. In their experiment Janis
and King used three different topics. Their results
revealed significantly greater attitude change for two of
these topics in comparison to the third. From their data
and subject interviews they observed that for the two
topics with greater change subjects improvised more in
their talks. For the third topic subjects stayed close to
the prepared outline made available for each topic. In
this case improvisation, or self-generated cognition, may
have operated as a critical factor in the greater attitude

12
change. In another experiment King and Janis (1956) had
subjects either read a persuasive communication to
themselves, read it into a tape recorder, or read and then
give their own improvised version of the message. Those
subjects who improvised changed their attitudes
significantly more than subjects in either the oral or
silent reading conditions. These results are supportive of
the notion that one's own cognitive responses on an issue
are the most compelling. In a further test of this
hypothesis Greenwald and Albert (1968) had subjects
improvise five arguments in response to instructions to
advocate either career preparatory or general liberal arts
undergraduate education. Subjects also read a set of
provided arguments supporting the opposite side. Results
indicated that subjects' attitudes tended to be in the
direction of the content of their own cognition. Moreover,
subjects recalled significantly more of their own
improvised arguments than provided arguments. They also
evaluated their own arguments as more original than those
provided to them.
Another area of research suggesting that a person's own
responses are important in mediating persuasion is
inoculation theory. McGuire (1964) suggests that
resistance to persuasion can be created by providing
information and arguments supportive of an individual's
original attitude or by "inoculating" the individual by

13
providing weak counterarguments to his or her attitude
accompanied by refutations. McGuire states that the
inoculation will pose a cognitive threat that will motivate
the individual to generate supportive arguments and
counterargument refutations for his or her original
attitude. This practice should produce greater resistance
to subsequent persuasion attempts. The point of interest
for the cognitive response approach is, does inoculation
lead to self-generated cognitive responses and thus greater
resistance to persuasion. McGuire and Papagioris (1961)
asked subjects to list thoughts in favor of their initial
position one week after being exposed to either supportive
defenses or refutational defenses (the inoculation
treatment). Subjects who had been exposed to the
refutational defenses listed more supportive thoughts than
subjects who had been exposed to supportive defenses. In
an experimental investigation of the effectiveness of
active and passive participation in the defense of one's
beliefs McGuire (1964) provided additional evidence for the
importance of one's own cognitive responses in mediating
attitudinal changes. In the active participation condition
he assigned subjects the task of writing either a
supportive or refutational defense. In the passive
participation condition subjects simply read a defense
provided by the investigator. The beliefs used in this
study were cultural truisms (e.g.,
mental illness is not

14
contagious) which people generally have little practice in
defending. According to inoculation theory individuals in
the active condition would perform more poorly than those
in the passive condition when exposed to immediate
counterattitudinal attack. However, they should
demonstrate increased resistance to attacks which are
delayed. His results conformed to this predicted
interaction. Active defenses increased in resistance when
the attack occurred one week later, while passive defenses
declined in resistance except for the passive refutational-
different defense. This passive defense, like the active
ones, presumably posed a threat that motivated the subjects
to generate additional cognitive responses in support of
their beliefs.
Further support for the principle that a person's own
cognitive responses are important in producing persuasion
is offered by studies investigating the relation of
attitudes to underlying beliefs and values. In the 1950s
and 1960s widespread racial prejudice caused many
psychologists to be concerned with changing prevailing
attitudes. Two studies designed to assess the impact upon
attitudes of changing underlying beliefs that illustrate
the importance of cognition were conducted by Carlson
(1956) and Stotland, Katz, and Patchen (1959). Carlson
(1956) found that in moderately prejudiced college students
more favorable attitudes toward racial integration could be

15
produced by making the students aware of how desegregation
would help in the attainment of some important goals (e.g.,
greater American prestige in other countries). Stotland et
al. (1959) had college students read a message designed to
give insight into the psychodynamics of racial prejudice.
They also assigned subjects to various manipulations
intended to facilitate the internal restructuring of
beliefs (e.g., ordering statements into cause-and-effect
sequences). Attitude measures showed no immediate
reduction in prejudice, but a significant reduction was
found in follow-up measures three to four weeks later.
These findings suggest that the restructuring of internal
beliefs with subsequent attitude change takes some time to
occur.
Thus far early research in support of a cognitive
response approach to persuasion has been presented from
three areas: 1) active versus passive participation in
persuasion, 2) inoculation theory, and 3) attitude
underlying beliefs linkages. Cognitive response
researchers have also provided empirical evidence for the
importance of cognition in producing persuasion. They
present four types of evidence for the cognitive response
approach based on forewarning manipulations, issue-relevant
thinking manipulations, argument quality manipulations, and
psychophysiological measures (Cacioppo, Petty, &
Stoltenberg 1985).

16
In an experiment utilizing the first approach, Petty
and Cacioppo (1977) forewarned some subjects and did not
forewarn others that they would be hearing a tape recording
prepared by the "Faculty Committee on Academic Affairs"
which recommended that seniors be required to pass a
comprehensive exam in their major as a prerequisite for
graduation. Half of the subjects completed a "thought
listing" technique (Cacioppo & Petty, 1981) in the latter
2.5 minutes of the 5 minute interval between the
forewarning and the message. Attitude survey results
indicated that those subjects who were forewarned, whether
or not they listed their thoughts, showed greater
resistance to persuasion than those who had not been
forewarned. Analyses of the thought-listings showed that
the forewarning elicited anticipatory counterargumentation.
Forewarned subjects listed twice as many thoughts that were
unfavorable than were favorable toward the recommendation.
Subjects who had not been forewarned did not generate any
issue-relevant thoughts, as they were unaware of the
upcoming message's topic.
In the second experiment reported in Petty and Cacioppo
(1977), they manipulated issue-relevant thinking. In this
study students in an introductory psychology class were
told that a guest lecturer, a psychologist from the
University Counseling Center, would speak to their class
that day, and in return for the visit they would complete a

17
questionnaire he had brought. Half of the students were
forewarned that the psychologist would be advocating that
all freshmen and sophomores be required to live in campus
dorms, a topic which pretesting showed most students to be
against. The remaining students were not forewarned of the
topic. Half of all of the students were then asked to list
their thoughts just prior to the speech. So far this study
replicates the conditions of the forewarning study
described above. The other half of the students, however,
were asked to list all of their thoughts on the topic of
requiring students to live in dorms. Thus, half of the
subjects were instructed to engage in issue-relevant
thinking whether or not they had been forewarned. Attitude
measures revealed that unwarned subjects who engaged in
issue-relevant thinking demonstrated resistance to
persuasion equal to that exhibited by the forewarned
groups. It appears that just being instructed to think
about a topic can produce cognitive responses which support
one's beliefs and, subsequently, lead to greater resistance
to persuasion.
In the third line of evidence argument quality has been
operationally defined such that "strong" arguments elicit
more favorable than unfavorable thoughts about a message,
whereas "weak" arguments elicit more unfavorable than
favorable thoughts concerning the message. In an
illustrative study, Cacioppo and Petty (1985) exposed

18
subjects either one or three times to a set of strong or
weak arguments in support of the recommendation that senior
comprehensive exams be instituted. According to their
theory, argument quality should differentiate people's
attitudes more after three than after one message
presentation. This should occur if it is the nature of
people's issue-relevant thoughts rather than the number of
different arguments they learn. Results indicate that
argument quality did differentiate people's attitudes more
after three exposures than after one. Subjects also
recalled more arguments after three presentations than
after one, but the amount of the message learned was
unrelated to their attitudes.
The fourth source of evidence comes from the use of
psychophysiological measures. Cacioppo and Petty (1979)
used integrated electromyographic (IEMG) activity as a
measure of subjects' silent language processing. In one
experiment, IEMG activity was recorded as subjects sat
quietly and as they listened to a tape-recorded message.
One minute prior to the message, subjects heard one of
three announcements about the message. One group only
heard that a message would be presented in one minute; a
second group was forewarned that they would hear a message
recommending more lenient visitation hours be instituted in
dormitories (a proattitudinal message according to
pretesting); and a third group was forewarned that they

19
would hear a message recommending that stricter visitation
hours be instituted (a counterattitudinal message). They
found that IEMG activity increased in all conditions during
the presentation of the message, which would be expected if
IEMG activity indicated the presence of silent language
processing. The finding of interest was that IEMG activity
had increased following the forewarning of the upcoming
counterattitudinal message.
In summary, early research prior to the formal
development of the cognitive response approach to pesuasion
supported the importance of people's construing of the
information in persuasive messages, rather than their
hearing of the material per se, as the most important
factor in determining their susceptibility or resistance to
persuasion. The methodologies described above have all
been used in research on a general framework for
understanding persuasion called the Elaboration Likelihood
Model (Petty & Cacioppo, 1986), which will be described
next.
The Elaboration Likelihood Model
The Elaboration Likelihood Model (ELM) of attitude
change is a general response theory, which arose out of
Petty and Cacioppo's (1981) attempts to account for the
differential persistance of communication-induced attitude
change. They hypothesized that two distinct routes to
persuasion existed, which they called the central and

20
peripheral routes. The central route involves a person's
thoughtful consideration and evaluation of the information
presented in the communication; whereas, the peripheral
route involves a person's selection from the persuasion
context of some cue (e.g., source trustworthiness) that
produces attitude change without the person having to
thoughtfully consider or evaluate the information
presented. Petty and Cacioppo found the central route to
be the more enduring of the two routes.
While people are neither universally thoughtful nor
universally mindless in evaluating persuasive messages,
they do vary in their motivation and ability to consider
carefully the information and arguments that comprise a
persuasive attempt. Petty and Cacioppo (1981) refer to
elaboration likelihood as the degree to which a person
thinks about the issue-relevant arguments contained in a
persuasion message. If conditions produce or enhance an
individual's motivation and ability to engage in issue
relevant thinking, then the elaboration likelihood is
considered to be high. Under conditions of high
elaboration likelihood a person allocates significant
cognitive resources to considering and critically
evaluating the presented message. If an individual's
motivation or ability to process a message is reduced, then
the elaboration likelihood is also decreased. When it is
low, a person is likely to forego a careful consideration

21
of issue-relevant information and rely upon positive or
negative cues in the persuasion context. A simple but
reasonable decision rule could also be used in the
persuasion context when elaboration likelihood is low.
In order to provide empirical support for distinct
central and peripheral routes to persuasion and the notion
of elaboration likelihood, researchers have used the
argument quality manipulation under high and low relevance
conditions. In high relevance conditions subjects were led
to believe that the issue had direct personal consequences
for them, while in low relevance condition subjects were
led to believe that the issue had few if any personal
consequences. The issues and messages used were easy to
understand so all subjects had the ability to think about
the information presented. Such a design suggests that
subjects in the high relevance conditions should follow the
central route to persuasion and subjects in the low
relevance conditions should follow the peripheral route.
Petty, Cacioppo, and Goldman (1981) conducted such a
study in which college students were exposed to a
counterattitudinal appeal (favoring senior comprehensive
exams). In addition to personal relevance and argument
quality, source expertise was also manipulated in this
study. Under conditions of high relevance, students were
led to believe that the exam policy would begin next year
and thus affect them. Under conditions of low relevance,

22
they were led to believe that the exam policy would begin
in 10 years and thus not affect them. For half of the
subjects the message was attributed to a Princeton
University Professor (high credible source), while for the
other half it was attributed to a high school student (low
credible source). Source credibility served as a
persuasion context cue in this study. Attitude rating
results revealed two significant interactions. First, a
relevance X message quality interaction showed that
argument quality was a more important determinant of
persuasion for high than low relevance subjects. Second, a
relevance X source credibility interaction suggested that
the source cue was a more important determinant for
persuasion for low than high relevance subjects. Thus,
under conditions of high relevance (high elaboration
likelihood), subjects exerted the effort to evaluate the
issue-relevant arguments presented. Under condition of low
elaboration likelihood, they were persuaded by a context
cue and appeared to be unaffected by argument quality.
While in the above study message factors were prepotent
under condition of central route processing and source
factors were prepotent under conditions of peripheral route
processing, the central/peripheral distinction is not one
between message and source factors. It refers to whether
issue-relevant thinking versus context cues or decision
rules leads to attitude change. Different message factors

23
could have differing effects depending upon the level of
elaboration likelihood. For example, the quality of the
message arguments should have a greater influence when
elaboration likelihood is high, while the actual number of
arguments could have a greater influence when elaboration
likelihood is low. In the latter situation, the person may
employ a decision rule such as, "the more arguments the
better."
To test these hypotheses Petty and Cacioppo (1984)
conducted two studies. In the first experiment, college
students received a message on the issue of instituting
senior comprehensive exams. Personal relevance was
manipulated as outlined in the Petty, Cacioppo, and Goldman
(1981) study described previously. Subjects received one
of four messages in support of the exam proposal: 1) three
strong arguments, 2) three weak arguments, 3) nine strong
arguments, and 4) nine weak arguments. Attitude rating
results revealed that number of arguments was a more
important determinant of persuasion under low than high
relevance but that quality of the arguments was more
important under high than low relevance condition. In the
second experiment, the message concerned a proposal to
increase tuition, but relevance was manipulated by either
stating that the proposal was for the student's own
university (high relevance) or for a distant university
(low relevance). In this study the message contained

24
either three strong argument, three weak argument, or six
arguments (three strong and three weak). Results indicate
that under high relevance conditions three strong arguments
elicited more attitude agreement with the proposal than
three weak arguments. The agreement for six arguments was
greater than that for the three weak arguments but less
than that for the three strong arguments. Under low
relevance conditions, six arguments produced the most
agreement followed by three strong and finally three weak.
Thus, in these two studies argument quantity served as a
cue under low relevance conditions, but quality of the
arguments was more important under high relevance
conditions.
This research suggests that two distinct routes to
persuasion do exist and differ in their level of
elaboration likelihood. The ELM provides a organizing
framework for understanding these major cognitive processes
underlying persuasion and how variables relate to these
processes. Petty and Cacioppo (1986) present the ELM in
postulate form as follows:
1. People are motivated to hold correct attitudes
(p. 127) .
2. Although people want to hold correct attitudes,
the amount and nature of issue-relevant elaboration
in which people are willing or able to engage to
evaluate a message vary with individual and
situational factors (p. 128).
3. Variables can affect the amount and direction
of attitude change by: (A) serving as persuasive
arguments, (B) serving as peripheral cues, and/or
(C) affecting the extent or direction of issue and
argument elaboration (p. 132) .

25
4. Affecting motivation and/or ability to process
a message in a relatively objective manner can do
so by either enhancing or reducing argument
scrutiny (p. 138) .
5. As motivation and/or ability to process
arguments is decreased, peripheral cues become
relatively more important determinants of
persuasion. Conversely, as argument scrutiny is
increased, peripheral cues become relatively less
important determinants of persuasion (p. 152).
6. Variables affecting message processing in a
relatively biased manner can produce either a
positive (favorable) or negative (unfavorable)
motivational and/or ability bias to the issue
relevant thoughts attempted (p. 163).
7. Attitude changes that result mostly from
processing issue-relevant arguments (central route)
will show greater temporal persistence, greater
prediction of behavior, and greater resistance to
counterpersuasion than attitude changes that result
mostly from peripheral cues (p. 175).
While reviewing the accumulated empirical evidence for
the ELM is beyond the scope of this chapter, relevant
research will be presented in the discussions of cognitive
processing and the variable of involvement.
Cognitive Processing
In the ELM Petty and Cacioppo (1981) propose that two
major cognitive routes, the central and the peripheral, are
used by people to process persuasive messages. The
understanding of these cognitive processes involves the
analysis of the structures and mechanisms that comprise
mental activity as well as the impact of various source,
message, and recipient variables upon mental activity.
Petty and Cacioppo (1986) further propose that variables
can affect the cognitive processing of message arguments so
that argument processing proceeds in a relatively objective

26
or a relatively biased manner. In objective processing,
some variable "either motivates or enables subjects to see
the strengths of cogent arguments and the flaws in specious
ones, or inhibits them from doing do" (p. 136). In biased
processing some variable "either motivates or enables
subjects to generate a particular kind of thought in
response to a message, or inhibits a particular kind of
thought" (p. 136). Objective processing appears to have
much in common with "bottom-up" or "data-driven"
processing, since the elaboration is primarily impartial
and data focused. On the other hand, biased processing
appears to share similarities with "top-down" or "theory-
driven" processing, as the elaboration may be guided by
prior knowledge, such as a relevant attitude schema
(Landman & Manis, 1983; Fiske & Taylor, 1984).
Argument processing lies at the heart of the ELM
(postulate 3) along with two other critical constructs:
argument quality and peripheral cues. While peripheral
cues are not investigated in this study, argument quality
is used as a manipulation to assess the degree to which
variables of involvement affect cognitive processing in
either a relatively objective or biased manner. In
accordance with the ELM, a persuasive message with strong
arguments should elicit more agreement when it is
considered carefully (high elaboration) than when
consideration is low (low elaboration), but a persuasive

27
message consisting of weak arguments should produce less
agreement when consideration is high rather than low. By
manipulating argument quality along with another variable,
it is possible to assess whether that variable enhances or
reduces argument processing in a relatively objective or
biased manner. In relatively objective processing if the
variable increases processing, subjects' attitudes and
thoughts should be more clearly distinguished when the
variable is present rather than absent. In the case of
biased processing a variable will produce varying levels of
effects depending upon the direction of the bias and
argument strength. A variable that biases thinking in a
positive direction should generally have a greater
influence on a strong than a weak message, because it will
be more difficult for an individual to generate favorable
thoughts to weak than strong arguments. On the other hand,
a variable that produces a negative bias should have a
greater effect on a weak than a strong message. This
should occur because it will generally be more difficult
for an individual to generate counterarguments to strong
than weak arguments. Research utilizing the argument
quality manipulation has provided empirical evidence for
the effect of numerous variables upon the extent of
cognitive processing in both relatively objective and
biased manners. A sampling of this research will be
presented next.

28
Objective Processing
In order to provide support for objective cognitive
processing ELM researchers have studied the effects of a
number of situational and motivational variables upon the
extent of argument processing. These have included 1)
distraction, 2) repetition, 3) personal responsibility, 4)
need for cognition, and 5) personal involvement.
Illustrative studies of the first four variables will be
presented next, but research on personal involvement will
be presented in a later section.
From the ELM viewpoint distraction is seen as affecting
cognitive processing in a relatively objective way.
Specifically, the ELM predicts a "thought disruption"
hypothesis concerning the effects of distraction upon
persuasion. The reasoning underlying this hypothesis
consists of the following arguments. Distraction from
information processing should enhance persuasion for a
message consisting of weak arguments. This should occur
because without distraction, weak arguments should
generally elicit predominantly unfavorable thoughts.
However, distraction should disrupt these unfavorable
thoughts and thus enhance agreement. For a message
consisting of strong arguments favorable thoughts would
generally be predominant, and distraction would disrupt
these thoughts leading to reduced agreement.

29
To test this hypothesis Petty, Wells, and Brock (1976)
presented college students with either strong or weak
arguments for the proposal that their tuition be raised by
20 percent accompanied by a distraction task. Their
distraction task consisted of having subjects record the
quadrant in which Xs flashed on a screen in front of them.
Distraction varied according to the rate of presentation of
the Xs with four levels: l)no distraction, 2) low
distraction (15 second intervals), 3) medium distraction (5
second intervals), and 4) high distraction (3 second
intervals). After hearing the messages subjects completed
attitude measures and listed their thoughts during the
message. Attitude rating results revealed a significant
argument quality X distraction interaction which is
consistent with the thought disruption hypothesis. Under
conditions of weak argument quality increasing distraction
was associated with more favorable attitudes; however,
under conditions of strong argument quality increasing
distraction was associated with less favorable attitudes.
Analyses of the thought listings showed that high
distraction decreased counterargument generation for the
weak message but not for the strong one. Moreover, high
distraction tended to decrease the amount of favorable
thoughts elicited by the strong message but not the weak
one. For similar results see Lammers and Becker (1980).

30
Repetition is another variable which, in moderate
amounts, can increase cognitive processing of information
in a relatively objective way. Once again, Cacioppo and
Petty (1985) used strong and weak arguments for the
institution of a senior comprehensive exam; this time,
however, half of their college student subjects heard the
message once, and half heard the message three times in
succession. Students showed greater attitudinal
differentiation of strong from weak messages when the
message was repeated three times rather than just presented
once. Similar results on the impact of repetition on
persuasion are reported by Petty and Cacioppo (1984).
Personal responsibility has also been found to enhance
objective processing of issue relevant arguments. Social
psychological research on the "social loafing" effect
(Latane & Darley, 1970; Latane, Williams, & Harkins, 1979)
suggests the classical finding of reduced personal
responsibility in group performance in comparison to
individual performance may result from a loss of motivation
in the group setting. For the persuasion context this
implies that if personal responsibility is increased, then
motivation to engage in significant cognitive processing
should also increase. In a test of this hypothesis Petty,
Harkins, and Williams (1980) asked college students to
provide peer feedback on editorial messages ostensibly
written by journalism students. Subjects were led to

31
believe that they were either one out of a group of ten
people who were evaluating the editorials or the only
person responsible for the evaluation. Editorials
contained either strong, weak, or a mixture of arguments
for the institution of comprehensive senior exams.
Attitude results revealed that individual evaluators were
significantly more favorable toward the strong message and
were significantly less favorable toward the weak message
than group evaluators. Responsibility did not affect
evaluations of the mixed message. Thus, as personal
responsibility increased, argument quality became a more
important determinant of the evaluation.
Finally, need for cognition, an individual difference
variable, refers to the extent to which people need to
structure information in reasonable, integrated ways
(Cohen, Stotland, & Wolfe, 1955). Cacioppo, Petty, and
Morris (1983) exposed high and low need for cognition
college students to either strong or weak arguments for a
proposal which would raise their tuition. After hearing
the message, subjects were asked to evaluate the message
and to give their personal opinion about the issue.
Results indicate that high need for cognition students
produced more polarized evaluations and attitudes for the
strong and weak messages. Furthermore, a significantly
higher correlation between argument evaluation and personal
opinion was found for the high (r = .70) than the low

32
(r = .22) need for cognition group. This would be expected
since high need for cognition subjects should be more
likely to deduce their attitudes from a careful
consideration of the central arguments in the proposal.
It is now apparent that numerous variables can affect
people's motivation and/or ability to process persuasive
arguments in a relatively objective manner. As noted
earlier, however, variables can also affect cognitive
processing in a relatively biased way.
Biased Processing
Studies of the impact of forewarning and prior
knowledge have provided evidence within the ELM framework
for the relatively biased cognitive processing of
persuasive messages. Forewarning has been seen to affect
an individual's motivation to process arguments in a biased
manner, whereas prior knowledge generally affects an
individual's ability to engage in biased processing.
Forewarning can be further distinguished as either warning
of message content or warning of persuasive intent.
In a study of the forewarning of message content, Petty
and Cacioppo (1977) manipulated argument quality,
forewarning, and also issue-relevant thinking. Details and
results of this study were presented earlier in the section
on ELM research in support of the cognitive response
approach. Briefly, if subjects engage in issue-relevant
thinking, they resisted the message whether they were

33
forewarned or not. These results support the notion that
it is not so much the warning but rather the accessing of
attitude-supportive cognitions and the subsequent
processing of the persuasive message in light of these
cognitions which facilitates resistance. It appears that
the forewarning served to motivate subjects to begin
thinking about their beliefs and the information and
thoughts that they already had to support those beliefs.
In line with this reasoning, a content forewarning is more
effective when there is some time delay between the warning
and message to allow thinking (Hass & Grady, 1975; Petty &
Cacioppo, 1977).
Hass and Grady (1975) also report that a forewarning of
persuasive intent is just as effective when it immediately
precedes a message as when it comes several minutes before
a message. This suggests that an intent warning functions
differently from a content warning. They propose that an
announcement of an intent to persuade may arouse a
psychological state of "reactance" that motivates a person
to defend their freedom to hold a particular attitude (cf.
Brehm, 1972). Such a state could produce biased
processing. In a study designed to explore whether
warnings of persuasive intent produced biased processing
Petty and Cacioppo (1979b) told students that they would be
evaluating radio editorials. Some students were given a
warning that the editorials were intended to persuade

34
college students, while others were just told that the
editorials were a journalism class project. Personal
relevance was also manipulated in this study by telling the
students that certain college regulations would change next
year (high relevance) or ten years from now (low
relevance). The editorial message used with all subjects
consisted of five strong arguments for the institution of
senior comprehensive exams as a requirement for graduation.
Attitude measure results revealed a main effect for warning
and a warning X relevance interaction. The interaction
showed that warning significantly decreased attitude
agreement only under the high relevance condition. Since
the warning reduced agreement even though the arguments
used were strong, this suggests that it induced biased
rather than objective processing. As this effect was
stronger for high than low relevance conditions the warning
did not appear to function as a simple rejection cue but
rather as a motivator for counterargument and resistance to
change.
The common thread that runs through the operation of
both types of forewarning effects is the presence of some
organized structure of prior knowledge within the recipient
of the persuasive attempt. In the social psychological
literature a person's organized structure of knowledge is
referred to as a schema (Landman & Manis, 1983; Markus &
Zajonc, 1985). Markus and Zajonc (1985) state that the

35
processing of persuasive messages "may be seen as
consisting of schema formation or activation, of the
integration of input with these schemas, and of the
updating or revision of these schema to accommodate new
input" (p. 150). While it is possible that schemas would
enable a person to process messages more objectively, the
literature suggests that when processing is guided by a
schema, it tends to be biased toward the perseverance of
the existing schema (Ross, Lepper, & Hubbard, 1975;
Crocker, Fiske, & Taylor, 1984; Fiske & Taylor, 1984).
Thus, if a person has a great deal of prior knowledge, or a
well-developed schema, he or she would tend to be able to
more effectively counterargue messages opposing their
initial attitudes and to cognitively argue for messages
supporting their initial attitudes (Lord, Ross, & Lepper,
1979) .
It should be noted that a schema as currently defined
within the literature is conceptually similar to George A.
Kelly's (1955) theorizing on personal constructs. Like the
schema theorists he sees people as information processors
interested in the organization of knowledge and prediction
of future events. His Fundamental Postulate states that "a
person's processes are psychologically channelized by the
ways in which he anticipate events" (p. 46). In addition,
the Organization Corollary states that, "each person
characteristically evolves, for his convenience in

36
anticipating events, a construction system embracing
ordinal relationships between constructs" (p. 56).
Constructs are an individual's basic interpretations of the
world and are bipolar in that they describe the way in
which certain things are perceived as alike and different
from others (Bannister and Mair, 1968). Constructs which
are superordinate in a person's system subsume other
constructs, have relatively more implications, and are
generally based on more information. Thus, they function
similarly to schemas.
In an investigation of the effects of prior knowledge
and schemas upon cognitive processing, Cacioppo, Petty, and
Sidera (1982) provide evidence that prior knowledge does
influence the processing of persuasive messages. Subjects
who characterized themselves using trait adjectives as
either "religious" or "legalistic" people were exposed to
strong or weak proattitudinal messages, which were either
schema-congruent or schema-incongruent. After presentation
of the message subjects rated its persuasiveness and listed
their thoughts. Results suggested that with proattitudinal
messages subjects who received schema-relevant information
were more positive about the quality of the communication's
arguments and in their listed thoughts. Thus, it appears
that self-schemas influence cognitive responses in a biased
or top-down fashion.

37
With counterattitudinal message it would be expected
that prior knowledge would increase a person's ability to
counterargue the message. To test this hypothesis Wood
(1982) divided subjects into high and low prior knowledge
groups for the topic of environmental preservation. One to
two weeks later subjects read a counterattitudinal message
consisting of four arguments against environmental
preservation. Attitude measure and thought listing
revealed the subjects who had high prior knowledge changed
less in the advocated direction than did subjects with low
prior knowledge. In addition, high prior knowledge
subjects generated more counterarguments and fewer
favorable thoughts toward the message.
Wood, Kallgren, and Priesler (1985) extended this
finding by adding argument quality and message length
manipulations. They divided subjects into high, medium,
and low prior knowledge about environmental preservation
groups and exposed them to one of four persuasive messages.
Two messages consisted of three strong arguments against
preservation, and two messages consisted of three weak
arguments against preservation. In the message length
manipulation two versions of each type of argument were
developed. One version contained short concise statements,
while the other contained more wordy versions of the same
arguments. Both versions were equal in argument strength
and ease of comprehension. Attitude measure results reveal

38
that high prior knowledge subjects were overall more
resistant to the different messages than low prior
knowledge subjects and that this effect was stronger for
weak than strong arguments. This result is consistent with
the notion that prior knowledge would enhance a person's
ability to counterargue an incongruent message. With weak
arguments it would be easier to counterargue than with
strong arguments. In addition, low prior knowledge
subjects' attitudes were affected by argument length, but
high prior knowledge subjects' attitudes were not.
In sum, research on forewarning and prior knowledge has
provided support for the view that when a person has a
well-developed level of organized knowledge on a topic,
message processing will be biased. This will occur,
because prior knowledge enables the counterarguing of
incongruent messages and the strengthening of congruent
ones. The presence of schemas or superordinate constructs
are indicators of such well-developed levels of prior
knowledge.
Involvement and Persuasion
In the preceding sections numerous variables (e.g.,
need for cognition, forewarning) which may affect cognitive
processing in either a relatively objective or relatively
biased manner have been discussed. One of the most
important variables which can affect the extent and type of
cognitive processing employed is personal involvement. In

39
past psychological research similar variables have been
alternately referred to as "issue involvement" (Kiesler,
Collins, and Miller, 1969), "personal involvement" (Sherif,
Kelly, Rodgers, Sarup, and Tittler, 1973), and "ego-
involvement" (Greenwald, 1980). This has occurred because
involvement can be judged in a variety of ways, such as the
degree to which an issue has "intrinsic importance" (Sherif
and Hovland, 1961), the centrality of the issue to a
person's self, and the number of personal consequences of
the issue.
Involvement became a central variable in research on
persuasion with the advent of social judgment theory
(Sherif and Hovland, 1961; Sherif, Sherif, and Nebergall,
1965). Social judgement theory views persuasion as a two-
stage process. First, one makes a judgment about the
position of a persuasive message in relation to one's own
position. Second, attitude change or no change occurs
depending upon the assimilation or contrast effects evoked
by the judged discrepancy between the message and one's own
position. Central to the theory are two assumptions: (1)
an individual's own stand on an issue serves as an internal
anchor for judging messages, and (2) the more "involved" an
individual is in the issue, the stronger the anchoring
effects of the initial opinion. Greater involvement thus
leads to more resistance to persuasion, because people were
postulated to hold broader "latitudes of rejection" as

40
involvement increased. For Sherif and Hovland, involvement
seemed to refer to both the intensity with which an
attitude is held and the importance of that attitude for
the self-identity.
Sherif and Hovland (1961) described two field studies
in support of social judgment theory. For illustrative
purposes the "prohibition study" (Hovland, Harvey, and
Sherif, 1957) will be presented here. The prohibition
study was conducted in Oklahoma shortly after a close
referendum which favored prohibition in the final outcome.
To insure that subjects were deeply involved with the
issue, dry-stand subjects were recruited from Women's
Christian Temperance Union groups, the Salvation Army, and
strict denominational colleges. Wet-side subjects were
selected from acquaintances of the experimenters. A group
of moderate-stand subjects were also included in the study.
Subjects were exposed to a 15 minute tape recorded message,
which presented either a wet, moderately wet, or dry stand
on prohibition. Dependent measures were obtained on the
following: (1) the communication's estimated position, (2)
subject's reactions to the communication in terms of
fairness and impartiality, and (3) subject's preferred
position and latitudes of acceptance and rejection.
In support of social judgment theory, Hovland, Harvey,
and Sherif (1957) found that for the majority of the
subjects with initially extreme stands, their latitudes of

41
rejection were broader than their latitudes of acceptance.
Furthermore, for the majority of the moderate subjects, the
width of their latitudes of acceptance exceeded the width
of their latitudes of rejection. Thus, it appears that the
two groups of subjects use very different reference scales
as internal anchors.
Results for the dependent measures reflect the
functioning of these different internal anchors. In
judging the speaker's position, extremely dry subjects saw
the message as advocating a much wetter position than it
did, while extremely wet subjects judged it as advocating a
drier position than it did. Moderate subjects were more
accurate in judging it as advocating a slightly wet
position. Moreover, the closer the communication was to
the respondent's own position, the more likely that subject
was to favorably evaluate the message in terms of fairness
and impartiality. Attitude change results indicated that
in comparison to moderate subjects, approximately twice as
many extreme subjects remained unchanged by the message.
Greater involvement thus leads to more resistance to
persuasion.
Other early research has also found that increasing
involvement was associated with resistance to persuasion.
Miller (1965) found that high issue involvement
consistently decreased the persuasive effect of a
discrepant communication on attitude measures, but

42
involvement level did not affect subjects' latitudes of
acceptance. According to social judgment theory, high
involvement should reduce latitudes of acceptance. Eagly
and Manis (1966) also report that highly involved subjects
react more negatively toward persuasive messages that
contradict their beliefs.
Other researchers, however, have not found results
consistent with the notion that higher levels of
involvement increase resistance to persuasion. Zimbardo
(1960) defined involvement in terms of an individual's
concern with the social consequences of his or her response
in a given situation rather than the intrinsic importance
of an issue for the individual. He found greater attitude
change with highly involved subjects than with less
involved subjects. Freedman (1964) also operationalized
involvement in terms of concern about a response. In
addition, he exposed subjects to messages which were
slightly, moderately, or extremely discrepant from their
initial positions. His results indicated that under low
involvement, there was more change with greater
discrepancy; but under high involvement, their relationship
was nonmonotonic with maximum change occurring at moderate
discrepancy. Moreover, under moderate discrepancy, both
high and low involved subjects changed by a similar amount.
Similar results were reported by Rhine and Severance
(1970).

43
Contradictory results have also been reported by
researchers who have manipulated forewarning along with
involvement. Apsler and Sears (1968) either forewarned or
did not forewarn subjects that they would read a proposal
calling for the replacement of professors by supervised
teaching assistants. Involvement was manipulated by
informing half of the subjects that the proposal would go
into effect ten years from now. Results indicated a
significant interaction with forewarning inhibiting
attitude change under high personal involvement but
facilitating change under low personal involvement. Dean,
Austin, and Watts (1971) found, however, that forewarning
inhibited attitude change for both high and low involvement
issues and its effect was even more pronounced for the low
involvement issue. On the other hand, Petty and Cacioppo
(1979b) found the inhibiting effect of forewarning to be
greater under high than low involvement conditions. In
these latter two studies the forewarning given was one of
persuasive intent not content, but all three studies
manipulated involvement in similar ways.
One interesting finding in the Petty and Cacioppo
(1979b) study was that when no forewarning was given, high
involved subjects tended to show more attitude change than
low involved subjects. Similar findings were also reported
by Apsler and Sears (1968) and Eagly (1967). Eagly (1967)
gave subjects favorable or unfavorable discrepant

44
information about either themselves (high involvement) or
another person (low involvement). She found that when
favorable information was provided, high involvement
subjects showed more attitude change than low involvement
subjects. When unfavorable information was provided, the
reverse was true.
In sum, research has produced contradictory findings
on the effects of level of involvement. To some degree
this has resulted from differing definitions of
involvement, but the contrary results have persisted even
when involvement was similarly defined. In hopes of
reconciling these opposing results, involvement will now be
examined in relation to cognitive processing.
Involvement and Cognitive Processing
From the ELM perspective Petty and Cacioppo (1979a)
propose that increasing involvement with an issue increases
one's motivation to cognitively process issue-relevant
information and can lead to either increased or decreased
persuasion. The research presented in the previous section
clearly demonstrates that high involvement can lead to
either enhanced or inhibited persuasion. Petty and
Cacioppo further postulate that whether persuasion is
enhanced or inhibited depends upon the individuals initial
opinion. If the persuasive message is contradictory to the
individual's initial attitudes, it is likely that the
individual is motivated and able to generate

45
counterarguments to the message. Therefore, as issue
relevant thinking increases, counterargumentation and
resistance to persuasion also increases. On the other
hand, if the persuasive message is congruent with the
individual's initial attitude, it is likely that the
individual is motivated and able to generate favorable
cognitions. As involvement and issue-relevant thinking
increases in this situation, more favorable thoughts might
be generated, and increased persuasion would result.
In order to test these hypotheses Petty and Cacioppo
(1979a) conducted two experiments. In the first experiment
subjects were exposed to either a proattitudinal or
counterattitudinal message concerning coed visitation
hours. Issue involvement was manipulated by stating that
the changes in visitation hours would go into effect at
their own university (high involvement) or at a distant
university (low involvement). Dependent measures consisted
of attitude scales and thought listings. Results on the
attitude index revealed a significant interaction between
involvement and type of message. Analysis of the
interaction revealed that increased involvement increased
agreement with the proattitudinal message but decreased
agreement with the counterattitudinal message. The thought
listing technique revealed similar results. Under high
involvement conditions subjects produced more positive
cognitions and fewer counterarguments to the proattitudinal

46
communication than to the counterattitudinal one. Message
direction showed no significant effects on cognitive
responses under low involvement conditions.
In the second experiment Petty and Cacioppo (1979a)
again varied involvement, but they presented only a
counterattitudinal message with either strong or weak
arguments. The message in this experiment concerned the
institution of senior comprehensive exams. The ELM
predicts that increased involvement will lead to more
issue-relevant thinking about the arguments presented in
the message. For the message with strong arguments, which
are difficult to counterargue, increased involvement should
be associated with more persuasion. For the message with
weak and easy-to-counterargue arguments, increased
involvement will be associated with decreased persuasion.
While high involvement may initially motivate an individual
to reject a counterattitudinal message, the enhanced
cognitive processing of the message should enable the
virtues and flaws of the arguments to be objectively
recognized. Results of the cognitive response measures
supported these hypotheses. Under high involvement
conditions subjects produced more favorable cognitions and
fewer counterarguments to the strong than to the weak
arguments. Argument quality produced no significant
effects on cognitive responses under low involvement
conditions. Furthermore, high involvement increased the

47
amount of counterarguments generated to weak arguments and
increased the amount of favorable thoughts generated to
strong arguments. These results show that increased
involvement increases the distinction of strong and weak
arguments.
This interaction of personal involvement and argument
quality has been replicated several times (Petty, Cacioppo,
and Goldman, 1981; Petty, Cacioppo, and Heesacker, 1981;
Petty and Cacioppo, 1984). For example, in the context of
examining the effects of source expertise under conditions
of high and low involvement, Petty, Cacioppo, and Goldman
(1981) found that under high conditions objective cognitive
processing increased distinction of argument quality and
that argument quality was the primary determinant of
persuasion. Under low involvement conditions, the
peripheral cue of source expertise exerted a greater effect
than argument quality.
Further support for the notion that increasing
involvement heightens message scrutiny was provided by
studies in which involvement was manipulated along with
source and message cues. Results of Chaiken's (1980) work
indicate that high issue involvement subjects tend to use a
systematic information processing strategy, whereas low
involvement subjects tend to use a heuristic processing
strategy. Subjects who use a systematic strategy focus on

48
the content of persuasive messages and carefully evaluate
the content in a relatively objective manner.
Although the research cited above suggests that people
become more likely to evaluate carefully and objectively
the issue-relevant arguments in a persuasive message as
personal involvement increases, circumstances may occur in
which cognitive processing becomes biased as involvement
increases. As suggested in the section on biased
processing, the presence of a well-organized structure of
prior knowledge may lead to biased processing. An
individual who is involved with an issue has done some
amount of prior thinking about the pool of issue-relevant
arguments. Some individuals may have done a great deal of
prior thinking and have a well-developed construct system
or schema for the issue in question. Such individuals may
have a greater ability to counterargue persuasive messages,
a greater store of supportive arguments for their own
beliefs, and/or little motivation to consider yet another
persuasive appeal.
Ostrom and Brock (1968) suggest that when an issue is
intimately connected with an individual's central values,
its personal involvement may be intense enough to generate
biased processing. Greenwald (1980) proposes that the ego
is characterized by a cognitive bias of conservatism or
resistance to change that serves to protect the self's
organization of knowledge. If the ego is threatened, it

49
will respond with cognitions aimed at preserving the status
quo. Thus, extremely high levels of involvement, or
personal importance for a person's central aspects of
identity, may also lead to biased processing. Biased
processing may then consist of either a negative bias for
counterattitudinal messages or a positive bias for
proattitudinal messages.
Involvement Level in Counseling
Clear cases of high or low involvement probably do not
exist in the therapy room. People generally seek help with
issues in which they are involved to some significant
degree. Prior to pursuing therapy, they will have given
thought to how to deal with the issue and perhaps will have
tried many self-generated change attempts. They may even
have sought the advice of friends, family, and others.
Strong (1971) suggested that different types of
counseling may elicit different levels of involvement from
clients, but it seems clear that clients may also bring
different levels of involvement with them to counseling.
Stoltenberg and McNeill (1984) showed that college students
who were undecided about a career regarded the issue of
career exploration as more personally involving than
students who had already chosen a career. They also found
undecided students agreed more with a message advocating a
career exploration course than did decided students.

50
This finding suggests that more involved clients may
see counseling as more beneficial and may attend more to
the counseling process. Thus, clients with a high level of
involvement should adhere more to central route cognitive
processing of therapeutic messages, whereas clients with a
low level of involvement may depend upon the peripheral
route. Moreover, as central route processing appears to
lead to more enduring attitude change (Petty and Cacioppo,
1981), it behooves the therapist to motivate clients to
evaluate carefully the pros and cons of their present
behaviors and to weigh the costs and benefits of change and
various means of change.
An example of a highly involved client perhaps would
be the articulate, well-educated, and young business person
seeking counseling concerning improving his or her
relationships with co-workers and superiors. One would
expect this person to consider carefully the issues and
suggestions in counseling. The low involved client may be
represented by the person forced to attend an alcohol
therapy group due to a DWI conviction. He or she may
rather be at home or even just spend the weekend in jail
than attend therapy weekly for 3 months. Such an
individual may elaborate very little on the issues raised
in counseling but may be influenced by persuasion cues
(e.g., counselor social attractiveness).

51
On the other hand, client factors may also serve to
influence cognitive processing in a biased manner. An
individual may be so extremely involved in an issue that he
or she avoids thinking about it altogether (e.g., the
denial of a person with alcohol dependence). Processing
may also become biased in the service of one's own ego or
for self-protection (e.g., Greenwald, 1980). If a person
organizes his or her identity around being a certain type
of individual (e.g., an easy to get along with, always
ready to please, nonassertive person), he or she would
probably be less objective in considering therapeutic
messages designed to change that behavior. This would
especially be the case if this identity had provided the
individual with rewarding secondary gains in the past.
Such a client may be well prepared to generate cognition in
support of his or her present attitudes and behaviors and
actively counterargue any therapeutic communications aimed
at changing that basic identity. Thus, this biased
processing would lead to greater resistance to change.
Hypotheses
As outlined in the preceding sections of this review,
cognitive processing of persuasive messages may follow
either a central information-processing route (a careful
consideration of the message content) or a peripheral
information-processing route (a cue or decision rule basis
for persuasion). Within the more central route, cognitive

52
processing may be relatively objective or relatively
biased. As a result of this, individuals with different
levels of involvement will engage in different types of
cognitive processing.
The present study will screen for four groups of
subjects: (1) high ego and high issue involved, (2) high
ego and low issue involved, (3) low ego and high issue
involved, and (4) low ego and low issue involved. Subjects
will be asked to consider a message advocating the use of
the cognitive restructuring counseling technique for the
treatment of assertiveness problems. The topic of
assertiveness was chosen for this investigation because of
its relevance for a college population. Using measures of
CFCR, outcome expectation, behavioral intention, and
spontaneous use of cognitive restructuring to assess the
subjects' cognitive responses to strong and weak arguments
advocating the use of the technique, the following
predictions are made:
1. Replicating the findings of Petty and Cacioppo
(1979b), there will be a significant interaction
between issue involvement and argument quality.
This effect will be such that there will be greater
distinction between strong and weak arguments under
conditions of high issue involvement than under
conditions of low issue involvement.
2. A significant interaction is also predicted
between ego-involvement and argument quality with a
greater distinction between strong and weak
messages occurring under high than low ego-
involvement.
3. A biasing effect is also predicted to be
reflected in a significant interaction between

issue involvement and ego-involvement. Under
conditions of high issue involvement, high
ego-involvement will exert a greater influence on
the weak than the strong messages. Under low
involvement, high ego-involvement will exert an
equal influence on both strong and weak messages.

CHAPTER III
METHODS
This study was designed to investigate the effects of
ego-involvement and issue involvement upon the cognitive
processing of strong and weak arguments for the use of
cognitive restructuring as a treatment technique for
assertiveness. The sample for the study consisted of 120
students drawn from the University of Florida introductory
psychology subject pool and an undergraduate sociology
class, Marriage and Family. Cognitive responses and
acceptance of arguments was assessed across four groups: 1)
high ego and high issue involvement, 2) high ego and low
issue involvement, 3) low ego and high issue involvement,
and 4) low ego and low issue involvement. Half of each
group was exposed to strong arguments and half to weak
arguments in favor of using cognitive restructuring.
Levels of ego-involvement were assessed by using the
Relationships Grid (see Appendix A), a form of the dyad
grid (Ryle and Lunghi, 1970). Levels of issue involvement
were measured through the use of the Simple Rathus
Assertiveness Schedule (McCormick, 1984). The preceding
instruments were administered to one General Psychology
class and one Marriage and Family class.
54

55
Cognitive responses to strong and weak arguments for
cognitive restructuring were assessed through the use of
the Thought-Listing Technique (Cacioppo and Petty, 1981).
Several seven-point Likert-type scales on which subjects
rated their attitudes toward cognitive restructuring and
their intentions to use it were also used to measure their
acceptance or rejection of the arguments. An additional
Thought-Listing Technique was used to assess the
spontaneous use of cognitive restructuring by subjects
during a covert rehearsal technique.
This chapter is divided into several sections. In the
first section, the research design and variables of
interest are discussed. In the second section, the
population of interest and the study's sample and sample
selection and recruitment procedures are discussed. In the
third section, the instruments used in this study are
presented. In the fourth section, the experimental
procedure and data collection is described. In the fifth
section, the method of data analyses are presented.
Methodological limitations of the study and a summary of
the methodology are presented in the final two sections.
Research Design
This study involved a 2 x 2 x 2 (Ego-involvement: high
or low X Issue involvement: high or low X Intervention
Quality: strong or weak arguments) factorial design.
Subjects were crossed on the first two factors according to

56
a median split for issue involvement and upper and lower
divisions of a trichotomization of ego-involvement and then
randomly assigned to either strong or weak intervention
quality. Ego-involvement was operationalized in terms of a
summed relationship score for the construct, "Assertive
Nonassertive" on the Relationships Grid. The summed
relationship score (Bannister, 1965) is a measure of the
total variance within a construct system accounted for by
one construct. The greater the total variance for a
construct the greater the implications and importance it
holds for a person. The variable, ego-involvement, had two
levels: high and low. The high ego-involvement level
refered to students who scored in the uppermost division of
a trichotomization of the sample's summed relationship
scores for the construct, AssertiveNonassertive. The low
ego-involvement level refered to students who scored in the
lowermost division for that score.
The second factor, issue involvement, was
operationalized by the use of an objective inventory
assessing personal assertiveness, the Simple Rathus
Assertiveness Schedule (McCormick, 1984). Subjects who
scored above the median for the pool of subjects on the
objective inventory (high trait assertiveness) were
considered to have low issue involvement. As scores below
the median on the objective inventory would indicate above

57
average concern with assertion, subjects who scored in the
below the median range comprised the high issue involvement
condition.
The third factor, argument quality, consisted of two
levels, strong arguments and weak arguments. Strong and
weak arguments for the use and effectiveness of the
cognitive restructuring technique were developed by Greg
Neimeyer and his research team (Unpublished data, 1986).
Following the procedure outlined by Petty and Cacioppo
(1981) for developing arguments for a topic, they generated
strong and weak arguments for the use of cognitive
restructuring in the treatment of eating disorders. In the
present study these arguments have been adopted to the use
of cognitive restructuring in the treatment of
assertiveness.
Cognitive Responses to Arguments
The variables of interest in the present study
consisted of the subjects' cognitive responses to arguments
for cognitive restructuring. Cognitive responses were
operationalized in terms of cognitive favorability toward
cognitive restructuring, outcome expectation and behavioral
intention ratings, and judges' ratings of spontaneous use
of cognitive restructuring by subjects during a covert
rehearsal condition.
Cognitive favorability toward cognitive restructuring
(CFCR) was derived from the Thought-Listing Technique

58
(TLT). The TLT is used to identify a person's subjective
thoughts or reactions to a topic. In the TLT subjects were
instructed to list the thoughts they had while listening to
a taped message on cognitive restructuring (CR).
Subsequently, they also rated whether their thoughts were
in favor of using CR, opposed to using CR, or neither in
favor of nor opposed to it. A general index of CFCR was
calculated for each subject by subtracting the number of
unfavorable thoughts from the number of favorable thoughts
listed and then dividing by the total number of relevant
thoughts yielding a ratio score (see Cacioppo and Petty,
1981) Higher CFCR scores indicated more positive or
favorable thoughts toward CR, while lower or negative CFCR
scores indicated less favorable thoughts or opposition to
the use of the CR technique. It was expected that under
conditions of high issue involvement individuals with high
ego-involvement would have lower CFCR than individuals with
low ego-involvement. It was also expected that under
conditions of low issue involvement individuals with high
ego-involvement would have greater CFCR than individuals
with low ego-involvement.
Cognitive responses toward the CR technique were also
operationalized in terms of outcome expectations and
behavioral intention ratings. The outcome expectation
asked subjects to indicate, "To what extent do you think
the cognitive restructuring technique would be beneficial

59
to you, personally?" As a message of behavioral intention,
subjects were asked to indicate "How likely are you to use
the cognitive restructuring technique in the future in
situations requiring assertion"? Subjects responded to
those items as well as a number of manipulation check and
ancillary items on a seven point Likert-type scale. It was
expected that analysis of outcome expectation ratings would
parallel the predictions for CFCR. In addition, under
conditions of high issue involvement it was expected that
subjects with low ego-involvement would have greater
intentions to use the CR technique than subjects with high
ego-involvement.
Finally, the spontaneous use of cognitive
restructuring during a covert rehearsal condition (see
Appendix E) was also used to operationalize subjects'
cognitive responses to the CR arguments. It was reasoned
that subjects with greater favorability toward cognitive
restructuring and greater behavioral intention to use it
will generate more CR thoughts during a covert rehearsal of
a situation requiring assertion.
Description of the Sample
The population of interest in this study included all
undergraduate students attending the University of Florida,
Gainesville, Florida.
The sample in this study was drawn from the University
of Florida introductory psychology subject pool and from an

60
undergraduate sociology course on marriage and family at
the University of Florida. During the first week of spring
semester 1987, students in one section of Psychology 2013
(General Psychology) and Sociology 2430 (Marriage and
Family) were asked to complete a pretest consisting of
identifying information, the Relationships Grid, and the
Personal Problems Inventory (see Appendices A & B).
On the basis of the scores from these two pretests,
subjects were divided into four separate groups: 1) high
ego and high issue involvement, 2) high ego and low issue
involvement, 3) low ego and high issue involvement, and 4)
low ego and low issue involvement. In order to assign
students to these four groups the following procedure was
followed. First, all students will be rank ordered
according to their summed relationship scores for the
construct, "AssertiveNonassertive" on the Relationships
Grid, and a trichotomization was performed on the basis of
these scores. The uppermost division constituted the high
ego-involvement group; the lowermost division constituted
the low ego-involvement group. Second, all students were
rank ordered according to their scores on the Simple Rathus
Assertiveness Schedule, which is entitled the Personal
Problem Inventory in this study. On the basis of these
scores a median split was performed, and the upper half of
the split consisted of the low involvement group, while the
lower half consisted of a high issue involvement group.

61
Third, students were assigned to the appropriate group
according to their high and/or low classifications on ego
and issue involvement. This procedure resulted in the
inclusion of 104 subjects (N=40, F=64). The percentage of
males within each group is presented in Table 1. Mean ego-
involvement and issue involvement scores and standard
deviations for all groups in this study are also listed in
Table 1.
Instrumentation
Six instruments were used in this study: 1) The
Relationships Grid, 2) Personal Problems Inventory (Simple
Rathus Assertiveness Schedule), 3) Assertion Survey, 4)
Thought Listing for Message, 5) Covert Rehearsal
Technique, and 6) Cognitive Assessment Questionnaire.
Relationships Grid
The Relationships Grid (RG; see Appendix A) is a form
of the dyad grid (Ryle & Lunghi, 1970). The RG is an
adaptation of the original repertory grid technique
developed by G.A. Kelly (1955) to elicit and measure
personal construct systems. "A grid may be defined as any
form of sorting task which allows for the assessment of
relationships between constructs and which yields these
primary data in matrix form" (Bannister and Mair, 1968, p.
136). The basic components of a grid are elements and
constructs. In the RG, like other dyad grids, the elements
used consist of relationships (e.g. my relationship with my

Table 1. Means with Standard Deviations in Parentheses for Levels of
Issue Involvement and Ego-Involvement
Variables
Independent
Descriptive
Issue Involvement
Ego-Involvement
n
%M
Rathus Score
Variance Score
High
High
30
24
77.08(10.60)
115.46(16.14)
High
Low
22
38
78.12 (6.11)
27.09(13.32)
Low
High
24
58
110.293(8.15)
119.80(24.81)
Low
Low
28
32
111.80 (8.50)
21.31(11.55)
N)

63
spouse, my relationship with my father) instead of
individuals as in the original repertory grid. Dyad grids
use relationships as elements in order to gain information
about a person's interpersonal behavior across a range of
important relationships in his/her life. Dyad grids have
been used in the past in clinical work with married couples
(Ryle and Breen, 1972a and 1972b; Ryle and Lipshitz, 1975
and 1976) and in individual therapy to provide hypotheses
for the conduct of therapy (Ryle and Lunghi, 1971; Ryle,
1979a; Ryle 1981) and to assess change (Ryle, 1979b; Ryle,
1980). Elements and constructs for this study's RG were
adapted from Ryle's Relationships Grid (Ryle, 1985) and
modified for the purposes of this study. For a complete
listing of elements and constructs see Appendix A. The
elements of interest consisted of the subjects'
relationships with several important figures in their lives
and how they saw themselves in those relationships.
Constructs described interpersonal behavior and were
presented in the classical bipolar form.
The RG, a ten by ten grid, involves having subjects
rate how they see themselves in their relationships with
ten interpersonal figures (e.g., mother, sibling, roommate)
along ten provided constructs using a seven-point Likert-
type scale. For example, a subject will be asked to rate
his/her relationship with mother along the construct,
Assertive +3 +2 +1 O -1 -2 -3 Nonassertive. The subject

64
may believe him/herself to be moderately nonassertive in
that relationship and place a -2_ in the appropriate square
in the grid. The left pole or side of each construct will
be assigned a positive valence, while the right pole will
be assigned a negative valence. The Relationships Grid and
complete instructions are provided in Appendix A.
Estimated time for completion of the RG is fifteen minutes.
Numerous measures can be derived from a grid, but the
measure of interest for this study is the summed
relationship score for the construct, "Assertive-
Nonassertive." The summed relationship score (Bannister,
1965) is a measure of the total variance within a construct
system accounted for by a single construct. The greater
the total variance accounted for by one construct; the more
closely it is related to all other constructs. A construct
with greater variance and more implications for the rest of
an individual's construct system would be more
superordinate and hold more importance for that individual
than a construct accounting for less variance in the
system. In this study subjects with higher summed
relationship scores for whom the construct, Assertive-
Nonassertive, accounted for a greater amount of variance in
their construct systems were designated as highly involved.
Subjects for whom the assertiveness dimension accounted for
less variance in the system were designated as low ego
involved.

65
The calculation of variance scores is described by
Bannister (1965) Pearson product-moment correlations are
first calculated between all constructs. Second, each
correlation is squared, multiplied by 100, and the original
sign retained. These scores are called relationship
scores. The relationship scores for any one construct may
be summed to give the total variance for that construct.
In this study the relationship scores for the construct,
"Assertive-Nonassertive," will be summed to give the total
variance accounted for by that construct. If this score is
then divided by the total variance in the system across all
constructs, then it yields an index of the proportion of
variance accounted for by the single construct Assertive-
Non Assertive. A trichotomization was performed along the
distribution of these scores. The uppermost division was
designated as high ego-involvement and the lowermost as low
ego-involvement.
Investigations into the psychometric properties of the
Rep Grid are complicated by the varieties of grids and
scorings in use. While the reliability and validity of
summed relationship scores is not reported in the
literature, psychometric properties of total variance
scores (Bannister, 1960) are reported. Bannister (1962)
reports an immediate retest reliability of 0.35 in a sample
of 30 normal subjects. Honess (1977) reports the same
reliability correlation in a rank order grid but a

66
correlation of 0.62 for a bi-polar implications grid.
While reliability for total variance scores is generally
low, this may be due to its sensitivity to construct system
change as the individual and/or situation changes. Total
variance scores have been shown to effectively discriminate
thought-disordered schizophrenics from normals and other
psychiatric groups (Bannister, 1962; Bannister and
Fransella, 1966).
Rather than using the total variance score, this study
uses the proportion of variance accounted for by the single
relevant construct. The psychometric properties of this
score are not yet established. It seems reasonable to say
that despite the instability of the overall variance in the
system, the proportion of variance accounted for by any one
construct might remain relatively constant. To test this
assertion 30 students from an undergraduate psychology
course were administered the RG twice with a 7 day interval
between administrations. Proportion scores for the
relevant construct were used to calculate a Pearson
product-moment correlation. This test-retest correlation
was .66, (p <.0001).
Personal Problems Inventory
The Personal Problems Inventory (PPI; see Appendix B)
was used to assess issue involvement concerning assertion.
It consisted of the Simple Rathus Assertiveness Schedule
(SRAS) (McCormick, 1984). The SRAS is an easier-to-read,

67
revised version of the Rathus Assertiveness Schedule
(Rathus, 1973). The SRAS is a 30-item self-report
assertiveness inventory, which provides a global rating of
perceived trait assertiveness. Respondents mark each item
in terms of how characteristic the behavior is of the
individual from 5 (very much like me) to 1 (very unlike
me). A total score is obtained by summing item scores
after correcting for reversed scoring weights.
In an investigation of the test-retest reliability of
the SRAS, thirty undergraduate psychology students were
administered the SRAS twice with a 7 day interval between
adminstrations. A Pearson product-moment correlation
yielded a coefficient of .91 (p <.0001).
McCormick (1984) provides evidence for a satisfactory
degree of equivalence between the Simple Rathus
Assertiveness Schedule (SRAS) and the Rathus Assertiveness
Schedule (RAS). He reports a mean interim correlation of
0.79 between the two versions of the test with all
correlations reaching statistical significance. The
correlation between the total scores of the two tests was
0.94. He also reports a correlation of 0.90 between total
odd and even item scores for both versions. Finally, in
his sample of 116 undergraduate students he found an almost
identical distribution of scores on both tests. Mean
scores and standard deviations for the two versions were
M=96.12, SD=23.85 (SRAS) and M=96.29, SD=24.94 (RAS).

68
Psychometric research on the RAS suggests that it has
moderately high stability (test-retest reliability) and
moderate to high homogeneity. Rathus (1973) presents a
five-week test-retest reliability of 0.78 and a split-half
Pearson product moment correlation of 0.77 between total
odd and total even item scores. Vaal (1975) presents
similar correlations of 0.76 for test-retest reliability
over an eight-week period and 0.77 for split-half internal
consistency. With college students seeking assertion
training Heimberg and Harrison (1980) obtained over an
eleven to fifteen day period a test-retest reliability
correlation of 0.80 and a split-half internal consistency
correlation also of 0.80. Other reports of internal
consistency correlations have ranged from 0.69 to 0.86
(Mann and Flowers, 1978; Quillan, Besing, and Dinning,
1977; Futch, Scheirer, and Lisman, 1982).
A base of validity information has also been
established for the RAS. When comparing RAS scores with
peer ratings on a seventeen-item semantic differential
Rathus (1973) found that total RAS scores correlated
positively with each of the items comprising the peer
rating scale's assertiveness factor. Several studies have
tested Wolpe's (1973) argument of an inverse relationship
between trait anxiety and assertiveness; conformation of
Wolpe's argument would provide indirect support for the
construct validity of the RAS. Orenstein, Orenstein, and

69
Carr (1975) found RAS scores to be significantly correlated
(range -0.60 to -0.75) with interpersonal fears as measured
by the social factors from the Wolpe-Lang (1964) Fear
Survey. Other studies by Morgan (1974) and Hollandsworth
(1976) have found lower correlations between RAS scores and
a social fear survey. In Morgan's study of college
students the correlations ranged from 0.172 to -0.239 and
in Hollandsworth's study a correlation of 0.436 was found.
The combined data from these studies does appear to provide
some support for the construct validity of the RAS.
Concurrent validity of the RAS has also been
established in several studies. Rathus (1973) reports a
correlation of 0.705 between RAS scores and observer
ratings of verbal behavior in response to five questions
asking for assertive behavior. Frankel (1977) reports
significant correlations between the RAS and the Conflict
Resolution Inventory and between the RAS and the Assertion
Inventory. MacDonald (1975/1974) also supports the
concurrent validity of the RAS in her findings of a
moderate relationship between RAS scores and behavioral
measures of assertiveness. In addition, the RAS has been
used in treatment outcome studies and has been shown to be
a sensitive index of pretreatment to post treatment change
(Rathus, 1973; Blanchard, Turner, Eschette, and Coury,
1977) .

70
Assertion Survey
The Assertion Survey (AS; see Appendix C) is a
questionnaire developed by the researcher to assess outcome
expectations. It consisted of eight 7-point Likert-type
items. Outcome expectation was assessed by the item, "To
what extent do you think the cognitive restructuring
technique would be beneficial to you, personally?"
Behavioral intention was assessed by the item, "How likely
are you to use the cognitive restructuring technique in the
future in situations requiring assertion?" As a check on
the argument quality manipulation, subjects answered a
seven-point Likert item asking them to rate the quality of
the arguments in the intervention: very poor arguments (1)
to very good arguments (7). In addition, as a general
check on the issue involvement manipulation, subjects
answered another Likert item asking them the extent to
which the message has implications for them personally:
not at all relevant to me (1) to very relevant to me (7).
Ancillary items assessed other aspects of the message
including voice quality, speaker qualification, and ratio
of delivery.
Thought Listing Techniques
The Thought Listing Techniques used in this study were
variations of the "thought-listing procedure" developed by
Brock (1967) and Greenwald (1968). The procedure has been
refined to become a suitable self-report technique for

71
obtaining a written listing of an individual's thoughts on
a topic and a general index of CFCR toward the topic
(Cacioppo and Petty, 1981). In the present study Thought
Listing was employed under two conditions: 1) after
listening to the taped arguments for cognitive
restructuring and completing the Assertion Inventory and 2)
as part of the covert rehearsal of a personal assertion
situation. Under the first condition the Thought Listing
Technique (see Appendix D) was used to compute a CFCR
index, and under the second condition (see Appendix E) it
was used to tally the spontaneous use of cognitive
restructuring. Under the first condition subjects listed
their thoughts concerning the arguments for CR according to
directions similar to those of Petty and Cacioppo (1977).
In addition, they rated their thoughts along the dimension
of favorableness toward cognitive restructuring. The
cognitive favorability index was calculated for each
subject by subtracting the number of unfavorable thoughts
from the number of favorable thoughts listed and then
dividing by the total number of relevant thoughts (Cacioppo
and Petty, 1981). Petty et al. (1976) report correlations
between subjects' own rating and independent judges ratings
of .82 for favorable thoughts and .79 for counterarguments.
Under the second condition, two undergraduate psychology
major students trained by the researcher and unaware of the
hypotheses of the study and the involvement group

72
classifications of the subjects judged the thoughts listed
for frequency of spontaneous use of cognitive
restructuring. This consisted of the post-coding of each
subject's listed thoughts into one of five categories: 1)
self-instructional statement, 2) self-assuring statement,
3) self-doubting statement, 4) presentation content
statement, and 5) other statement. Inter-judge reliability
Pearson product-moment correlations were .81 for self-
instructional statements, .85 for self-assuring statements,
.92 for self-doubting statements, .72 for presentation
content, and .34 for other statements.
Support for the Thought Listing Technique (TLT) as a
reliable measure can be found in a study by Cullen
(1969/1968). She compared the split-half reliability and
test-retest reliability of the TLT with Likert attitude and
Thurstone attitude scales on two topics, birth control and
segregation. .She found average split-half reliabilities of
0.78 for Thought-Listing, 0.83 for Likert scales, and 0.55
for Thurstone scales. The average test-retest
reliabilities were 0.64 for the TLT, 0.83 for Likert
scales, and 0.53 for Thurstone scales.
In indirect support for the construct validity of the
TLT, Petty, Harkins, and Williams (1980) found that the
implied or real presence of others working on the same
cognitive task decreased the thoughts listed by an
individual in that implied or real presence of others when

73
compared to the individual working alone. This finding
corresponds to the well-known "social loafing" argument
(Latane and Darley, 1970).
Cognitive Assessment Questionnaire
The Cognitive Assessment Questionnaire (CAQ; see
Appendix F) was used to assess subjects' tendency to engage
in and enjoy effortful cognitive endeavors. It consisted
of the short form of the Need for Cognition Scale (NCS)
(Cacioppo, Petty, and Kao, 1984). The short NCS is an 18-
item self-report inventory which provides a global
assessment of need for cognition. Cacioppo, Petty, and Kao
(1984) report a significant correlation (r = .95, p <,001)
between subjects' scores on original and 18-item versions
of the NCS. The NCS shows strong internal consistency (r =
.76, p<.001), and some evidence exists for its content and
predictive validity (Cacioppo & Petty, 1982).
Procedures
Procedures were set up to insure standardization of
administration. Instruments were presented to each subject
by cassette tape (see Appendix G for a complete transcript
of the tapes for each experimental condition).
Subjects were run in group administrations in a
language laboratory classroom. They were seated and
provided with a package of written materials, the measures
of interest. They were then instructed by the researcher
to place their headphones on and follow the instructions

74
that they heard on the tape carefully. All further
instructions to the subject were on the taped message.
The taped message consisted of a general introduction
to the study and the cognitive restructuring technique for
all subjects. Half of the subjects in each involvement
level group were randomly assigned to one of two
experimental conditions, either strong arguments in favor
of the effectiveness of cognitive restructuring or weak
arguments favoring cognitive restructuring. (See Petty &
Cacioppo, 1984, for a discussion of the development of
these messages). The next section of the tape that the
subjects heard presented either strong or weak arguments
for cognitive restructuring depending upon the experimental
condition being conducted. This was followed by
instructions for the completion of the Assertion Survey
followed by three minutes of silence to enable the
completion of the questions. Next, instructions for the
thought listing for the message were presented and followed
by three minutes of silence during which subjects were
instructed to "List all of the thoughts that came to mind
as you listened to the taped message." Instructions for
the Cogntive Assessment Questionnaire were presented next
and followed by five minutes of silence during which the
subjects completed the questionnaire. The tape resumed
with instructions for covert rehearsal of a personal
assertion situation and its associated thought listing,

75
again allowing for three minutes to complete the task.
Finally, subjects were instructed to remove their
headphones, turn in their materials to the researcher, and
receive their experimental credit from the researcher.
All subjects participating in the study will be given
a subject number code which will identify their responses
while maintaining their anonymity. All subjects also
received either experimental credit or extra class credit
for their participation in the study.
Analyses of Data
The analyses compared the impact of high and low
levels of ego-involvement, issue involvement, and message
quality on attitudes toward cognitive restructuring.
Dependent variables included an index of cognitive
favorability in response to the taped messages, outcome
expectation attitude ratings, behavioral intention ratings,
and post-coding of covert rehearsal thoughts. A series of
2x2x2 analyses of variance were performed to determine
the effects of ego-involvement, issue involvement, argument
quality, and interactions on these variables. Duncan's
Multiple Comparison Procedure was used to evaluate the
effects predicted in the research hypotheses.
Summary
This study was designed to test and compare the
varying levels of ego-involvement, issue involvement, and
argument quality upon a subject's cognitive responses to

76
the cognitive restructuring technique presented in a
message advocating its use in treating assertiveness
problems.
The design was a 2 x 2 x 2 between subjects factorial
design. The factors were ego-involvement (high and low),
issue involvement (high and low), and argument quality
(strong or weak). The dependent variables consisted of the
following: 1) cognitive favorability index, 2) outcome
expectations for the use of the CR technique, 3) behavioral
intention toward cognitive restructuring, and 4)
spontaneous use of CR during covert rehearsal.
The sample was drawn from the University of Florida
introductory psychology subject pool and an undergraduate
sociology class. It consisted of students pretested to
form four distinct groups: 1) high ego involved, high
issue involved; 2) high ego, low issue involved; 3) low
ego, high issue involved; and 4) low ego, low issue
involved.
Instruments used in this study included the
Relationships Grid, Personal Problems Inventory (which
consisted of the Simple Rathus Assertiveness Schedule),
Assertion Survey, Cognitive Assessment Questionnaire (Need
for Cognition Scale), and Thought Listing Techniques for
messages about cognitive restructuring and covert

77
rehearsal. Procedures for collection and evaluation of
data were standardized in order to prevent bias between
groups.
Analysis of data compared the four involvement groups
and argument quality. A 2 x 2 x 2 ANOVA for each cognitive
response variable determined the effects of ego-
involvement, issue involvement, argument quality, and
interactions.

CHAPTER IV
RESULTS
Manipulation Checks
Before investigating whether the manipulations produced
the desired effects on the dependent measures, the
effectiveness of the manipulations should be ascertained.
Manipulation checks were planned for argument quality and
issue involvement and analyzed using a series of 2 x 2 x 2
analyses of variance.
Argument quality
To see whether subjects differentiated the strength of
the arguments used in the message quality manipulation,
subjects answered a seven-point Likert item asking them to
rate the quality of the arguments in the intervention:
very poor arguments (1) to very good arguments (7).
Results of the Analyses of Variance indicate that the
manipulation successfully influenced how strongly the
subjects rated the quality of the two groups of arguments,
F(1,96)=30.43, p<.0001: strong arguments M=5.0, weak
arguments M=3.5. The involvement variables (issue and ego)
did not significantly affect this manipulation check,
either alone or in interaction with the argument quality
manipulation.
78

79
In addition, as predicted by the ELM subjects who heard
the strong argument quality message generated a
significantly greater proportion of positive thoughts
(M=.76) than did subjects who heard the weak message
(M=.55), F(l,90) = 12.21, p<.0007. Furthermore, subjects
who heard the message with weak arguments generated a
significantly greater proportion of negative thoughts
(M=.45) than did subjects who heard the strong message
(M=.23) F(1,90) = 12.21, p<.0007.
Issue Involvement
The other category of manipulation checks investigated
the degree of issue involvement that subjects believed that
they had with the use of the cognitive restructuring
technique for treating personal problems in assertion. As
an initial check on issue involvement a seven-point Likert
item asking subjects to indicate the extent to which the
message had implications for them personally was included.
An analysis of variance revealed a trend toward a main
effect of issue involvement, F(l,96)=2.8, p=.10; high issue
M=4.25, low issue M=3.72. All other variables (ego and
argument quality) also did not significantly affect this
manipulation check. As these results cast doubt on the
effectiveness of the issue involvement manipulation, an
alternative operationalization of involvement was pursued.
Petty and Cacioppo (1986) suggest that as personal
involvement increases, people become more motivated to

80
process the issue-relevant arguments presented. One way in
which argument processing could be measured would be by the
number of relevant cognitions that a person generates. In
order to assign subjects to high and low levels of argument
processing (high and low issue involvement) a median split
was performed on scores of total number of relevant
thoughts generated by subjects on the thought listing
technique. The median score was deleted and subjects in
the upper half of the distribution constituted the high
issue involvement group, while subjects in the lower half
constituted the low issue involvement group. Analysis of
variance of the involvement manipulation check revealed
that high issue involvement subjects believed that the
message had significantly more implications for them than
did the low issue involvement group, F(1,96)=6.17, p<.01:
high issue involvement M=4.4, low issue involvement M=3.6.
The other variables (ego-involvement and argument quality)
did not significantly affect this manipulation check.
Dependent Measures
Data for each of the dependent measures was analyzed by
a 2(High and Low Issue Involvement) x2(High and Low Ego-
Involvement) x2(Strong and Weak Argument Quality) Analyses
of Variance. Analyses for each dependent measure are
presented separately.

81
Cognitive Favorability (CFCR)
The analysis of variance revealed a significant main
effect of argument quality for the CFCR score,
F(l,90)=12.21, p<.001, such that subjects who heard the
strong arguments had a greater proportion score for overall
cognitive favorability toward cognitive restructuring
(M=0.54) than did subjects who heard the weak arguments
(M=0.10) (see Table 2). Means and standard deviations for
CFCR for all levels of the design are presented in Table 3.
Attitude Rating
The analysis of variance revealed a significant main
effect of issue involvement for the outcome expectation
attitude rating, F(1,96)=9.23, p<.003, such that subjects
in the high issue involvement condition believed that the
cognitive restructuring technique would be significantly
more beneficial to them (M=4.58) than did subjects in the
low issue involvement condition (M=3.76) (see Table 4).
Means and standard deviations for attitude rating are
presented in Table 5.
The results also revealed a trend toward a main effect
of argument quality for the attitude rating, F(1,96)=3.54,
p=.06. These results suggest that the group that heard the
strong arguments had a tendency to believe that the
cognitive restructuring technique would be more beneficial
to them (M=4.4) than did the group that heard the weak
arguments (M=3.9).

Table 2. Analysis of Variance for CFCR Scores
Source
Sum of squares
df
Mean Square
F
P
Issue Involvement
0.989
1
0.989
0.22
.642
Ego
0.004
1
0.004
0.01
.919
Quality
5.011
1
5.011
12.21
.001
Issue Involvement x Ego
0.109
1
0.109
0.27
.607
Issue Involvement x Quality
0.016
1
0.016
0.04
.845
Ego x Quality
0.221
1
0.221
0.54
.465
Issue x Ego x Quality
0.374
1
0.374
0.91
. 342
Error
36.929
90
0.410
03

Table 3
Means with Standard Deviations in Parentheses for CFCR Score
for the Three-Way Interaction of Issue Involvement, Ego-
Involvement, and Argument Quality
Variables
Independent
Dependent
Issue Involvement Ego-
Involvement
Argument Quality
n
CFCR Score
High
High
Strong
11
0.47(0.64)
High
High
Weak
12
0.01(0.64)
High
Low
Strong
13
0.52(0.40)
High
Low
Weak
14
0.12(0.57)
Low
High
Strong
14
0.50(0.61)
Low
High
Weak
14
0.24(0.81)
Low
Low
Strong
11
0.67(0.49)
Low
Low
Weak
9
-0.04(0.89)

Table 4. Analysis of Variance for Attitide Rating Scores
Source
Sum of Squares
df
Mean Square
F
P
Issue
Involvement
19.254
1
19.254
9.23
.003
Ego
4.507
1
4.507
2.16
.145
Quality
7.358
1
7.358
3.53
.063
Issue
Involvement x
Ego
1.178
1
1.178
0.57
.454
Issue
Involvement x
Qual
1.338
1
1.338
0.64
.425
Ego x
Quality
0.575
1
0.575
0.28
.601
Issue
x Ego x Qual
0.241
1
0.241
0.12
.734
Error
200.187
96
2.085
CO
4^

Table 5. Means with Standard Deviaions in Parentheses for Attitude Rating and
Behavioral Intention Score for the Three-Way Intervention of Issue
Involvement, Ego-Involvement, and Argument Quality
Variables
Independent
Dependent
Issue Involement
Ego-Involvement Argument
Quality
n Attitude Rating Behavioral Intention
Score
High
High
Strong
11
4.73 (1.19)
4.36(1.57)
High
High
Weak
12
4.67(1.78)
4.25(2.05)
High
Low
Strong
13
4.77(1.48)
4.46(1.66)
High
Low
Weak
14
4.21(1.58)
4.36(1.45)
Low
High
Strong
15
4.40(1.35)
4.40(1.72)
Low
High
Weak
16
3.69(1.49)
3.94(1.65)
Low
Low
Strong
11
3.82(1.08)
3.91(1.04)
Low
Low
Weak
12
3.00(1.41)
3.33(1.97)
oo

86
Behavioral Intention Rating
The analysis of variance revealed no significant
findings for the behavioral intention score. Means and
standard deviations for the behavioral intention score are
presented in Table 5.
Spontaneous Use of Cognitive Restructuring
Only the post-coding categories of self-instructional
(SI), self-assuring (SA), and self-doubting (SD) statements
were deemed to have adequate inter-judge reliability (range
.81 to .92). These three categories were analyzed by
separate 2x2x2 Analyses by Variance.
Self-instructional statements. The analysis of variance
revealed no significant findings for the SI score.
Self-assuring statements. The analysis of variance
revealed no significant findings for the SA score.
However, the results did reveal a trend toward an
interaction effect of ego-involvement x argument quality
for the SA score, F(l,89)=2.8, p=.10. Examination of the
means suggests that under the high ego-involvement
condition argument quality did not influence the number of
self-assuring statements generated, but under the low ego-
involvement condition subjects who heard the message with
strong agruments generated significantly less self assuring
statements than subjects who heard the message with weak
arguments.

87
Self-doubting statements. The analysis of variance
revealed no significant findings for the SD score.
Additional Analyses
Correlations computed between eight of the design's
variables using Pearson product-moment correlations yielded
several significant relationships. The correlations most
germane to the present study included those between the
need for cognition variable and the other variables in the
study. These analyses yielded only one significant
correlation between the need for cognition score and the
score on the Simple Rathus Assertiveness Scale (SRAS), r =
.372, pC.0001. This result suggests that as subjects
increase in need for cognition, they see themselves as more
assertive. Perhaps even more interesting is the
nonsignificant correlation between the need for cognition
variable and the total number of relevant thoughts
generated by the subjects. This nonsignificant correlation
does not support Cacioppo and Petty's (1984) notion that
individuals high in need for cognition would tend to think
more extensively about messages presented to them.
Correlations are presented in Table 6.

Table 6. Intercorrelation of Descriptive and Dependent Variables
Ego-
Involvement
Argument
Quality
Attitude
Rating
Behavioral
Intention
CFCR
Need for
Cognition
Rathus
Scores
Relevant
Thoughts
Issue -.114
Involvement3 p<.249
-.001
p< .988
-.274
p< 005
-.132
p< 181
.065
p< 525
.006
p<.948
-.034
p<.730
-.841
p< .0001
Ego-
Involvement3
.001
p< .988
-.112
p<.259
-.055
p<.576
.017
p<.866
. 117
p<.234
.129
p<.191
.080
p<.419
Argument
Quality3
-.184
p< 06 2
-.097
p<.3 28
-.333
p< .0008
. 121
p<.222
-.019
p< .842
-.018
p< .852
Attitude
Rating
.730
p< .0001
.359
p< .0003
.012
p<.902
-.085
p<. 389
.243
p<.013
Behavioral
Intention
.424
p< 0001
. 131
p<.18 6
-.042
p<.668
.109
p<.271
CFCR
.034
p<.7 37
.052
p<.614
.055
p<.590
Need for
Cognition
.372
P<.0001
.014
p<.88 8
aReverse scored.
00
00

CHAPTER V
DISCUSSION
The primary purpose of the present investigation was
to examine the effects of issue involvement and ego-
involvement upon the cognitive responses that individuals
generate to arguments advocating the use of cognitive
restructuring to treat problems in assertiveness.
Hypotheses for these effects were developed from a review
of the social psychological literature on the Elaboration
Likelihood Model of Attitude Change. The present study
attempted to extend this model to include messages that may
be used in a counseling context.
In discussing variables that can affect a person's
ability to process message arguments Petty and Cacioppo
(1986) cite personal relevance/involvement as "perhaps the
most important variable in this regard" (p. 144). Their
ELM suggests that as issue involvement increases, people
become more motivated to engage in the cognitive work
necessary to evaluate the issue-relevant arguments
presented in a message. Petty and Cacioppo (1979b) provide
evidence consistent with this view. The first hypothesis
of the present study was also based upon this view. More
specifically, it predicted a significant interaction
89

90
between issue involvement and argument quality such that
there would be greater distinction between strong and weak
arguments under conditions of high issue involvement than
under conditions of low issue involvement. The predicted
issue involvement X argument quality interaction was not
found to be significant for any of the dependent measures.
The second hypothesis made a similar prediction of a
significant interaction between ego-involvement and
argument quality such that there would be a greater
distinction between strong and weak arguments under
condition of high ego-involvement than under conditions of
low ego-involvement. Once again, the predicted interaction
was not found for any of the dependent measures.
The results do indicate that the strength of the
argument quality significantly differentiated scores on the
CFCR dependent measure and demonstrated a strong tendency
to differentiate scores on the attitude rating dependent
measure. These results, coupled with the significant
argument quality manipulation check, suggest that overall
the content (arguments) of a counseling message made a
major difference in its effectiveness in advocating
cognitive restructuring for treating assertiveness
problems. Also consistent with the ELM Model was the
finding that strong messages produced more positive
cognitive responses, whereas weak messages generated more
negative thoughts. For the counseling situation this

91
implies that it may be advantageous for the therapist to
screen out distractions, to repeat important messages over
a number of sessions, and even to provide the client with a
written summary of central points made during therapy, all
manipulations that may also have the effect of enhancing
client involvement.
This effect of the message variable also has other
important implications for counseling. According to the
ELM, clients who carefully evaluate message content are
more likely to have enduring, behavior-related, central-
route attitude change than are those who attend only to
peripheral cues, such as counselor personal
characteristics. Message variables (such as argument
quality) may play crucial roles in eliciting behavior
change and in the maintenance of therapy gains. Future
research should investigate message factors to determine
those that are most helpful to counselors in bringing about
client change.
Results also show that issue involvement produced a
significant main effect for attitude rating such that
subjects who generated more thoughts about the message
believed that cognitive restructuring had more personal
benefits than did subjects who generated fewer thoughts.
Theoretically, an explanation of this finding could be
found in Zajonc's (1980) theory of mere exposure, which
describes changes in evaluation of objects as a result of

92
relatively primitive affective and associational processes.
It may be that those subjects who thought more about the
message associated it more with counseling and the common
notion that counseling is beneficial. Because a main
effect for issue involvement was not significant for the
CFCR measure, it does not appear that high issue
involvement subjects saw the cognitive restructuring
technique itself as being better than did the low issue
involvement subjects. In other words, greater favorability
toward the technique was related to more thinking in
general, but not to more favorable thinking per se.
This theoretical explanation further suggests some
methodological factors that may have affected the results
of the study. It may be that the high and low levels of
issue involvement in this study do not correspond to the
high and low levels of issue involvement of previous ELM
research reported in the social psychological literature.
In previous research involvement manipulations were
designed to induce relatively pure forms of the central and
peripheral routes to persuasion. For example, in the
Petty, Cacioppo, and Goldman (1981) study high involvement
subjects were told that the message they were to evaluate
had relatively important consequences for them (i.e., if
they did not pass the senior comprehensive exam that was to
be instituted the next year, they would not graduate). In
contrast to this, low involvement subjects were led to

93
believe that the proposal would have no personal
consequences for them. These two groups of subjects
represent relatively extreme levels of involvement. In the
present study, and in the counseling situation in general,
the issues of concern are applicable to almost everyone.
That is to say, most topics discussed in counseling (for
example, assertion, problems in personal relationships) are
by definition highly involving since such issues are only
presented if they are personally problematic concerns.
Therefore, the levels of issue involvement in the present
study might best be described as moderately high and
moderately low. Further support for a moderate level of
involvement came from the nonsignificant difference on
Rathus scores between high and low issue involvement
conditions. This could explain the current failure to
duplicate the interaction between issue involvement and
argument quality found in previous ELM research. This same
argument could also be extended to the ego-involvement
variable.
Another methodology question concerns the
operationalization of issue involvement in counseling-
related research. Past investigations have attempted to
manipulate issue involvement in a number of ways. These
attempts have included manipulation of clients' "perceived
need" or requests for help (Dixon & Claiborn, 1981; Heppner
& Dixon, 1978) and motivations for counseling (Heppner &

94
Heesacker, 1982). Such manipulations have resulted in no
differential effects. More recent studies (Stoltenberg &
McNeill, 1984; Heesacker, 1986) have utilized approaches
that focused on one specific problem that may be presented
in counseling. For example, in the Stoltenberg and McNeill
(1984) study conditions of high and low issue involvement
were designated by a median split on the Decisiveness scale
of the Career Maturity Inventory. The manipulations of
issue involvement used in the latter two studies have
resulted in mixed findings. As a result, all of the
authors cited above have called for better operational
definition and further research in regard to client
involvement.
The present study originally defined issue involvement
in a similar manner to the latter two studies cited above.
However, because the manipulation check did not yield
significant results, issue involvement was redefined in an
empirical manner based upon previous ELM research. Based
on a median split on scores of total number of relevant
thoughts, subjects were assigned to high and low issue
involvement conditions. According to the ELM, those two
groups by definition represent high and low levels of
elaboration likelihood within the sample. Yet significant
interactions effects between issue involvement and argument
quality did not appeal in the results. Once again, while
these two groups were high and low groups for the sample,

95
they probably did not represent relatively pure levels of
high and low issue involvement.
As noted earlier, these same criticisms may also be
applied to the ego-involvement variable. No significant
effects were found for ego-involvement for any dependent
measure. The absence of a manipulation check precludes the
conclusion that ego-involvement did not influence subjects'
cognitive responses to the counseling message. Therefore,
it is not surprising that the biasing effect predicted by
hypothesis three was not found. Hypothesis three predicted
that under conditions of high issue involvement, high ego-
involvement would exert a greater influence on the weak
than on the strong messages. Under low issue involvement,
it predicted that ego-involvement would exhibit an equal
influence on both strong and weak messages.
In summary, although the results of this study did not
support the hypotheses of either issue involvement or ego-
involvement or a combination of the two interacting with
argument quality to influence cognitive responses to a
counseling message, the results did not necessarily negate
those results due to possible methodological problems. The
results do suggest that the quality of the arguments used
in a counseling message have a significant impact upon
acceptance of the message with messages containing strong
arguments being more accepted than messages containing weak
arguments.

96
Need for Cognition
Individual differences in motivation to think have
relevant implications for message processing and
elaboration likelihood. Cacioppo and Petty (1982)
developed the need for cognition scale (NCS) in order to
distinguish people who dispositionally tend to engage in
and enjoy effortful analytic activity from those who do
not. Petty and Cacioppo (1986) further speculate that
individuals high in need for cognition should be more
likely to carefully evaluate and elaborate upon the issue
relevant arguments provided in persuasive messages.
However, additional correlational analysis conducted in the
present study did not support this notion. The correlation
between need for cognition score and total number of
relevant thoughts was nonsignificant; r=-.01, p=.88. This
finding does not support Petty and Cacioppo's notion of a
positive direct relationship between need for cognition and
elaboration likelihood.
Future Considerations
The investigation of the effects of issue involvement
and ego-involvement upon acceptance of counseling messages
is a relatively undeveloped area of inquiry. Only the four
studies cited earlier in this discussion have attempted to
manipulate issue involvement within research with a
counseling context. The present investigation is the first
to investigate the possible biasing effect of ego-

97
involvement under conditions of high issue involvement.
All of these studies appear to have methodological problems
in the manipulation of issue involvement. In the present
study the Simple Rathus Assertiveness Scale failed to
successfully differentiate subjects into high and low issue
involvement groups, and the theory-based measure appears
also not to have divided the sample into pure high and low
groups but into moderately high and moderately low groups.
As suggested earlier, however, these problems in
manipulating issue involvement probably arise from the
counseling context of the message and the ubiquitous
concerns that people have with interpersonal relations.
Future research may benefit from using alternative
approaches to manipulating issue involvement. For example,
crossing a measure of assertiveness level with a measure of
relevant thoughts generated may produce a high involvement
group, but it may be necessary to add a distraction
manipulation to the low involvement group in order to
obtain a relatively pure low level of elaboration
likelihood.
Similar steps may also be beneficial in refining the
ego-involvement manipulation. Within Personal Construct
methodology a number of ways exist to operationally define
superordinancy of a construct. Besides the amount of
relative variance accounted for by a single construct,
extremity ratings, resistance-to-change grids, and the

98
subject's own estimate of importance are all possible
methods of determining construct superordinancy. Since the
present ego-involvement manipulation produced no effects,
perhaps one of these alternative operationalizations should
be pursued in future research.
In addition to the above two changes in methodology,
future research may also benefit from investigating the
influence of various message factors upon acceptance of
counseling messages. Previous ELM research (Cacioppo &
Petty, 1979b) shows that repeating a persuasive
communication tends first to increase and then to decrease
argument. What effects might message repetition produce in
counseling and when should a counselor change messages?
Another interesting question is how should a counselor word
a message. Petty, Cacioppo, and Heesacker (1981) offer
evidence that when personal relevance is low, the use of
arguments in the form of rhetorical questions increases
elaboration, but when personal relevance is high, argument
scrutiny is reduced by the use of rhetorical questions.
The high personal relevance subjects reported that they
found the rhetorical questions to be distracting.
Further, future research may also benefit from more
refinement of behavioral measures appropriate for
counseling research. Eliciting behavior change as well as
attitude change is central to successful counseling. In
the present study the covert rehearsal technique (CRT) was

99
included to investigate the possibilities of it's use as a
behavioral measure of change related to counseling
messages. No hypotheses involving the CRT were made, and
no significant results were found. Development of
behavioral measures for counseling research is a complex
task due to the need to match the measure to presenting
problem and the need to determine what is appropriate
behavior changes.
One final point which must be addressed by future
research concerns the generalizability of experimental
studies (such as the present one) to actual counseling
settings. While the messages used in the present study
could be used in an actual counseling situation, the
differences between the two settings far outweigh the
similarities. Strong (1971) concludes that the
implications for counseling derived from laboratory
research are markedly limited due to the wide gap that
exists between the lab and the counseling room.
In sum, the ELM postulates several mediational
processes in which variables can have an impact upon
persuasion in general and acceptance of counseling messages
in particular. The importance of argument quality is a
solid finding in both basic persuasion ELM research and
counseling-related ELM research. The application of ELM
postulates concerning involvement and other variables which
may affect argument scrutiny is just beginning. While it

100
shows promise in providing a general framework from which
to understand counseling as a social influence process, the
results of the present study suggest that the application
of the ELM to the counseling setting may not be as fruitful
as basic ELM researchers believe. The nature of counseling
itself may place limitations on the ELM's usefulness in
understanding the social influence process as it occurs in
the therapy room.

APPENDIX A
RELATIONSHIPS GRID
instrument
in several
is designed to assess the ways you view
different types of relationships,
a list of the relationships found on the grid,
need to think of a person in your life who fits the
relationship described. Please use a different
each role described. In each case we are
you see yourself in the relationship.
This
yourself
Below is
You
person for
interested in how
1.Yourself in your relationship with your mother or
stepmother.
2. Yourself in your relationship with your current
roommate. If you don't have a roommate now, the
last person with whom you shared living quarters.
3. Yourself in your relationship with your father or
stepfather.
4.Yourself in your relationship with your current
romantic partner. If you currently don't have a
partner, think of your last boy/girlfriend.
5.Yourself in your relationship with the sibling
that you feel closest to. This may be either a
brother or sister. If you have no siblings think
of a person who has been most like a sibling to
you.
6.
Yourself in your relationship
boss. If you are not working
last boss. If you have never
someone who has had authority
with your current
now, think of your
worked, think of
over you.
7.Yourself in your relationship with your favorite
teacher or former teacher.
8.Yourself in your relationship with your closest
present friend of the same sex as yourself.
9.Yourself in your relationship with a person with
whom you usually feel most uncomfortable.
101

102
10. Yourself in your relationship with a former friend
of either sex whom you once thought was a close
friend but in whom you were later disappointed.
Each relationship is identified by a single word in the
columns along the top of the grid. They are presented from
left to right in the same order as above.
What you are to do is to think about your relationships
and rate them along the ten descriptors which appear along
the right side of the grid. You will use the rating scale
at the top of the descriptors. (+3 +2 +1 0 -1 -2 -3). For
example, in the first case you will be considering your
relationship with your mother along the descriptor, "calm
anxious." If you see yourself as very calm in your
relationship with her, place a +3 in the first box on the
grid. Use a +2 if you see yourself as fairly calm; use a
+1 if you see yourself as slightly calm with your mother.
A "0" rating is a neutral, whereas -1, -2, and -3 indicate
that you see yourself as correspondingly more anxious in
the relationship. When you have finished this rating move
down to the next descriptor and indicate how "respectful--
not respectful" you consider yourself to be with your
mother. Continue in this way until you have rated your
relationship with your mother along all ten descriptors.
Then move over to the next column and consider your
relationship with your roommate and rate yourself in that
relationship along each of the ten descriptors. Continue
your ratings in this way until you have rated each
relationship along all of the descriptors and every box in
the grid is filled in.
Please turn this page over and begin your ratings.

RELATIONSHIPS GRID
(D
lI
X
r0
-P
TJ
0)
P
G
-p
p
Cr
P
o
p
(0
p
(u
G
0)
t
fat
H
QJ

c
H
X
C
g
P
X
fc
x
p
i1
tn

CL)
0
M-t
X
0
-p
p
X!
Cfl
to
H
o
1
0
0
f0
(0
rl
0
(1)
P
G
X
S
fa
fa
fa
fa
CQ
H
fa
fa
w
Calm Anxious
Respectful Not respectful
Assertive Nonassertive
Dependent Independent
Does not
Good listener listen
Lacking in
Understanding understanding
Fair Unfair
Kind, helpful Hurtful
Open Closed
Satisfied Dissatisfied
103

APPENDIX B
PERSONAL PROBLEM INVENTORY
WHAT TO DO: Read each sentence carefully. Write down on
each line whatever number is correct for you.
5 very much like me
4 somewhat like me
3 neutral
2 somewhat unlike me
1 very unlike me
1. Most people stand up for themselves more than
I do.
2. At times I have not made or gone on dates
because of my shyness.
3. When I am eating out and the food I am served
is not cooked the way I like it, I complain
to the person serving it.
4. I am careful not to hurt other people's
feelings, even when I feel hurt.
5. If a person serving in a store has gone to a
lot of trouble to show me something which I
do not really like, I have a hard time saying
"no."
6. When I am asked to do something, I always
want to know why.
7. There are times when I look for a good strong
argument.
8. I try as hard to get ahead in life as most
people like me do.
9. To be honest, people often get the better of
me.
10. I enjoy meeting and talking with people for
the first time.
104

105
11. I often don't know what to say to good
looking people of the opposite sex.
12. I do not like making phone calls to
businesses or companies.
13. I would rather apply for jobs by writing
letters than by going to talk to the people.
14. I feel silly if I return things I don't like
to the store that I bought them from.
15. If a close relative that I liked was
upsetting me, I would hide my feelings rather
than say that I was upset.
16. I have sometimes not asked questions for fear
of sounding stupid.
17. During an argument I am sometimes afraid that
I will get so upset that I will shake all
over.
18. If a famous person were talking in a crowd
and I thought he or she was wrong, I would
get up and say what I thought.
19. I don't argue over prices with people selling
things.
20. When I do something important or good, I try
to let others know about it.
21. I am open and honest about my feelings.
22. If someone has been telling false and bad
stories about me, I see him (her) as soon as
possible to "have a talk" about it.
23. I often have a hard time saying "No."
24. I tend not to show my feelings rather than
upsetting others.
25. I complain about poor service when I am
eating out or in other places.
26. When someone says I have done very well, I
sometimes just don't know what to say.

106
27. If a couple near me in the theatre were
talking rather loudly, I would ask them to
be quiet or to go somewhere else and talk.
28. Anyone trying to push ahead of me in a line
is in for a good battle.
29. I am quick to say what I think.
30. There are times when I just can't say
anything.

APPENDIX C
ASSERTION SURVEY
Please circle the number that best represents your answer
to each question below:
1. How effective do you think this technique would be in
helping most people to become more assertive?
(Not at all effective) 1234567 (very effective)
2. To what extent do you think the cognitive restructuring
technique would be beneficial to you, personally?
(Not at all beneficial) 1234567 (Very beneficial)
3. How likely are you to use the cognitive restructuring
technique in the future in situations requiring
assertion?
(Not at all likely) 1234567 (Very likely)
4. Please rate the following aspects of the tape:
a. Voice Quality
Good 1234567 Bad
b. Rate of Delivery
Too Fast 1234567 Too Slow
c. Enthusiasm for Topic
Good 1234567 Bad
5. To what extent did the tape hold your attention?
Not at all 1234567 Very much
6. To what extent did the taped message have implications
for you personally?
Not at all relevant to me 1234567 Extremely relevant
to me
107

108
7. How would you rate the quality of the arguments used by
the speaker in support of the advocated topic?
Very poor arguments 1234567 Very good arguments
8. How likely are you to pursue counseling within the next
three months
Not at all likely 1234567 Very likely

APPENDIX D
THOUGHT LISTING TECHNIQUE
Thoughts Regarding The Message
We would now like to get an idea of the thoughts that
crossed your mind while listening to the tape. The next
page contains the form we have prepared for you to use to
record your thoughts and ideas. Simply write down the
first idea that comes to mind on the first line, the second
idea on the second line, etc. Please put only one idea or
thought on a line. You should try to record those thoughts
and ideas as concisely as possiblea phrase is sufficient.
Ignore spelling, grammar, and punctuation.
You will have three minutes to write down your
thoughts. We have deliberately provided more space than we
think most people will need to insure that everyone would
have plenty of room to write his/her ideas and thoughts.
So don't worry if you don't fill every space. Just write
down whatever your thoughts were while listening to the
tape. Please be completely honest and list all of the
thoughts that you hadputting only one thought per space.
Do not turn the page until instructed to do so.
109

110
Please list the thoughts you had while listening to the
tape, listing one thought per line.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18 .
19.
20.
21.
22.
23.
24

Ill
Do not turn the page until instructed to do so.
Now, please go back to the thoughts you have just
listed on the previous page. In the space at the end of
each line, rate each thought as follows:
If the thought was in favor of the idea of using
cognitive restructuring, mark the thought with a plus (+).
If the thought was opposed to the idea of using
cognitive restructuring, mark the thought with a minus (-).
If the thought was neither in favor of using cognitive
restructuring nor opposed to it, mark the thought with a
zero (0) .
Please do not turn page until instructed to do so.

APPENDIX E
COVERT REHEARSAL TECHNIQUE
As a final part of this study, we would like for you to
take a minute to imagine a situation that required you to
talk in front of a large class. Imagine that you are
giving a ten minute presentation on the thing that you
value most in life. What we would like for you to do is
to take the next thirty seconds to conjure up the clearest
possible image that you can of that situation. Imagine a
picture of yourself, the room, the other students, your
professor, etc. Do this now.
Now that you have the situation in mind we would like
to know how clear it is for you. People often differ in
how sharp their mental pictures are, and we would like for
you to rate how clear your mental image is on the scale
below. Do this now.
1.
Very fuzzy and unclear
2.
Somewhat fuzzy
3.
Midpoint: not real clear but not
real
fuzzy
4.
Fairly clear
5.
Very sharp and clear.
Now
that you have indicated how clear
this
image
you, we would like to have you conjure up the image again.
This time concentrate on the things that you are saying to
yourself or thinking in the situation. In other words, as
you imagine yourself in this situation, what kind of
thoughts run through your head? During the next three
minutes we would like for you to write down every thought
that runs through your mind in this situation. As you did
before ignore spelling, grammar, and punctuation. Just
write down one thought per line on the following page.
Please turn the page now and begin.
112

113
Please list the thoughts that run through your mind as you
imagine yourself giving the class presentation. Please
list one thought per line.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24 .

APPENDIX F
COGNITIVE ASSESSMENT QUESTIONNAIRE
Please read the statements below and indicate how
characteristic each statement is of you, using the
following rating scale:
1. extremely uncharacteristic of me
2. fairly uncharacteristic of me
3.
neutral, neither characteristic nor uncharacteristic
of me
4.
5.
fairly characteristic of me
extremely characteristic of me
1.
, I would prefer complex to simple problems.
2. I like to have the responsibility of handling a
situation that required a lot of thinking.
3.
. Thinking is not my idea of fun.
4.
. I would rather do something that required little
thought than something that is sure to challenge
my thinking abilities.
5.
, I try to anticipate and avoid situations where
there is likely some chance I will have to think
in depth about something.
6.
, I find satisfaction in deliberating hard and for
long hours.
7.
, I only think as hard as I have to.
8.
I prefer to think about small, daily projects to
long-term ones.
9.
I like tasks that require little thought once I've
learned them.
10. The idea of relying on thought to make my way to
the top appeals to me.
114

115
11. I really enjoy a task that involves coining up with
new solutions to problems.
12. Learning new ways to think doesn't excite me very
much.
13. I prefer my life to be filled with puzzles that I
must solve.
14. The notion of thinking abstractly is appealing to
me.
15. I would prefer a task that is intellectual,
difficult, and important to one that is somewhat
important but does not require much thought.
16. I feel relief rather than satisfaction after
completing a task that required a lot of mental
effort.
17. It's enough for me that something gets the job
done; I don't care how or why it works.
18. I usually end up deliberating about issues even
when they do not effect me personally.

APPENDIX G
TRANSCRIPT OF TAPED INSTRUCTIONS
Cognitive restructuring is a therapeutic technique
which is based on a simple observation. The things we
think about affect the way we feel. If you were asked to
make yourself feel anxious, for example, you could probably
do this by thinking about an upcoming exam, picturing
yourself giving a talk in front of a large class, or
imagining yourself asking someone new out for a date. The
cognitive restructuring technique relies on the fact that
just as you can make yourself feel upset, you can also make
yourself feel better by controlling the things you think
about and tell yourself. The basic procedure consists of
three steps. First, identifying the things you say to
yourself. Second, assessing how realistic those things
are. And third, changing your unrealistic thoughts to more
realistic or reasonable ones. For example, if you are
thinking of asking someone out for a date, and you are
telling yourself. "He or she will never go out with me.
They will just laugh at me." You could ask yourself how
realistic this thought really is, and then change it to
something more probable like, "They may or may not go out
116

117
with me, but I won't know until I ask. All they can do is
say no, and I can deal with that." Although this technique
can be used in a variety of disorders, one common
application is in the treatment of inadequate
assertiveness. Problems in personal assertion can be quite
common in college populations. Personal assertion refers
to self-expression through which one stands up for his or
her basic rights without violating the basic human rights
of others. Inadequate assertiveness may involve either
passive or aggressive behavior in place of assertive
behavior. With passive behavior one sacrifices one's own
basic rights and needs and desires through limited self-
expression. Passive people often lack friends, miss career
opportunities, and have low self-esteem. On the other
hand, people may get what they need or want, but they use
aggressive behavior to get it and thus hurt others in the
process. They often are not trusted by others, have
limited close relationships, and are often confronted with
anger from other people. Cognitive restructuring can help
people with problems in personal assertion because it
involves identifying dysfunctional thoughts that people
tell themselves and changing these thoughts to more
constructive, healthy messages. If you would like to know
more about this technique please listen to the following
message concerning its effectiveness in treating problems
in personal assertion.

118
(Depending upon the subject's assignment to either
strong or weak argument quality, he or she will next hear
the appropriate arguments for the use of cognitive
restructuring. In this transcript strong arguments are
presented next and then weak arguments.)
Strong argument quality intervention
First, cognitive restructuring works. Controlled,
laboratory research published in the journal, Cognitive
Psychotherapy and Research, Dore & Caplan (1982) has shown
that the use of the cognitive restructuring technique has
enabled people with problems in personal assertion to
become more assertive 74 percent of the time and to become
more consistent in their self-expression. Moreover,
research has demonstrated significant reductions in the
anxiety, negative feelings, and poor self-images associated
with this problem. It also helped alleviate the tension
and the stress that often accompanies problems in personal
assertion.
Second, cognitive restructuring enables you to analyze
your problems in asserting yourself giving you greater
understanding of yourself. You can look at yourself more
objectively and see yourself as others see you. You also
become more aware of automatic thoughts and irrational
beliefs and the impact that they have had on the way you
f eel.

119
Third, because it reduces the stress associated with
problems in personal assertion, cognitive restructuring
provides many additional health benefits. For example, the
use of cognitive restructuring has been associated with
lower blood pressure, higher levels of energy, and general
nutritional improvement.
Finally, this technique can be used in the treatment of
other problems as well. It is effective in helping you
gain control over many areas of your life in which you
might feel anxious or depressed. After using this
technique people typically report feeling better about
themselves, more confident, relaxed and self-assured.
In summary, cognitive restructuring works. Research
indicates its effectiveness. It also provides greater
self-awareness and understanding. It is associated with
general health benefits and is useful in many other aspects
of your life as well.
Weak argument quality intervention
Cognitive restructuring is quick and easy to learn.
Most people can master the technique without difficulty and
can begin to use it soon after they learn it. So for this
reason cognitive restructuring is a very simple technique
to use for treating problems in personal assertion.
Because it doesn't require a great deal of effort, it's
worth a try. Even if it isn't especially beneficial, at
least it isn't harmful.

120
Secondly, I've tried cognitive restructuring once, and
even though it didn't help me with my inadequate personal
assertion that doesn't mean it wouldn't work more
effectively for you. Most techniques work better for some
people than for others, and you may be one that it works
for. I think that everyone should at least try it and see
if it works for them.
Thirdly, although I haven't been able to interview
everyone that has used this technique, I personally know of
several people who did use it and thought it was okay.
They seemed to think that it was better than some of the
other methods they had used to treat their assertion
problems.
Finally, most people waste a lot of their free time.
These are times often spent watching television, sleeping,
or in a similarly unproductive activity. At least
cognitive restructuring is constructive. It is probably
more useful and interesting than what you would ordinarily
be doing with your free time. So it makes sense to try the
technique.
In summary, cognitive restructuring is a quick and easy
technique for treating problems in personal assertion.
I've tried it, and though it didn't work for me it may work
for you. Additionally, I've talked to others that have
tried it, and they thought that it worked better than some
other techniques they had tried. Finally, at least it's a

121
constructive way to spend your free time and gives you
something interesting and useful to do.
This completes the message portion of our tape. We are
now interested in feedback about that message. Please turn
the page and begin answering the Assertion Survey leaving
the headphones on for further instructions.
(Silence lasting 2 minutes)
Now that you have completed your responses please turn
to the next page and read the instructions with me.
We would like to get an idea of the thoughts that
crossed your mind while listening to the tape. The next
page contains the form we have prepared for you to use to
record your thoughts and ideas. Simply write down the
first idea that comes to mind on the first line, the second
idea on the second line, etc. You should try to record
those thoughts and ideas as concisely as possible. A
phrase is sufficient. Ignore spelling, grammar, and
punctuation.
You will have three minutes to write down your
thoughts. We have deliberately provided more space than we
think most people will need to insure that everyone would
have plenty of room to write his/her ideas and thoughts.
So don't worry if you don't fill every space. Just write
down whatever your thoughts were while listening to the
tape. Please be completely honest and list all of the

122
thoughts that you had putting only one thought per space.
Please turn the page now and begin.
(Silence lasting three minutes)
Please stop and turn to the next page and follow the
instructions at the top of the page. You will have two
minutes to complete those instructions. Then please listen
for further instructions.
(Silence lasting two minutes)
Please stop and turn to the next page and follow the
instructions at the top of the page. You will have five
minutes to complete those instructions. Then please listen
for further instructions.
(Silence lasting five minutes)
Now that you have completed your ratings please turn to
the next page and read the instructions with me.
As a final part of this study, we would like for you to
take a minute to imagine a situation that requires that you
take a risk and express yourself. The situation that we
would like for you to imagine is that of you giving a ten
minute oral class presentation on the thing that you value
most in life. What we would like for you to do is to take
the next thirty seconds to conjure up the clearest possible
image that you can of that situation. Imagine a picture of
yourself, the room, the other students, your professor,
etc. Do this now.

123
(Silence lasting thirty seconds)
Now that you have the situation in mind we would like
to know how clear it is for you. People often differ in
how sharp their mental pictures are, and we would like for
you to rate how clear your mental image is on the scale
below. Do this now.
(Silence lasting twenty seconds)
Now that you have indicated how clear the image is for
you, we would like to have you conjure up the image again.
This time concentrate on the things that you are saying to
yourself or thinking in the situation. In other words, as
you imagine yourself in this situation, what kind of
thoughts run through your head? During the next three
minutes we would like for you to write down every thought
that runs through your mind in this situation. As you did
before ignore spelling, grammar, and punctuation. Just
write down one thought per line on the following page.
Please turn the page now and begin.
(Silence lasting three minutes)
You have now finished the study. Thank you. Please
turn your materials in to the researcher to receive your
credit, and thank you again.

REFERENCES
Apsler, R., & Sears, D. 0. (1968). Warning, personal
involvement, and attitude change. Journal of
Personality and Social Psychology, 9_, 162-166,
Bannister, D. (1960). Conceptual structure in thought
disordered schizophrenics. Journal of Mental science,
108, 825.
Bannister, D. (1962). The nature and measurement of
schizophrenic thought disorder. Journal of Mental
Science, 108, 825-842.
Bannister, D. (1965). The rationale and clinical relevance
of repertory grid technique. British Journal of
Psychiatry, 111, 977-982.
Bannister, D. & Fransella, F. (1966). A grid test of
schizophrenic thought disorder. British Journal of
Social and Clinical Psychology, 5^, 95.
Bannister, D., & Mair, M. (1968). The Evaluation of
Personal Constructs. New York: Academic Press.
Blanchard, E. B., Turner, J., Eschette, N., & Coury,
V. M. (1977). Assertiveness training for dental
students. Journal of Dental Education, 41, 206-208.
Brehm, J. W. (1972). Responses to loss of freedom: A
theory of psychological reactance. Morristown, NJ:
General Learning Press.
Brock, T. C. (1967). Communication discrepancy and intent
to persuade as determinants of counterargument
production. Journal of Experimental Social
Psychology, 3_, 296-309.
Cacioppo, J. T., & Petty, R. E. (1979). Attitudes and
cognitive response: An electrophysiological approach.
Journal of Personality and Social Psychology, 37, 2181-
2199.
124

125
Cacioppo, J. T., & Petty, R. E. (1981). Social
psychological procedures for cognitive response
assessment: The thought listing technique. In T.
Merluzzi C. Glass, & M. Genest (Eds.), Cognitive
assessment. New York: Guilford.
Cacioppo, J. T., & Petty, R. E. (1982). The need for
cognition. Journal of Personality and Social
Psychology, 42, 116-131.
Cacioppo, J. T., & Petty, R. E. (1984). The need for
cognition: Relationship to attitudinal processes. In
R. P. McGlynn, J. E. Maddux, C. D. Stoltenberg, & J. H.
Harvey (Eds.), Social perception in clinical and
counseling psychology (pp. 113-139). Lubbock, TX:
Texas Tech Press.
Cacioppo, J. T., & Petty, R. E. (1985). Central and
peripheral routes to persuasion: The role of message
repetition. In L. Alwitt & A. Mitchell (Eds.),
Psychological processes and advertising effects:
Theory, research, and applications (pp. 91-111).
Hillsdale, NJ: Erlbaum.
Cacioppo, J. T., Petty, R. E., & Kao, C. (1984). The
efficient assessment of need for cognition. Journal of
Personality Assessment, 48, 306-307.
Cacioppo, J. T., Petty, R. E., & Morris K. (1983). Effects
of need for cognition of message evaluation, recall,
and persuasion. Journal of Personality and Social
Psychology, 45, 805-818.
Cacioppo, J. T., Petty, R. E., & Sidera, J. A. (1982). The
effects of a salient self-schema on the evaluation of
proattitudinal editorials: Top-down versus bottom-up
message processing. Journal of Experimental Social
Psychology, 18, 324-338.
Cacioppo, F. T., Petty, R. E., & Stoltenberg, C. D. (1985).
Processes of social influence: The elaboration
likelihood model of persuasion. in P. Kendall (Ed.),
Advances in Cognitive-Behavioral Research and Therapy,
(Vol. pp. 215-274) New York: Academic Press .
Carlson, E. R. (1956). Attitude change through
modification of attitude structure. Journal of
Abnormal and Social Psychology, 52, 256-261.

126
Chaiken, S. (1980). Heuristic versus systematic
information processing and the use of source versus
message cues in persuasion. Journal of Personality and
Social Psychology, 39, 752-766.
Cohen, A., Stotland, E., & Wolfe, D. (1955). An
experimental investigation of need for cognition.
Journal of Abnormal and Social Psychology, 51, 291-294.
Corrigan, J. D., Dell, D. M., Lewis, K. N., & Schmidt, L.
D. (1980). Counseling as a social influence process:
A review [Monograph]. Journal of Counseling
Psychology, 27, 395-441"!
Crocker, J., Fiske, S. T. & Taylor, S. E. (1984).
Schematic bases of belief change. In J.R. Eiser (Ed.).
Attitudinal judgment (pp. 197-226). New York:
Springer-Ver lag.
Cullen, D. M. (1969). Attitude measurement by cognitive
sampling. (Doctoral dissertation, The Ohio State
University, 1968). Dissertation Abstracts, 29, 1597A.
Dean, R., Austin, J., & Watts, W. (1971). Forewarning
effects in persuasion: Field and classroom
experiments. Journal of Personality and Social
Psychology, 18, 210-221.
Dixon, D. N., & Claiborn, C. D. (1981). Effects of need
and commitment on career exploration behaviors.
Journal of Counseling Psychology, 28, 411-415.
Eagly, A.H. (1967). Involvement as a determinant of
response to favorable and unfavorable information.
Journal of Personality and Social Psychology, 7, (3,
pt. 2), 1-15.
Eagly, A. H., & Mannis, M. (1966). Evaluation of message
and communication as a function of involvement.
Journal of Personality and Social Psychology, 3, 482-
485.
Ellis, A. (1985). Overcoming resistance: Rational-
emotive therapy with difficult clients. New York:
Springer.
Fiske, S. T., & Taylor, S. E. (1984). Social cognition.
Reading, MA: Addison-Wesley.
Frank, J. D. (1961). Persuasion and healing. Baltimore:
John Hopkins.

127
Frankel, P. B. (1977). A factor analytic study of measures
of assertiveness. (Doctoral dissertation, California
School of Professional Psychology, 1976) Dissertation
Abstracts International, 37, 4676B-4677B.
Freedman, J. D. (1964). Involvement, discrepancy, and
change. Journal of Abnormal and Social Psychology, 69,
290-295.
Freud, S. (1939). The interpretation of dreams. In A. A.
Brill (Ed. and Trans.), The Basic Writings of Sigmund
Freud. (pp.179-549). New York: Random House.
(Original work published 1900).
Freud, S. (1959). The defence neuro-psychoses. In E.
Jones (Ed.) and J. Richman (Trans.), Sigmund Freud:
Collected Pages (Vol. 1, pp. 59-75). New York:
Basic Books. (Original work published 1894).
Futch, E. J., Scheirer, C. J., & Lisman, S. A. (1982).
Factor analyzing a scale of assertiveness: A critique
and demonstration. Behavior Modification, 6_, 23-43.
Greenwald, A. G. (1968). Cognitive learning, cognitive
responses to persuasion, and attitude change. In A.
G. Greenwald, T.C. Brock, & T.M. Ostrom (Eds.),
Psychological Foundations of Attitudes, (pp. 147-170).
New York: Academic Press.
Greenwald, A. G. (1980). The totalitarian ego:
Fabrication and revision of personal history. American
Psychologist, 35, 603-618.
Greenwald, A. G., & Albert, R. D. (1968). Acceptance and
recall of improvised arguments. Journal of Personality
and Social Psychology, 8_, 31-34.
Hass, R. G., & Grady, K. (1975). Temporal delay, type of
forewarning, and resistance to influence. Journal of
Experimental Social Psychology, 11, 459-469.
Heesacker, M. (1986). Counseling pretreatment and the
elaboration likelihood model of attitude change.
Journal of Counseling Psychology, 33, 107-114.
Heimberg, R. G., & Harrison, D. F. (1980). Use of the
Rathus Assertiveness Schedule with offenders: A
question of questions. Behavior Therapy, 11, 278-281.

128
Heppner, P. P., & Dixon, D. N. (1978). The effects of
client perceived need and counselor role on client's
behavior. Journal of Counseling Psychology, 25, 514-
519.
Heppner, P. P., & Dixon, D. N. (1981). A review of the
interpersonal influence process in counseling.
Personnel and Guidance Journal, 59, 542-550.
Heppner, P., & Heesacker, M. (1982). Interpersonal
influence process in real-life counseling:
Investigating client perceptions, counselor experience
level, and counselor power over time. Journal of
Counseling Psychology, 29, 215-223.
Hinkle, D. (1966). The change of personal constructs from
the viewpoint of a theory of construct implications.
(Doctoral dissertation, The Ohio State University,
1965). Dissertation Abstracts, 26, 6837.
Hollandsworth, J. G. (1976). Further investigation of the
relationship between expressed social fear and
assertiveness. Behavior Research and Therapy, 14, 85-
87.
Honess, T. (1977). A comparison of the implication and
repretory grid techniques. Unpublished manuscript,
University of Exeter, England.
Hovland, C. I. (1951). Changes in attitude through
communication. Journal of Abnormal and Social
Psychology, 46, 424-437.
Hovland, C. I., Harvey, 0. J., & Sherif, M. (1957).
Assimilation and contrast effects in reactions to
communication and attitude change. Journal of Abnormal
and Social Psychology, 55, 244-252.
Janis, I. L., & King, B. (1954). The influence of role-
playing on opinion change. Journal of Abnormal and
Social Psychology, 49, 211-218.
Jones, E. E. (1985). Major developments in social
psychology during the past five decades. In G. Lindzey
& E. Aronson (Eds.), Handbook of Social Psychology:
Vol. I. Theory and Method (pp. 47-107). New York:
Random House.
Kaplan, D. A. (1982). Behavioral, cognitive, and
behavioral-cognitive approaches to group assertion
training therapy. Cognitive Therapy and Research, 6,
301-314.

129
Kelly, G. A. (1955) The psychology of personal
constructs. New York: Norton.
Kiesler, C. A., Collins, B. E., & Miller, N. (1969).
Attitude change: A critical analysis of theoretical
approaches. New York: Wiley.
King, B., & Janis, I. L. (1956). Comparison of the
effectiveness of improvised versus non-improvised role-
playing in producing opinion changes. Human Relations,
9, 177-186.
Lammers, H. B., & Becker, L. A. (1980). Distraction:
Effects on the perceived extremity of a communication
on cognitive responses. Personality and Social
Psychology Bulletin, 6_, 261-266.
Landman, J., & Manis, M. (1983). Social cognition: Some
historical and theoretical perspectives. In L.
Berkowitz (Ed.), Advances in Experimental Social
Psychology (Vol. 16, pp. 49-123). New York: Academic
Press.
Latane, B., & Darley, J. M. (1970). The unresponsive
bystander: Why doesn't he help? New York: Appleton-
Century-Crofts.
Latane, B., Williams, K., & Harkins, S. G. (1979). Many
hands make light the work: The causes and consequences
of social loafing. Journal of Personality and Social
Psychology, 37, 822-832.
Lewin, K. (1947). Group decision and social change. In T.
Newcomb & E. Hartley (Eds.), Readings in social
psychology. New York: Holt, Rinehart & Winston.
Lord, C. G., Ross L., & Lepper, M. R. (1979). Biased
assimilation and attitude polarization: The effects of
prior theories on subsequently considered evidence.
Journal of Personality and Social Psychology, 37, 2098-
2109.
MacDonald, M. L. (1975). A behavioral assessment
methodology applied to the measurement of assertion.
(Doctoral dissertation, University of Illinios, 1974.
Dissertation Abstracts International, 35, 6101B.
McCormick, I. A. (1984). A simple version of the Rathus
Assertiveness Schedule. Behavioral Assessment, 7, 95-
99.

130
McGuire, W. J. (1964). Inducing resistance to persuasion:
Some contemporary approaches. In L. Berkowitz (Ed.),
Advances in Experimental Social Psychology (Vol. 1, pp.
191-229). New York: Academic Press.
McGuire, W., & Papageorgis, D. (1961). The relative
efficacy of various types of prior belief defense in
producing immunity against persuasion. Journal of
Abnormal and Social Psychology, 62, 327-337.
Mann, R. J., & Flowers, J. V. (1978). An investigation of
the reliability and validity of the Rathus
Assertiveness Schedule. Psychological Reports, 42,
632-634.
Markus, H., & Sentis, K. (1982). The self in social
information processing. In J. Suls (Ed.). Social
psychological perspectives on the self (pp. 41-70).
Hillsdale, NJ: Erlbaum.
Markus, J., & Zajonc, R. B. (1985). The cognitive
perspective in social psychology. In G. Lindsey & E.
Aronson (Eds.), Handbook of social psychology:
Vol. I. Theory and method (pp. 137-230). New York:
Random House.
Miller, N. (1965). Involvement and dogmatism as inhibitors
of attitude change. Journal of Experimental Social
Psychology, .1, 121-132.
Morgan, W. G. (1974). The relationship between expressed
social fears and assertiveness and its treatment
implications. Behavior Research and Therapy, 12, 155-
257.
Orenstein, H., Orenstein, E., & Carr, J. (1975).
Assertiveness and anxiety: A correlational study.
Journal of Behavior Therapy and Experimental
Psychiatry, 6, 203-207.
Ostrom, T. M., & Brock, T. C. (1968). A cognitive model of
attitudinal involvement. In R. Abelson, E. Aronson, W.
J. McGuire, T. M. Newcomb, J. j. Rosenberg, & P. H.
Tannenbaum (Eds.), Theories of cognitive consistency:
A sourcebook. Chicago: Rand McNally.
Petty, R. E., & Cacioppo, J. T. (1977). Forewarning,
cognitive responding, and resistance to persuasion.
Journal of Personality and Social Psychology, 35, 645-
65in ~

131
Petty, R. E., & Cacioppo, J. T. (1979a). Effects of
forewarning of persuasive intent and involvement on
cognitive responses and persuasion. Personality and
Social Psychology Bulletin, 5^ 173-176.
Petty, R. E. & Cacioppo, J. T. (1979b). Issue involvement
can increase or decrease persuasion by enhancing
message relevant cognitive responses. Journal of
Personality and Social Psychology, 37, 1915-1926.
Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and
persuasion: Classic and contemporary approaches.
Dubuque, IAl William C. Brown.
Petty, R. E., & Cacicoppo, J. T. (1984). The effects of
involvement on responses to argument quantity and
quality: Central and peripheral routes to persuasion.
Journal of Personality and Social Psychology, 46, 69-
81.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration
likelihood model of persuasion. In L. Berkiwitz (Ed.),
Advances in Experimental Social Psychology (Vol. 19,
pp. 123-205). Orlando, FL: Academic Press.
Petty, R. E., Cacioppo, J. T. & Goldman, R. (1981).
Personal involvement as a determinant of argument-based
persuasion. Journal of Personality and Social
Psychology, 41, 847-855.
Petty, R. E., Cacioppo, J. T., & Heesacker, M. (1981). The
use of rhetorical questions in persuasion: A cognitive
response analysis. Journal of Personality and Social
Psychology, 40, 432-440.
Petty, R. E., Harkins, S. G., & Williams, K. 0. (1980).
The effects of group diffusion of cognitive effort on
attitudes: An information processing view. Journal of
Personality and Social Psychology, 38, 81-92.
Petty, R. E., Wells, G. L., & Brock, T. C. (1976).
Distraction can enhance or reduce yielding to
propaganda: Thought disruption versus effort
justification. Journal of Personality and Social
Psychology, 34, 874-884.
Quillan, J., Besing, S., & Dinning, D. (1977).
Standardization of the Rathus Assertiveness Schedule.
Journal of Clinical Psychology, 33, 418-422.
Rathus, S. A. (1973). A 30-item schedule for assessing
assertive behavior. Behavior Therapy, 4, 398-406.

132
Rhine, R.J. & Severance, L. J. (1970). Ego-involvement,
discrepancy, source credibility, and attitude change.
Journal of Personality and Social Psychology, 16, 17 5-
190.
Ross, L., Lepper, M. R., & Hubbard, M. (1975).
Perseverance in self-perception and social perception:
Biased attributional processes in the debriefing
paradigm. Journal of Personality and Social
Psychology, 32, 880-892.
Ryle, A. (1979a). The focus in brief interpretive
psychotherapy: Dilemmas, traps and snaps as target
problems. British Journal of Psychiatry, 134, 46-54.
Ryle, A. (1979b). Defining goals and assessing change in
brief psychotherapy: A pilot study using target
ratings and the dyad grid. British Journal of Medical
Psychology, 52, 223-233.
Ryle, A. (1980). Some measures of goal attainment in
focussed integrated active psychotherapy: A study of
fifteen cases. British Journal of Psychiatry, 137,
475-486.
Ryle, A. (1981). Dyad grid dilemmas in patient and control
subjects. British Journal and Medical Psychology, 54,
353-358.
Ryle, A. (1985). The dyad grid and psychotherapy research.
In N. Beail (Ed.), Repertory grid technique and
personal constructs: Applications in clinical and
educational settings (pp~I 190-206) Cambridge, MA:
Bookline Books.
Ryle, A., & Breen, D. (1972a). The use of the double dyad
in the clinical setting. British Journal of Medical
Psychology, 45, 383-389.
Ryle, A., & Breen D. (1972b). A comparison of adjusted
and maladjusted couples using the double dyad grid.
British Journal of Medial Psychology, 45, 375-382.
Ryle, A., & Lipshitz, S. (1975). Recording change in
marital therapy with the reconstruction grid. British
Journal of Medical Psychology, 48, 39-48.
Ryle, A., & Lipshitz, S. (1976). Repretory grid
elucidation of a difficult conjoint therapy. British
Journal of Medical Psychology, 49, 281-285.

133
Ryle, A., & Lunghi, M. (1970). The dyad grid: A
modification of repertory grid technique. British
Journal of Psychiatry, 117, 323-327.
Ryle, A., & Lunghi, M. (1971). A therapist's prediction of
a patient's dyad grid. British Journal of Psychiatry,
118, 555-560.
Sherif, M. & Hovland, C.I. (1961). Social judgment:
Assimilation and contrast effects in communication and
attitude change. New Haven, CT: Yale University
Press.
Sherif, C. W., Kelly, M., Rodgers, H. L., Sarup, G., &
Tittler, B. (1973). Personal involvement, social
judgment, and action. Journal of Personality and
Social Psychology. 27 311-3 27 .
Sherif, C. W., Sherif, M., & Nebergall, R. E. (1965).
Attitude and attitude change: The social judgment-
involvement approach-! Philadelphia: Saunders.
Stoltenberg, C. C., & McNeil, B. W. (1984). Effects of
expertise and issue involvement on perceptions of
counseling. Journal of Social and Clinical Psychology,
2, 314-325.
Stotland, E., Katz, D., & Patchen, M. (1959). The
reduction of prejudice through the arousal of self
insight. Journal of Personality, 27, 507-531.
Strong, S. R. (1968). Counseling: An interpersonal
influence process. Journal of Counseling Psychology,
15, 215-224.
Strong, S. R. (1971). Experimental laboratory research in
counseling. Journal of Counseling Psychology, 18, 106-
110.
Tally, J. E., Barrow, J. C., Fulkerson, K. F., & Moore, C.
A. (1983). A comparison of telephone vs. mail
strategies in conducting a university psychological
services needs assessment. Journal of American College
Health, 32, 101-103.
Vaal, J. j. (1975). The Rathus Assertiveness Schedule:
Reliability at the junior high school level. Behavior
Therapy, £, 566-567.
Wolpe, J. (1973). Supervision transcript: V. Mainly
about assertive training. Journal of Behavior Therapy
and Experimental Psychiatry" T, 141-148.

134
Wolpe, J., & Lang, P. J. (1964). A fear survey for use in
behavior therapy. Behavior Research and Therapy, 2,
27-30.
Wood, W. (1982). Retrieval of attitude-relevant
information from memory: Effects on susceptibility to
persuasion and on intrinsic motivation. Journal of
Personality and Social Psychology, 42, 798-810.
Wood, W., Kallgren, C., & Priesler, R. (1985). Access to
attitude relevant information in memory as a
determinant of persuasion. Journal of experimental
Social Psychology, 21, 73-85.
Zajonc, R. B. (1980). Feeling and thinking: Preferences
need no inferences. American Psychologist, 35, 151-
175.
Zimbardo, P. G. (1960). Involvement and communication
discrepancy as determinants of opinion conformity.
Journal of Abnormal and Social Psychology, 60, 86-94.

BIOGRAPHICAL SKETCH
Steven Hugh Williams was born on June 10 1957 in
Clarksburg, West Virginia. In 1975, he graduated from
Bridgeport Senior High School in Bridgeport, West Virginia.
He received his Bachelor of Arts degree in psychology in
1979 from West Virginia University in Morgantown, West
Virginia.
In August, 1981, Steven began graduate studies in
counseling psychology at the University of Florida. He
received his Master of Science degree in 1984. In August,
1986, Steven completed his predoctoral internship at the
Veterans Administration Medical Center in Bay Pines,
Florida. He expects to graduate with the Doctor of
Philosophy degree in May, 1987.
On May 7, 1983, Steven married Barbara Jessie Heier.
Barbara is currently employed as a pediatric intensive care
nurse at Shands Teaching Hospital at the University of
Florida. They have one son, Adam Benjamin.
135

I certify that I
opinion it conforms
presentation and is
as a dissertation for
have read this study and that in my
to acceptable standards of scholarly
fully adequate, in scope and quality,
the degrqe of Doctor of Philosophy.
V
£11
g J,
Associate
imeyer, Chairme
Professor of Psychology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Psychology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Mar^t Alicki
Assistant Professor of Psychology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Assistant Professor of Sociology
This dissertation was submitted to the Graduate Faculty of
the Department of Psychology in the College of Liberal Arts
and Sciences and to the Graduate School and was accepted as
partial fulfillment of the requirements for the degree of
Doctor of Philosophy.
May, 1987
Dean, Graduate School



113
Please list the thoughts that run through your mind as you
imagine yourself giving the class presentation. Please
list one thought per line.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24 .


91
implies that it may be advantageous for the therapist to
screen out distractions, to repeat important messages over
a number of sessions, and even to provide the client with a
written summary of central points made during therapy, all
manipulations that may also have the effect of enhancing
client involvement.
This effect of the message variable also has other
important implications for counseling. According to the
ELM, clients who carefully evaluate message content are
more likely to have enduring, behavior-related, central-
route attitude change than are those who attend only to
peripheral cues, such as counselor personal
characteristics. Message variables (such as argument
quality) may play crucial roles in eliciting behavior
change and in the maintenance of therapy gains. Future
research should investigate message factors to determine
those that are most helpful to counselors in bringing about
client change.
Results also show that issue involvement produced a
significant main effect for attitude rating such that
subjects who generated more thoughts about the message
believed that cognitive restructuring had more personal
benefits than did subjects who generated fewer thoughts.
Theoretically, an explanation of this finding could be
found in Zajonc's (1980) theory of mere exposure, which
describes changes in evaluation of objects as a result of


Ill
Do not turn the page until instructed to do so.
Now, please go back to the thoughts you have just
listed on the previous page. In the space at the end of
each line, rate each thought as follows:
If the thought was in favor of the idea of using
cognitive restructuring, mark the thought with a plus (+).
If the thought was opposed to the idea of using
cognitive restructuring, mark the thought with a minus (-).
If the thought was neither in favor of using cognitive
restructuring nor opposed to it, mark the thought with a
zero (0) .
Please do not turn page until instructed to do so.


issue involvement and ego-involvement. Under
conditions of high issue involvement, high
ego-involvement will exert a greater influence on
the weak than the strong messages. Under low
involvement, high ego-involvement will exert an
equal influence on both strong and weak messages.


80
process the issue-relevant arguments presented. One way in
which argument processing could be measured would be by the
number of relevant cognitions that a person generates. In
order to assign subjects to high and low levels of argument
processing (high and low issue involvement) a median split
was performed on scores of total number of relevant
thoughts generated by subjects on the thought listing
technique. The median score was deleted and subjects in
the upper half of the distribution constituted the high
issue involvement group, while subjects in the lower half
constituted the low issue involvement group. Analysis of
variance of the involvement manipulation check revealed
that high issue involvement subjects believed that the
message had significantly more implications for them than
did the low issue involvement group, F(1,96)=6.17, p<.01:
high issue involvement M=4.4, low issue involvement M=3.6.
The other variables (ego-involvement and argument quality)
did not significantly affect this manipulation check.
Dependent Measures
Data for each of the dependent measures was analyzed by
a 2(High and Low Issue Involvement) x2(High and Low Ego-
Involvement) x2(Strong and Weak Argument Quality) Analyses
of Variance. Analyses for each dependent measure are
presented separately.


57
average concern with assertion, subjects who scored in the
below the median range comprised the high issue involvement
condition.
The third factor, argument quality, consisted of two
levels, strong arguments and weak arguments. Strong and
weak arguments for the use and effectiveness of the
cognitive restructuring technique were developed by Greg
Neimeyer and his research team (Unpublished data, 1986).
Following the procedure outlined by Petty and Cacioppo
(1981) for developing arguments for a topic, they generated
strong and weak arguments for the use of cognitive
restructuring in the treatment of eating disorders. In the
present study these arguments have been adopted to the use
of cognitive restructuring in the treatment of
assertiveness.
Cognitive Responses to Arguments
The variables of interest in the present study
consisted of the subjects' cognitive responses to arguments
for cognitive restructuring. Cognitive responses were
operationalized in terms of cognitive favorability toward
cognitive restructuring, outcome expectation and behavioral
intention ratings, and judges' ratings of spontaneous use
of cognitive restructuring by subjects during a covert
rehearsal condition.
Cognitive favorability toward cognitive restructuring
(CFCR) was derived from the Thought-Listing Technique


50
This finding suggests that more involved clients may
see counseling as more beneficial and may attend more to
the counseling process. Thus, clients with a high level of
involvement should adhere more to central route cognitive
processing of therapeutic messages, whereas clients with a
low level of involvement may depend upon the peripheral
route. Moreover, as central route processing appears to
lead to more enduring attitude change (Petty and Cacioppo,
1981), it behooves the therapist to motivate clients to
evaluate carefully the pros and cons of their present
behaviors and to weigh the costs and benefits of change and
various means of change.
An example of a highly involved client perhaps would
be the articulate, well-educated, and young business person
seeking counseling concerning improving his or her
relationships with co-workers and superiors. One would
expect this person to consider carefully the issues and
suggestions in counseling. The low involved client may be
represented by the person forced to attend an alcohol
therapy group due to a DWI conviction. He or she may
rather be at home or even just spend the weekend in jail
than attend therapy weekly for 3 months. Such an
individual may elaborate very little on the issues raised
in counseling but may be influenced by persuasion cues
(e.g., counselor social attractiveness).


12
change. In another experiment King and Janis (1956) had
subjects either read a persuasive communication to
themselves, read it into a tape recorder, or read and then
give their own improvised version of the message. Those
subjects who improvised changed their attitudes
significantly more than subjects in either the oral or
silent reading conditions. These results are supportive of
the notion that one's own cognitive responses on an issue
are the most compelling. In a further test of this
hypothesis Greenwald and Albert (1968) had subjects
improvise five arguments in response to instructions to
advocate either career preparatory or general liberal arts
undergraduate education. Subjects also read a set of
provided arguments supporting the opposite side. Results
indicated that subjects' attitudes tended to be in the
direction of the content of their own cognition. Moreover,
subjects recalled significantly more of their own
improvised arguments than provided arguments. They also
evaluated their own arguments as more original than those
provided to them.
Another area of research suggesting that a person's own
responses are important in mediating persuasion is
inoculation theory. McGuire (1964) suggests that
resistance to persuasion can be created by providing
information and arguments supportive of an individual's
original attitude or by "inoculating" the individual by


99
included to investigate the possibilities of it's use as a
behavioral measure of change related to counseling
messages. No hypotheses involving the CRT were made, and
no significant results were found. Development of
behavioral measures for counseling research is a complex
task due to the need to match the measure to presenting
problem and the need to determine what is appropriate
behavior changes.
One final point which must be addressed by future
research concerns the generalizability of experimental
studies (such as the present one) to actual counseling
settings. While the messages used in the present study
could be used in an actual counseling situation, the
differences between the two settings far outweigh the
similarities. Strong (1971) concludes that the
implications for counseling derived from laboratory
research are markedly limited due to the wide gap that
exists between the lab and the counseling room.
In sum, the ELM postulates several mediational
processes in which variables can have an impact upon
persuasion in general and acceptance of counseling messages
in particular. The importance of argument quality is a
solid finding in both basic persuasion ELM research and
counseling-related ELM research. The application of ELM
postulates concerning involvement and other variables which
may affect argument scrutiny is just beginning. While it


23
could have differing effects depending upon the level of
elaboration likelihood. For example, the quality of the
message arguments should have a greater influence when
elaboration likelihood is high, while the actual number of
arguments could have a greater influence when elaboration
likelihood is low. In the latter situation, the person may
employ a decision rule such as, "the more arguments the
better."
To test these hypotheses Petty and Cacioppo (1984)
conducted two studies. In the first experiment, college
students received a message on the issue of instituting
senior comprehensive exams. Personal relevance was
manipulated as outlined in the Petty, Cacioppo, and Goldman
(1981) study described previously. Subjects received one
of four messages in support of the exam proposal: 1) three
strong arguments, 2) three weak arguments, 3) nine strong
arguments, and 4) nine weak arguments. Attitude rating
results revealed that number of arguments was a more
important determinant of persuasion under low than high
relevance but that quality of the arguments was more
important under high than low relevance condition. In the
second experiment, the message concerned a proposal to
increase tuition, but relevance was manipulated by either
stating that the proposal was for the student's own
university (high relevance) or for a distant university
(low relevance). In this study the message contained


129
Kelly, G. A. (1955) The psychology of personal
constructs. New York: Norton.
Kiesler, C. A., Collins, B. E., & Miller, N. (1969).
Attitude change: A critical analysis of theoretical
approaches. New York: Wiley.
King, B., & Janis, I. L. (1956). Comparison of the
effectiveness of improvised versus non-improvised role-
playing in producing opinion changes. Human Relations,
9, 177-186.
Lammers, H. B., & Becker, L. A. (1980). Distraction:
Effects on the perceived extremity of a communication
on cognitive responses. Personality and Social
Psychology Bulletin, 6_, 261-266.
Landman, J., & Manis, M. (1983). Social cognition: Some
historical and theoretical perspectives. In L.
Berkowitz (Ed.), Advances in Experimental Social
Psychology (Vol. 16, pp. 49-123). New York: Academic
Press.
Latane, B., & Darley, J. M. (1970). The unresponsive
bystander: Why doesn't he help? New York: Appleton-
Century-Crofts.
Latane, B., Williams, K., & Harkins, S. G. (1979). Many
hands make light the work: The causes and consequences
of social loafing. Journal of Personality and Social
Psychology, 37, 822-832.
Lewin, K. (1947). Group decision and social change. In T.
Newcomb & E. Hartley (Eds.), Readings in social
psychology. New York: Holt, Rinehart & Winston.
Lord, C. G., Ross L., & Lepper, M. R. (1979). Biased
assimilation and attitude polarization: The effects of
prior theories on subsequently considered evidence.
Journal of Personality and Social Psychology, 37, 2098-
2109.
MacDonald, M. L. (1975). A behavioral assessment
methodology applied to the measurement of assertion.
(Doctoral dissertation, University of Illinios, 1974.
Dissertation Abstracts International, 35, 6101B.
McCormick, I. A. (1984). A simple version of the Rathus
Assertiveness Schedule. Behavioral Assessment, 7, 95-
99.


118
(Depending upon the subject's assignment to either
strong or weak argument quality, he or she will next hear
the appropriate arguments for the use of cognitive
restructuring. In this transcript strong arguments are
presented next and then weak arguments.)
Strong argument quality intervention
First, cognitive restructuring works. Controlled,
laboratory research published in the journal, Cognitive
Psychotherapy and Research, Dore & Caplan (1982) has shown
that the use of the cognitive restructuring technique has
enabled people with problems in personal assertion to
become more assertive 74 percent of the time and to become
more consistent in their self-expression. Moreover,
research has demonstrated significant reductions in the
anxiety, negative feelings, and poor self-images associated
with this problem. It also helped alleviate the tension
and the stress that often accompanies problems in personal
assertion.
Second, cognitive restructuring enables you to analyze
your problems in asserting yourself giving you greater
understanding of yourself. You can look at yourself more
objectively and see yourself as others see you. You also
become more aware of automatic thoughts and irrational
beliefs and the impact that they have had on the way you
f eel.


47
amount of counterarguments generated to weak arguments and
increased the amount of favorable thoughts generated to
strong arguments. These results show that increased
involvement increases the distinction of strong and weak
arguments.
This interaction of personal involvement and argument
quality has been replicated several times (Petty, Cacioppo,
and Goldman, 1981; Petty, Cacioppo, and Heesacker, 1981;
Petty and Cacioppo, 1984). For example, in the context of
examining the effects of source expertise under conditions
of high and low involvement, Petty, Cacioppo, and Goldman
(1981) found that under high conditions objective cognitive
processing increased distinction of argument quality and
that argument quality was the primary determinant of
persuasion. Under low involvement conditions, the
peripheral cue of source expertise exerted a greater effect
than argument quality.
Further support for the notion that increasing
involvement heightens message scrutiny was provided by
studies in which involvement was manipulated along with
source and message cues. Results of Chaiken's (1980) work
indicate that high issue involvement subjects tend to use a
systematic information processing strategy, whereas low
involvement subjects tend to use a heuristic processing
strategy. Subjects who use a systematic strategy focus on


130
McGuire, W. J. (1964). Inducing resistance to persuasion:
Some contemporary approaches. In L. Berkowitz (Ed.),
Advances in Experimental Social Psychology (Vol. 1, pp.
191-229). New York: Academic Press.
McGuire, W., & Papageorgis, D. (1961). The relative
efficacy of various types of prior belief defense in
producing immunity against persuasion. Journal of
Abnormal and Social Psychology, 62, 327-337.
Mann, R. J., & Flowers, J. V. (1978). An investigation of
the reliability and validity of the Rathus
Assertiveness Schedule. Psychological Reports, 42,
632-634.
Markus, H., & Sentis, K. (1982). The self in social
information processing. In J. Suls (Ed.). Social
psychological perspectives on the self (pp. 41-70).
Hillsdale, NJ: Erlbaum.
Markus, J., & Zajonc, R. B. (1985). The cognitive
perspective in social psychology. In G. Lindsey & E.
Aronson (Eds.), Handbook of social psychology:
Vol. I. Theory and method (pp. 137-230). New York:
Random House.
Miller, N. (1965). Involvement and dogmatism as inhibitors
of attitude change. Journal of Experimental Social
Psychology, .1, 121-132.
Morgan, W. G. (1974). The relationship between expressed
social fears and assertiveness and its treatment
implications. Behavior Research and Therapy, 12, 155-
257.
Orenstein, H., Orenstein, E., & Carr, J. (1975).
Assertiveness and anxiety: A correlational study.
Journal of Behavior Therapy and Experimental
Psychiatry, 6, 203-207.
Ostrom, T. M., & Brock, T. C. (1968). A cognitive model of
attitudinal involvement. In R. Abelson, E. Aronson, W.
J. McGuire, T. M. Newcomb, J. j. Rosenberg, & P. H.
Tannenbaum (Eds.), Theories of cognitive consistency:
A sourcebook. Chicago: Rand McNally.
Petty, R. E., & Cacioppo, J. T. (1977). Forewarning,
cognitive responding, and resistance to persuasion.
Journal of Personality and Social Psychology, 35, 645-
65in ~


6
Goldman (1981) provided evidence that under conditions of
high involvement a thoughtful evaluation of message content
is the most important determinant of attitude change. But
is the "involvement" of the laboratory comparable to that
brought to counseling or is it more complex? This study
will attempt to explore the complexity of personal
involvement and its influence upon the cognitive processing
of a counseling technique.
In the research on susceptibility to influence,
personal involvement or relevance has been defined as the
extent to which the issue under consideration is of
personal importance (Petty and Cacioppo, 1979). Variables
similar to this definition of personal involvement have
been alternately referred to as "issue involvement"
(Kiesler, Collins, and Miller, 1969), "personal
involvement" (Sherif, Kelly, Rodgers, Sarup, and Tittler,
1973), and "ego-involvement" (Rhine and Severance, 1970;
Greenwald, 1980). As mentioned earlier, any person seeking
counseling is likely to be highly involved with the problem
or issue presented.
Another factor to consider is that people who are
personally involved with an issue have done some amount of
prior thinking on the topic before seeking counseling. It
is possible that this information has been organized into a
schema or guiding principle for the perception of new
information (Fiske and Taylor, 1984). Schema-driven


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ii
ABSTRACT vi
CHAPTERS
IINTRODUCTION 1
IIREVIEW OF THE LITERATURE 9
The Cognitive Response Approach to Persuasion. 9
The Elaboration Likelihood Model 19
Cognitive Processing 25
Objective Processing 28
Biased Processing 32
Involvement and Persuasion 38
Involvement and Cognitive Processing 44
Involvement Level in Counseling 49
Hypotheses 51
IIIMETHODS 54
Research Design 55
Cognitive Responses to Arguments 57
Description of the Sample 59
Instrumentation 61
Relationships Grid 61
Personal Problems Inventory 66
Assertion Survey 69
Thought Listing Techniques 70
Cognitive Assessment Questionnaire 73
Procedures 73
Analyses of Data 75
Summary 75
IV


18
subjects either one or three times to a set of strong or
weak arguments in support of the recommendation that senior
comprehensive exams be instituted. According to their
theory, argument quality should differentiate people's
attitudes more after three than after one message
presentation. This should occur if it is the nature of
people's issue-relevant thoughts rather than the number of
different arguments they learn. Results indicate that
argument quality did differentiate people's attitudes more
after three exposures than after one. Subjects also
recalled more arguments after three presentations than
after one, but the amount of the message learned was
unrelated to their attitudes.
The fourth source of evidence comes from the use of
psychophysiological measures. Cacioppo and Petty (1979)
used integrated electromyographic (IEMG) activity as a
measure of subjects' silent language processing. In one
experiment, IEMG activity was recorded as subjects sat
quietly and as they listened to a tape-recorded message.
One minute prior to the message, subjects heard one of
three announcements about the message. One group only
heard that a message would be presented in one minute; a
second group was forewarned that they would hear a message
recommending more lenient visitation hours be instituted in
dormitories (a proattitudinal message according to
pretesting); and a third group was forewarned that they


24
either three strong argument, three weak argument, or six
arguments (three strong and three weak). Results indicate
that under high relevance conditions three strong arguments
elicited more attitude agreement with the proposal than
three weak arguments. The agreement for six arguments was
greater than that for the three weak arguments but less
than that for the three strong arguments. Under low
relevance conditions, six arguments produced the most
agreement followed by three strong and finally three weak.
Thus, in these two studies argument quantity served as a
cue under low relevance conditions, but quality of the
arguments was more important under high relevance
conditions.
This research suggests that two distinct routes to
persuasion do exist and differ in their level of
elaboration likelihood. The ELM provides a organizing
framework for understanding these major cognitive processes
underlying persuasion and how variables relate to these
processes. Petty and Cacioppo (1986) present the ELM in
postulate form as follows:
1. People are motivated to hold correct attitudes
(p. 127) .
2. Although people want to hold correct attitudes,
the amount and nature of issue-relevant elaboration
in which people are willing or able to engage to
evaluate a message vary with individual and
situational factors (p. 128).
3. Variables can affect the amount and direction
of attitude change by: (A) serving as persuasive
arguments, (B) serving as peripheral cues, and/or
(C) affecting the extent or direction of issue and
argument elaboration (p. 132) .


15
produced by making the students aware of how desegregation
would help in the attainment of some important goals (e.g.,
greater American prestige in other countries). Stotland et
al. (1959) had college students read a message designed to
give insight into the psychodynamics of racial prejudice.
They also assigned subjects to various manipulations
intended to facilitate the internal restructuring of
beliefs (e.g., ordering statements into cause-and-effect
sequences). Attitude measures showed no immediate
reduction in prejudice, but a significant reduction was
found in follow-up measures three to four weeks later.
These findings suggest that the restructuring of internal
beliefs with subsequent attitude change takes some time to
occur.
Thus far early research in support of a cognitive
response approach to persuasion has been presented from
three areas: 1) active versus passive participation in
persuasion, 2) inoculation theory, and 3) attitude
underlying beliefs linkages. Cognitive response
researchers have also provided empirical evidence for the
importance of cognition in producing persuasion. They
present four types of evidence for the cognitive response
approach based on forewarning manipulations, issue-relevant
thinking manipulations, argument quality manipulations, and
psychophysiological measures (Cacioppo, Petty, &
Stoltenberg 1985).


56
a median split for issue involvement and upper and lower
divisions of a trichotomization of ego-involvement and then
randomly assigned to either strong or weak intervention
quality. Ego-involvement was operationalized in terms of a
summed relationship score for the construct, "Assertive
Nonassertive" on the Relationships Grid. The summed
relationship score (Bannister, 1965) is a measure of the
total variance within a construct system accounted for by
one construct. The greater the total variance for a
construct the greater the implications and importance it
holds for a person. The variable, ego-involvement, had two
levels: high and low. The high ego-involvement level
refered to students who scored in the uppermost division of
a trichotomization of the sample's summed relationship
scores for the construct, AssertiveNonassertive. The low
ego-involvement level refered to students who scored in the
lowermost division for that score.
The second factor, issue involvement, was
operationalized by the use of an objective inventory
assessing personal assertiveness, the Simple Rathus
Assertiveness Schedule (McCormick, 1984). Subjects who
scored above the median for the pool of subjects on the
objective inventory (high trait assertiveness) were
considered to have low issue involvement. As scores below
the median on the objective inventory would indicate above


65
The calculation of variance scores is described by
Bannister (1965) Pearson product-moment correlations are
first calculated between all constructs. Second, each
correlation is squared, multiplied by 100, and the original
sign retained. These scores are called relationship
scores. The relationship scores for any one construct may
be summed to give the total variance for that construct.
In this study the relationship scores for the construct,
"Assertive-Nonassertive," will be summed to give the total
variance accounted for by that construct. If this score is
then divided by the total variance in the system across all
constructs, then it yields an index of the proportion of
variance accounted for by the single construct Assertive-
Non Assertive. A trichotomization was performed along the
distribution of these scores. The uppermost division was
designated as high ego-involvement and the lowermost as low
ego-involvement.
Investigations into the psychometric properties of the
Rep Grid are complicated by the varieties of grids and
scorings in use. While the reliability and validity of
summed relationship scores is not reported in the
literature, psychometric properties of total variance
scores (Bannister, 1960) are reported. Bannister (1962)
reports an immediate retest reliability of 0.35 in a sample
of 30 normal subjects. Honess (1977) reports the same
reliability correlation in a rank order grid but a


120
Secondly, I've tried cognitive restructuring once, and
even though it didn't help me with my inadequate personal
assertion that doesn't mean it wouldn't work more
effectively for you. Most techniques work better for some
people than for others, and you may be one that it works
for. I think that everyone should at least try it and see
if it works for them.
Thirdly, although I haven't been able to interview
everyone that has used this technique, I personally know of
several people who did use it and thought it was okay.
They seemed to think that it was better than some of the
other methods they had used to treat their assertion
problems.
Finally, most people waste a lot of their free time.
These are times often spent watching television, sleeping,
or in a similarly unproductive activity. At least
cognitive restructuring is constructive. It is probably
more useful and interesting than what you would ordinarily
be doing with your free time. So it makes sense to try the
technique.
In summary, cognitive restructuring is a quick and easy
technique for treating problems in personal assertion.
I've tried it, and though it didn't work for me it may work
for you. Additionally, I've talked to others that have
tried it, and they thought that it worked better than some
other techniques they had tried. Finally, at least it's a


119
Third, because it reduces the stress associated with
problems in personal assertion, cognitive restructuring
provides many additional health benefits. For example, the
use of cognitive restructuring has been associated with
lower blood pressure, higher levels of energy, and general
nutritional improvement.
Finally, this technique can be used in the treatment of
other problems as well. It is effective in helping you
gain control over many areas of your life in which you
might feel anxious or depressed. After using this
technique people typically report feeling better about
themselves, more confident, relaxed and self-assured.
In summary, cognitive restructuring works. Research
indicates its effectiveness. It also provides greater
self-awareness and understanding. It is associated with
general health benefits and is useful in many other aspects
of your life as well.
Weak argument quality intervention
Cognitive restructuring is quick and easy to learn.
Most people can master the technique without difficulty and
can begin to use it soon after they learn it. So for this
reason cognitive restructuring is a very simple technique
to use for treating problems in personal assertion.
Because it doesn't require a great deal of effort, it's
worth a try. Even if it isn't especially beneficial, at
least it isn't harmful.


61
Third, students were assigned to the appropriate group
according to their high and/or low classifications on ego
and issue involvement. This procedure resulted in the
inclusion of 104 subjects (N=40, F=64). The percentage of
males within each group is presented in Table 1. Mean ego-
involvement and issue involvement scores and standard
deviations for all groups in this study are also listed in
Table 1.
Instrumentation
Six instruments were used in this study: 1) The
Relationships Grid, 2) Personal Problems Inventory (Simple
Rathus Assertiveness Schedule), 3) Assertion Survey, 4)
Thought Listing for Message, 5) Covert Rehearsal
Technique, and 6) Cognitive Assessment Questionnaire.
Relationships Grid
The Relationships Grid (RG; see Appendix A) is a form
of the dyad grid (Ryle & Lunghi, 1970). The RG is an
adaptation of the original repertory grid technique
developed by G.A. Kelly (1955) to elicit and measure
personal construct systems. "A grid may be defined as any
form of sorting task which allows for the assessment of
relationships between constructs and which yields these
primary data in matrix form" (Bannister and Mair, 1968, p.
136). The basic components of a grid are elements and
constructs. In the RG, like other dyad grids, the elements
used consist of relationships (e.g. my relationship with my


95
they probably did not represent relatively pure levels of
high and low issue involvement.
As noted earlier, these same criticisms may also be
applied to the ego-involvement variable. No significant
effects were found for ego-involvement for any dependent
measure. The absence of a manipulation check precludes the
conclusion that ego-involvement did not influence subjects'
cognitive responses to the counseling message. Therefore,
it is not surprising that the biasing effect predicted by
hypothesis three was not found. Hypothesis three predicted
that under conditions of high issue involvement, high ego-
involvement would exert a greater influence on the weak
than on the strong messages. Under low issue involvement,
it predicted that ego-involvement would exhibit an equal
influence on both strong and weak messages.
In summary, although the results of this study did not
support the hypotheses of either issue involvement or ego-
involvement or a combination of the two interacting with
argument quality to influence cognitive responses to a
counseling message, the results did not necessarily negate
those results due to possible methodological problems. The
results do suggest that the quality of the arguments used
in a counseling message have a significant impact upon
acceptance of the message with messages containing strong
arguments being more accepted than messages containing weak
arguments.


98
subject's own estimate of importance are all possible
methods of determining construct superordinancy. Since the
present ego-involvement manipulation produced no effects,
perhaps one of these alternative operationalizations should
be pursued in future research.
In addition to the above two changes in methodology,
future research may also benefit from investigating the
influence of various message factors upon acceptance of
counseling messages. Previous ELM research (Cacioppo &
Petty, 1979b) shows that repeating a persuasive
communication tends first to increase and then to decrease
argument. What effects might message repetition produce in
counseling and when should a counselor change messages?
Another interesting question is how should a counselor word
a message. Petty, Cacioppo, and Heesacker (1981) offer
evidence that when personal relevance is low, the use of
arguments in the form of rhetorical questions increases
elaboration, but when personal relevance is high, argument
scrutiny is reduced by the use of rhetorical questions.
The high personal relevance subjects reported that they
found the rhetorical questions to be distracting.
Further, future research may also benefit from more
refinement of behavioral measures appropriate for
counseling research. Eliciting behavior change as well as
attitude change is central to successful counseling. In
the present study the covert rehearsal technique (CRT) was


48
the content of persuasive messages and carefully evaluate
the content in a relatively objective manner.
Although the research cited above suggests that people
become more likely to evaluate carefully and objectively
the issue-relevant arguments in a persuasive message as
personal involvement increases, circumstances may occur in
which cognitive processing becomes biased as involvement
increases. As suggested in the section on biased
processing, the presence of a well-organized structure of
prior knowledge may lead to biased processing. An
individual who is involved with an issue has done some
amount of prior thinking about the pool of issue-relevant
arguments. Some individuals may have done a great deal of
prior thinking and have a well-developed construct system
or schema for the issue in question. Such individuals may
have a greater ability to counterargue persuasive messages,
a greater store of supportive arguments for their own
beliefs, and/or little motivation to consider yet another
persuasive appeal.
Ostrom and Brock (1968) suggest that when an issue is
intimately connected with an individual's central values,
its personal involvement may be intense enough to generate
biased processing. Greenwald (1980) proposes that the ego
is characterized by a cognitive bias of conservatism or
resistance to change that serves to protect the self's
organization of knowledge. If the ego is threatened, it


CHAPTER III
METHODS
This study was designed to investigate the effects of
ego-involvement and issue involvement upon the cognitive
processing of strong and weak arguments for the use of
cognitive restructuring as a treatment technique for
assertiveness. The sample for the study consisted of 120
students drawn from the University of Florida introductory
psychology subject pool and an undergraduate sociology
class, Marriage and Family. Cognitive responses and
acceptance of arguments was assessed across four groups: 1)
high ego and high issue involvement, 2) high ego and low
issue involvement, 3) low ego and high issue involvement,
and 4) low ego and low issue involvement. Half of each
group was exposed to strong arguments and half to weak
arguments in favor of using cognitive restructuring.
Levels of ego-involvement were assessed by using the
Relationships Grid (see Appendix A), a form of the dyad
grid (Ryle and Lunghi, 1970). Levels of issue involvement
were measured through the use of the Simple Rathus
Assertiveness Schedule (McCormick, 1984). The preceding
instruments were administered to one General Psychology
class and one Marriage and Family class.
54


17
questionnaire he had brought. Half of the students were
forewarned that the psychologist would be advocating that
all freshmen and sophomores be required to live in campus
dorms, a topic which pretesting showed most students to be
against. The remaining students were not forewarned of the
topic. Half of all of the students were then asked to list
their thoughts just prior to the speech. So far this study
replicates the conditions of the forewarning study
described above. The other half of the students, however,
were asked to list all of their thoughts on the topic of
requiring students to live in dorms. Thus, half of the
subjects were instructed to engage in issue-relevant
thinking whether or not they had been forewarned. Attitude
measures revealed that unwarned subjects who engaged in
issue-relevant thinking demonstrated resistance to
persuasion equal to that exhibited by the forewarned
groups. It appears that just being instructed to think
about a topic can produce cognitive responses which support
one's beliefs and, subsequently, lead to greater resistance
to persuasion.
In the third line of evidence argument quality has been
operationally defined such that "strong" arguments elicit
more favorable than unfavorable thoughts about a message,
whereas "weak" arguments elicit more unfavorable than
favorable thoughts concerning the message. In an
illustrative study, Cacioppo and Petty (1985) exposed


40
involvement increased. For Sherif and Hovland, involvement
seemed to refer to both the intensity with which an
attitude is held and the importance of that attitude for
the self-identity.
Sherif and Hovland (1961) described two field studies
in support of social judgment theory. For illustrative
purposes the "prohibition study" (Hovland, Harvey, and
Sherif, 1957) will be presented here. The prohibition
study was conducted in Oklahoma shortly after a close
referendum which favored prohibition in the final outcome.
To insure that subjects were deeply involved with the
issue, dry-stand subjects were recruited from Women's
Christian Temperance Union groups, the Salvation Army, and
strict denominational colleges. Wet-side subjects were
selected from acquaintances of the experimenters. A group
of moderate-stand subjects were also included in the study.
Subjects were exposed to a 15 minute tape recorded message,
which presented either a wet, moderately wet, or dry stand
on prohibition. Dependent measures were obtained on the
following: (1) the communication's estimated position, (2)
subject's reactions to the communication in terms of
fairness and impartiality, and (3) subject's preferred
position and latitudes of acceptance and rejection.
In support of social judgment theory, Hovland, Harvey,
and Sherif (1957) found that for the majority of the
subjects with initially extreme stands, their latitudes of


43
Contradictory results have also been reported by
researchers who have manipulated forewarning along with
involvement. Apsler and Sears (1968) either forewarned or
did not forewarn subjects that they would read a proposal
calling for the replacement of professors by supervised
teaching assistants. Involvement was manipulated by
informing half of the subjects that the proposal would go
into effect ten years from now. Results indicated a
significant interaction with forewarning inhibiting
attitude change under high personal involvement but
facilitating change under low personal involvement. Dean,
Austin, and Watts (1971) found, however, that forewarning
inhibited attitude change for both high and low involvement
issues and its effect was even more pronounced for the low
involvement issue. On the other hand, Petty and Cacioppo
(1979b) found the inhibiting effect of forewarning to be
greater under high than low involvement conditions. In
these latter two studies the forewarning given was one of
persuasive intent not content, but all three studies
manipulated involvement in similar ways.
One interesting finding in the Petty and Cacioppo
(1979b) study was that when no forewarning was given, high
involved subjects tended to show more attitude change than
low involved subjects. Similar findings were also reported
by Apsler and Sears (1968) and Eagly (1967). Eagly (1967)
gave subjects favorable or unfavorable discrepant


22
they were led to believe that the exam policy would begin
in 10 years and thus not affect them. For half of the
subjects the message was attributed to a Princeton
University Professor (high credible source), while for the
other half it was attributed to a high school student (low
credible source). Source credibility served as a
persuasion context cue in this study. Attitude rating
results revealed two significant interactions. First, a
relevance X message quality interaction showed that
argument quality was a more important determinant of
persuasion for high than low relevance subjects. Second, a
relevance X source credibility interaction suggested that
the source cue was a more important determinant for
persuasion for low than high relevance subjects. Thus,
under conditions of high relevance (high elaboration
likelihood), subjects exerted the effort to evaluate the
issue-relevant arguments presented. Under condition of low
elaboration likelihood, they were persuaded by a context
cue and appeared to be unaffected by argument quality.
While in the above study message factors were prepotent
under condition of central route processing and source
factors were prepotent under conditions of peripheral route
processing, the central/peripheral distinction is not one
between message and source factors. It refers to whether
issue-relevant thinking versus context cues or decision
rules leads to attitude change. Different message factors


36
anticipating events, a construction system embracing
ordinal relationships between constructs" (p. 56).
Constructs are an individual's basic interpretations of the
world and are bipolar in that they describe the way in
which certain things are perceived as alike and different
from others (Bannister and Mair, 1968). Constructs which
are superordinate in a person's system subsume other
constructs, have relatively more implications, and are
generally based on more information. Thus, they function
similarly to schemas.
In an investigation of the effects of prior knowledge
and schemas upon cognitive processing, Cacioppo, Petty, and
Sidera (1982) provide evidence that prior knowledge does
influence the processing of persuasive messages. Subjects
who characterized themselves using trait adjectives as
either "religious" or "legalistic" people were exposed to
strong or weak proattitudinal messages, which were either
schema-congruent or schema-incongruent. After presentation
of the message subjects rated its persuasiveness and listed
their thoughts. Results suggested that with proattitudinal
messages subjects who received schema-relevant information
were more positive about the quality of the communication's
arguments and in their listed thoughts. Thus, it appears
that self-schemas influence cognitive responses in a biased
or top-down fashion.


96
Need for Cognition
Individual differences in motivation to think have
relevant implications for message processing and
elaboration likelihood. Cacioppo and Petty (1982)
developed the need for cognition scale (NCS) in order to
distinguish people who dispositionally tend to engage in
and enjoy effortful analytic activity from those who do
not. Petty and Cacioppo (1986) further speculate that
individuals high in need for cognition should be more
likely to carefully evaluate and elaborate upon the issue
relevant arguments provided in persuasive messages.
However, additional correlational analysis conducted in the
present study did not support this notion. The correlation
between need for cognition score and total number of
relevant thoughts was nonsignificant; r=-.01, p=.88. This
finding does not support Petty and Cacioppo's notion of a
positive direct relationship between need for cognition and
elaboration likelihood.
Future Considerations
The investigation of the effects of issue involvement
and ego-involvement upon acceptance of counseling messages
is a relatively undeveloped area of inquiry. Only the four
studies cited earlier in this discussion have attempted to
manipulate issue involvement within research with a
counseling context. The present investigation is the first
to investigate the possible biasing effect of ego-


97
involvement under conditions of high issue involvement.
All of these studies appear to have methodological problems
in the manipulation of issue involvement. In the present
study the Simple Rathus Assertiveness Scale failed to
successfully differentiate subjects into high and low issue
involvement groups, and the theory-based measure appears
also not to have divided the sample into pure high and low
groups but into moderately high and moderately low groups.
As suggested earlier, however, these problems in
manipulating issue involvement probably arise from the
counseling context of the message and the ubiquitous
concerns that people have with interpersonal relations.
Future research may benefit from using alternative
approaches to manipulating issue involvement. For example,
crossing a measure of assertiveness level with a measure of
relevant thoughts generated may produce a high involvement
group, but it may be necessary to add a distraction
manipulation to the low involvement group in order to
obtain a relatively pure low level of elaboration
likelihood.
Similar steps may also be beneficial in refining the
ego-involvement manipulation. Within Personal Construct
methodology a number of ways exist to operationally define
superordinancy of a construct. Besides the amount of
relative variance accounted for by a single construct,
extremity ratings, resistance-to-change grids, and the


Table 4. Analysis of Variance for Attitide Rating Scores
Source
Sum of Squares
df
Mean Square
F
P
Issue
Involvement
19.254
1
19.254
9.23
.003
Ego
4.507
1
4.507
2.16
.145
Quality
7.358
1
7.358
3.53
.063
Issue
Involvement x
Ego
1.178
1
1.178
0.57
.454
Issue
Involvement x
Qual
1.338
1
1.338
0.64
.425
Ego x
Quality
0.575
1
0.575
0.28
.601
Issue
x Ego x Qual
0.241
1
0.241
0.12
.734
Error
200.187
96
2.085
CO
4^


72
classifications of the subjects judged the thoughts listed
for frequency of spontaneous use of cognitive
restructuring. This consisted of the post-coding of each
subject's listed thoughts into one of five categories: 1)
self-instructional statement, 2) self-assuring statement,
3) self-doubting statement, 4) presentation content
statement, and 5) other statement. Inter-judge reliability
Pearson product-moment correlations were .81 for self-
instructional statements, .85 for self-assuring statements,
.92 for self-doubting statements, .72 for presentation
content, and .34 for other statements.
Support for the Thought Listing Technique (TLT) as a
reliable measure can be found in a study by Cullen
(1969/1968). She compared the split-half reliability and
test-retest reliability of the TLT with Likert attitude and
Thurstone attitude scales on two topics, birth control and
segregation. .She found average split-half reliabilities of
0.78 for Thought-Listing, 0.83 for Likert scales, and 0.55
for Thurstone scales. The average test-retest
reliabilities were 0.64 for the TLT, 0.83 for Likert
scales, and 0.53 for Thurstone scales.
In indirect support for the construct validity of the
TLT, Petty, Harkins, and Williams (1980) found that the
implied or real presence of others working on the same
cognitive task decreased the thoughts listed by an
individual in that implied or real presence of others when


10
material not in the message itself. From their prior
knowledge and evaluation of the communication the person
may generate cognitions that agree, disagree, or are
irrelevant to the persuasive message under consideration.
To the extent that the message elicits supportive cognitive
responses, attitude change in the advocated direction
should be facilitated. To the extent that the message
elicits negative cognitive responses, attitude change
should be inhibited.
The concept that an individual's cognitive responses
are an important mediator of attitudes is not a recent
development in psychology. Freud's (1900/1939) method of
free association was an early attempt at the measurement of
cognitive responses and their role as mediators of
attitudes in a clinical context. In the early study of
attitude change Hovland (1951) suggested that the accurate
and complete recording of an audience's thoughts as they
listened to a communication would constitute the best
method for investigating the internal process of change.
While the cognitive response approach to persuasion was
not formally proposed until 1968, researchers have been
concerned with cognitive responses in persuasion since the
early research of attitude change. In a classic study
investigating active versus passive participation in the
persuasion process Lewin (1947) compared individual
instruction with group discussion. The goal of both


REFERENCES
Apsler, R., & Sears, D. 0. (1968). Warning, personal
involvement, and attitude change. Journal of
Personality and Social Psychology, 9_, 162-166,
Bannister, D. (1960). Conceptual structure in thought
disordered schizophrenics. Journal of Mental science,
108, 825.
Bannister, D. (1962). The nature and measurement of
schizophrenic thought disorder. Journal of Mental
Science, 108, 825-842.
Bannister, D. (1965). The rationale and clinical relevance
of repertory grid technique. British Journal of
Psychiatry, 111, 977-982.
Bannister, D. & Fransella, F. (1966). A grid test of
schizophrenic thought disorder. British Journal of
Social and Clinical Psychology, 5^, 95.
Bannister, D., & Mair, M. (1968). The Evaluation of
Personal Constructs. New York: Academic Press.
Blanchard, E. B., Turner, J., Eschette, N., & Coury,
V. M. (1977). Assertiveness training for dental
students. Journal of Dental Education, 41, 206-208.
Brehm, J. W. (1972). Responses to loss of freedom: A
theory of psychological reactance. Morristown, NJ:
General Learning Press.
Brock, T. C. (1967). Communication discrepancy and intent
to persuade as determinants of counterargument
production. Journal of Experimental Social
Psychology, 3_, 296-309.
Cacioppo, J. T., & Petty, R. E. (1979). Attitudes and
cognitive response: An electrophysiological approach.
Journal of Personality and Social Psychology, 37, 2181-
2199.
124


14
contagious) which people generally have little practice in
defending. According to inoculation theory individuals in
the active condition would perform more poorly than those
in the passive condition when exposed to immediate
counterattitudinal attack. However, they should
demonstrate increased resistance to attacks which are
delayed. His results conformed to this predicted
interaction. Active defenses increased in resistance when
the attack occurred one week later, while passive defenses
declined in resistance except for the passive refutational-
different defense. This passive defense, like the active
ones, presumably posed a threat that motivated the subjects
to generate additional cognitive responses in support of
their beliefs.
Further support for the principle that a person's own
cognitive responses are important in producing persuasion
is offered by studies investigating the relation of
attitudes to underlying beliefs and values. In the 1950s
and 1960s widespread racial prejudice caused many
psychologists to be concerned with changing prevailing
attitudes. Two studies designed to assess the impact upon
attitudes of changing underlying beliefs that illustrate
the importance of cognition were conducted by Carlson
(1956) and Stotland, Katz, and Patchen (1959). Carlson
(1956) found that in moderately prejudiced college students
more favorable attitudes toward racial integration could be


21
of issue-relevant information and rely upon positive or
negative cues in the persuasion context. A simple but
reasonable decision rule could also be used in the
persuasion context when elaboration likelihood is low.
In order to provide empirical support for distinct
central and peripheral routes to persuasion and the notion
of elaboration likelihood, researchers have used the
argument quality manipulation under high and low relevance
conditions. In high relevance conditions subjects were led
to believe that the issue had direct personal consequences
for them, while in low relevance condition subjects were
led to believe that the issue had few if any personal
consequences. The issues and messages used were easy to
understand so all subjects had the ability to think about
the information presented. Such a design suggests that
subjects in the high relevance conditions should follow the
central route to persuasion and subjects in the low
relevance conditions should follow the peripheral route.
Petty, Cacioppo, and Goldman (1981) conducted such a
study in which college students were exposed to a
counterattitudinal appeal (favoring senior comprehensive
exams). In addition to personal relevance and argument
quality, source expertise was also manipulated in this
study. Under conditions of high relevance, students were
led to believe that the exam policy would begin next year
and thus affect them. Under conditions of low relevance,


134
Wolpe, J., & Lang, P. J. (1964). A fear survey for use in
behavior therapy. Behavior Research and Therapy, 2,
27-30.
Wood, W. (1982). Retrieval of attitude-relevant
information from memory: Effects on susceptibility to
persuasion and on intrinsic motivation. Journal of
Personality and Social Psychology, 42, 798-810.
Wood, W., Kallgren, C., & Priesler, R. (1985). Access to
attitude relevant information in memory as a
determinant of persuasion. Journal of experimental
Social Psychology, 21, 73-85.
Zajonc, R. B. (1980). Feeling and thinking: Preferences
need no inferences. American Psychologist, 35, 151-
175.
Zimbardo, P. G. (1960). Involvement and communication
discrepancy as determinants of opinion conformity.
Journal of Abnormal and Social Psychology, 60, 86-94.


64
may believe him/herself to be moderately nonassertive in
that relationship and place a -2_ in the appropriate square
in the grid. The left pole or side of each construct will
be assigned a positive valence, while the right pole will
be assigned a negative valence. The Relationships Grid and
complete instructions are provided in Appendix A.
Estimated time for completion of the RG is fifteen minutes.
Numerous measures can be derived from a grid, but the
measure of interest for this study is the summed
relationship score for the construct, "Assertive-
Nonassertive." The summed relationship score (Bannister,
1965) is a measure of the total variance within a construct
system accounted for by a single construct. The greater
the total variance accounted for by one construct; the more
closely it is related to all other constructs. A construct
with greater variance and more implications for the rest of
an individual's construct system would be more
superordinate and hold more importance for that individual
than a construct accounting for less variance in the
system. In this study subjects with higher summed
relationship scores for whom the construct, Assertive-
Nonassertive, accounted for a greater amount of variance in
their construct systems were designated as highly involved.
Subjects for whom the assertiveness dimension accounted for
less variance in the system were designated as low ego
involved.


106
27. If a couple near me in the theatre were
talking rather loudly, I would ask them to
be quiet or to go somewhere else and talk.
28. Anyone trying to push ahead of me in a line
is in for a good battle.
29. I am quick to say what I think.
30. There are times when I just can't say
anything.


EFFECTS OF ISSUE INVOLVEMENT AND
EGO-INVOLVEMENT UPON ACCEPTANCE OF A
COUNSELING MESSAGE
By
STEVEN HUGH WILLIAMS
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR
THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1987


108
7. How would you rate the quality of the arguments used by
the speaker in support of the advocated topic?
Very poor arguments 1234567 Very good arguments
8. How likely are you to pursue counseling within the next
three months
Not at all likely 1234567 Very likely


APPENDIX A
RELATIONSHIPS GRID
instrument
in several
is designed to assess the ways you view
different types of relationships,
a list of the relationships found on the grid,
need to think of a person in your life who fits the
relationship described. Please use a different
each role described. In each case we are
you see yourself in the relationship.
This
yourself
Below is
You
person for
interested in how
1.Yourself in your relationship with your mother or
stepmother.
2. Yourself in your relationship with your current
roommate. If you don't have a roommate now, the
last person with whom you shared living quarters.
3. Yourself in your relationship with your father or
stepfather.
4.Yourself in your relationship with your current
romantic partner. If you currently don't have a
partner, think of your last boy/girlfriend.
5.Yourself in your relationship with the sibling
that you feel closest to. This may be either a
brother or sister. If you have no siblings think
of a person who has been most like a sibling to
you.
6.
Yourself in your relationship
boss. If you are not working
last boss. If you have never
someone who has had authority
with your current
now, think of your
worked, think of
over you.
7.Yourself in your relationship with your favorite
teacher or former teacher.
8.Yourself in your relationship with your closest
present friend of the same sex as yourself.
9.Yourself in your relationship with a person with
whom you usually feel most uncomfortable.
101


133
Ryle, A., & Lunghi, M. (1970). The dyad grid: A
modification of repertory grid technique. British
Journal of Psychiatry, 117, 323-327.
Ryle, A., & Lunghi, M. (1971). A therapist's prediction of
a patient's dyad grid. British Journal of Psychiatry,
118, 555-560.
Sherif, M. & Hovland, C.I. (1961). Social judgment:
Assimilation and contrast effects in communication and
attitude change. New Haven, CT: Yale University
Press.
Sherif, C. W., Kelly, M., Rodgers, H. L., Sarup, G., &
Tittler, B. (1973). Personal involvement, social
judgment, and action. Journal of Personality and
Social Psychology. 27 311-3 27 .
Sherif, C. W., Sherif, M., & Nebergall, R. E. (1965).
Attitude and attitude change: The social judgment-
involvement approach-! Philadelphia: Saunders.
Stoltenberg, C. C., & McNeil, B. W. (1984). Effects of
expertise and issue involvement on perceptions of
counseling. Journal of Social and Clinical Psychology,
2, 314-325.
Stotland, E., Katz, D., & Patchen, M. (1959). The
reduction of prejudice through the arousal of self
insight. Journal of Personality, 27, 507-531.
Strong, S. R. (1968). Counseling: An interpersonal
influence process. Journal of Counseling Psychology,
15, 215-224.
Strong, S. R. (1971). Experimental laboratory research in
counseling. Journal of Counseling Psychology, 18, 106-
110.
Tally, J. E., Barrow, J. C., Fulkerson, K. F., & Moore, C.
A. (1983). A comparison of telephone vs. mail
strategies in conducting a university psychological
services needs assessment. Journal of American College
Health, 32, 101-103.
Vaal, J. j. (1975). The Rathus Assertiveness Schedule:
Reliability at the junior high school level. Behavior
Therapy, £, 566-567.
Wolpe, J. (1973). Supervision transcript: V. Mainly
about assertive training. Journal of Behavior Therapy
and Experimental Psychiatry" T, 141-148.


APPENDIX F
COGNITIVE ASSESSMENT QUESTIONNAIRE
Please read the statements below and indicate how
characteristic each statement is of you, using the
following rating scale:
1. extremely uncharacteristic of me
2. fairly uncharacteristic of me
3.
neutral, neither characteristic nor uncharacteristic
of me
4.
5.
fairly characteristic of me
extremely characteristic of me
1.
, I would prefer complex to simple problems.
2. I like to have the responsibility of handling a
situation that required a lot of thinking.
3.
. Thinking is not my idea of fun.
4.
. I would rather do something that required little
thought than something that is sure to challenge
my thinking abilities.
5.
, I try to anticipate and avoid situations where
there is likely some chance I will have to think
in depth about something.
6.
, I find satisfaction in deliberating hard and for
long hours.
7.
, I only think as hard as I have to.
8.
I prefer to think about small, daily projects to
long-term ones.
9.
I like tasks that require little thought once I've
learned them.
10. The idea of relying on thought to make my way to
the top appeals to me.
114


5
task (in this case cognitive restructuring), he/she would
more likely actually perform the therapeutic task outside
the counseling room. Two variables which could affect the
client's attitude toward cognitive restructuring are the
quality of the arguments for its use presented by the
counselor and the degree of involvement the client
possesses toward changing or retaining his/her present
level of assertiveness.
In the process of getting a client to try a particular
technique, such as cognitive restructuring, the counselor
will present any number of arguments for its use. The
quality of these arguments will influence the favorability
of the client's attitude toward the therapeutic task. In
order to provide a means to empirically test the ELM Petty
and Cacioppo (1979b) devised a method for developing
"strong" and "weak" messages. This method will be used in
this study to provide the means to investigate the
influence of two different types of subjects involvement
upon their cognitive responses toward the therapeutic task
of cognitive restructuring.
Because people would have a high level of involvement
in a particular life problem before seeking counseling, it
would seem that their elaboration likelihood would
generally be high. According to the ELM they would thus be
more likely to engage in central route processing rather
than peripheral route processing. Petty, Cacioppo, and


4
or direction of issue and argument elaboration" (Petty and
Cacioppo, 1981, p. 132).
From the ELM viewpoint much of the interpersonal
influence research in counseling has focused on peripheral
cues such as counselor characteristics. One aim of this
study is to expand the focus of interpersonal influence
research in counseling to include the study of message and
recipient/client variables. In order to provide a context
for the study which would approximate to a counseling
analogue and provide a framework for the testing of these
variables, a life problem that would likely be presented in
counseling needed to be identified. The problem chosen
concerns social assertiveness. In a needs assessment
survey conducted on the Duke University Campus, Talley,
Barrow, Fulkerson, and Moore (1983) found that of the 52
needs they asked college students to rate the current
importance of the need to assertively stand up for myself
was ranked number seven in overall importance. Thus,
problems in personal assertion is very likely to be an
issue that students will present to college counselors.
In counseling the therapist often presents the client
with some task to perform in order to bring about
therapeutic change. In the treatment of assertiveness
problems, cognitive restructuring has been found to be an
effective treatment technique (Kaplan, 1982). If the
client acquires a favorable attitude toward the therapeutic


RELATIONSHIPS GRID
(D
lI
X
r0
-P
TJ
0)
P
G
-p
p
Cr
P
o
p
(0
p
(u
G
0)
t
fat
H
QJ

c
H
X
C
g
P
X
fc
x
p
i1
tn

CL)
0
M-t
X
0
-p
p
X!
Cfl
to
H
o
1
0
0
f0
(0
rl
0
(1)
P
G
X
S
fa
fa
fa
fa
CQ
H
fa
fa
w
Calm Anxious
Respectful Not respectful
Assertive Nonassertive
Dependent Independent
Does not
Good listener listen
Lacking in
Understanding understanding
Fair Unfair
Kind, helpful Hurtful
Open Closed
Satisfied Dissatisfied
103


Table 6. Intercorrelation of Descriptive and Dependent Variables
Ego-
Involvement
Argument
Quality
Attitude
Rating
Behavioral
Intention
CFCR
Need for
Cognition
Rathus
Scores
Relevant
Thoughts
Issue -.114
Involvement3 p<.249
-.001
p< .988
-.274
p< 005
-.132
p< 181
.065
p< 525
.006
p<.948
-.034
p<.730
-.841
p< .0001
Ego-
Involvement3
.001
p< .988
-.112
p<.259
-.055
p<.576
.017
p<.866
. 117
p<.234
.129
p<.191
.080
p<.419
Argument
Quality3
-.184
p< 06 2
-.097
p<.3 28
-.333
p< .0008
. 121
p<.222
-.019
p< .842
-.018
p< .852
Attitude
Rating
.730
p< .0001
.359
p< .0003
.012
p<.902
-.085
p<. 389
.243
p<.013
Behavioral
Intention
.424
p< 0001
. 131
p<.18 6
-.042
p<.668
.109
p<.271
CFCR
.034
p<.7 37
.052
p<.614
.055
p<.590
Need for
Cognition
.372
P<.0001
.014
p<.88 8
aReverse scored.
00
00


42
involvement level did not affect subjects' latitudes of
acceptance. According to social judgment theory, high
involvement should reduce latitudes of acceptance. Eagly
and Manis (1966) also report that highly involved subjects
react more negatively toward persuasive messages that
contradict their beliefs.
Other researchers, however, have not found results
consistent with the notion that higher levels of
involvement increase resistance to persuasion. Zimbardo
(1960) defined involvement in terms of an individual's
concern with the social consequences of his or her response
in a given situation rather than the intrinsic importance
of an issue for the individual. He found greater attitude
change with highly involved subjects than with less
involved subjects. Freedman (1964) also operationalized
involvement in terms of concern about a response. In
addition, he exposed subjects to messages which were
slightly, moderately, or extremely discrepant from their
initial positions. His results indicated that under low
involvement, there was more change with greater
discrepancy; but under high involvement, their relationship
was nonmonotonic with maximum change occurring at moderate
discrepancy. Moreover, under moderate discrepancy, both
high and low involved subjects changed by a similar amount.
Similar results were reported by Rhine and Severance
(1970).


131
Petty, R. E., & Cacioppo, J. T. (1979a). Effects of
forewarning of persuasive intent and involvement on
cognitive responses and persuasion. Personality and
Social Psychology Bulletin, 5^ 173-176.
Petty, R. E. & Cacioppo, J. T. (1979b). Issue involvement
can increase or decrease persuasion by enhancing
message relevant cognitive responses. Journal of
Personality and Social Psychology, 37, 1915-1926.
Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and
persuasion: Classic and contemporary approaches.
Dubuque, IAl William C. Brown.
Petty, R. E., & Cacicoppo, J. T. (1984). The effects of
involvement on responses to argument quantity and
quality: Central and peripheral routes to persuasion.
Journal of Personality and Social Psychology, 46, 69-
81.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration
likelihood model of persuasion. In L. Berkiwitz (Ed.),
Advances in Experimental Social Psychology (Vol. 19,
pp. 123-205). Orlando, FL: Academic Press.
Petty, R. E., Cacioppo, J. T. & Goldman, R. (1981).
Personal involvement as a determinant of argument-based
persuasion. Journal of Personality and Social
Psychology, 41, 847-855.
Petty, R. E., Cacioppo, J. T., & Heesacker, M. (1981). The
use of rhetorical questions in persuasion: A cognitive
response analysis. Journal of Personality and Social
Psychology, 40, 432-440.
Petty, R. E., Harkins, S. G., & Williams, K. 0. (1980).
The effects of group diffusion of cognitive effort on
attitudes: An information processing view. Journal of
Personality and Social Psychology, 38, 81-92.
Petty, R. E., Wells, G. L., & Brock, T. C. (1976).
Distraction can enhance or reduce yielding to
propaganda: Thought disruption versus effort
justification. Journal of Personality and Social
Psychology, 34, 874-884.
Quillan, J., Besing, S., & Dinning, D. (1977).
Standardization of the Rathus Assertiveness Schedule.
Journal of Clinical Psychology, 33, 418-422.
Rathus, S. A. (1973). A 30-item schedule for assessing
assertive behavior. Behavior Therapy, 4, 398-406.


117
with me, but I won't know until I ask. All they can do is
say no, and I can deal with that." Although this technique
can be used in a variety of disorders, one common
application is in the treatment of inadequate
assertiveness. Problems in personal assertion can be quite
common in college populations. Personal assertion refers
to self-expression through which one stands up for his or
her basic rights without violating the basic human rights
of others. Inadequate assertiveness may involve either
passive or aggressive behavior in place of assertive
behavior. With passive behavior one sacrifices one's own
basic rights and needs and desires through limited self-
expression. Passive people often lack friends, miss career
opportunities, and have low self-esteem. On the other
hand, people may get what they need or want, but they use
aggressive behavior to get it and thus hurt others in the
process. They often are not trusted by others, have
limited close relationships, and are often confronted with
anger from other people. Cognitive restructuring can help
people with problems in personal assertion because it
involves identifying dysfunctional thoughts that people
tell themselves and changing these thoughts to more
constructive, healthy messages. If you would like to know
more about this technique please listen to the following
message concerning its effectiveness in treating problems
in personal assertion.


121
constructive way to spend your free time and gives you
something interesting and useful to do.
This completes the message portion of our tape. We are
now interested in feedback about that message. Please turn
the page and begin answering the Assertion Survey leaving
the headphones on for further instructions.
(Silence lasting 2 minutes)
Now that you have completed your responses please turn
to the next page and read the instructions with me.
We would like to get an idea of the thoughts that
crossed your mind while listening to the tape. The next
page contains the form we have prepared for you to use to
record your thoughts and ideas. Simply write down the
first idea that comes to mind on the first line, the second
idea on the second line, etc. You should try to record
those thoughts and ideas as concisely as possible. A
phrase is sufficient. Ignore spelling, grammar, and
punctuation.
You will have three minutes to write down your
thoughts. We have deliberately provided more space than we
think most people will need to insure that everyone would
have plenty of room to write his/her ideas and thoughts.
So don't worry if you don't fill every space. Just write
down whatever your thoughts were while listening to the
tape. Please be completely honest and list all of the


105
11. I often don't know what to say to good
looking people of the opposite sex.
12. I do not like making phone calls to
businesses or companies.
13. I would rather apply for jobs by writing
letters than by going to talk to the people.
14. I feel silly if I return things I don't like
to the store that I bought them from.
15. If a close relative that I liked was
upsetting me, I would hide my feelings rather
than say that I was upset.
16. I have sometimes not asked questions for fear
of sounding stupid.
17. During an argument I am sometimes afraid that
I will get so upset that I will shake all
over.
18. If a famous person were talking in a crowd
and I thought he or she was wrong, I would
get up and say what I thought.
19. I don't argue over prices with people selling
things.
20. When I do something important or good, I try
to let others know about it.
21. I am open and honest about my feelings.
22. If someone has been telling false and bad
stories about me, I see him (her) as soon as
possible to "have a talk" about it.
23. I often have a hard time saying "No."
24. I tend not to show my feelings rather than
upsetting others.
25. I complain about poor service when I am
eating out or in other places.
26. When someone says I have done very well, I
sometimes just don't know what to say.


90
between issue involvement and argument quality such that
there would be greater distinction between strong and weak
arguments under conditions of high issue involvement than
under conditions of low issue involvement. The predicted
issue involvement X argument quality interaction was not
found to be significant for any of the dependent measures.
The second hypothesis made a similar prediction of a
significant interaction between ego-involvement and
argument quality such that there would be a greater
distinction between strong and weak arguments under
condition of high ego-involvement than under conditions of
low ego-involvement. Once again, the predicted interaction
was not found for any of the dependent measures.
The results do indicate that the strength of the
argument quality significantly differentiated scores on the
CFCR dependent measure and demonstrated a strong tendency
to differentiate scores on the attitude rating dependent
measure. These results, coupled with the significant
argument quality manipulation check, suggest that overall
the content (arguments) of a counseling message made a
major difference in its effectiveness in advocating
cognitive restructuring for treating assertiveness
problems. Also consistent with the ELM Model was the
finding that strong messages produced more positive
cognitive responses, whereas weak messages generated more
negative thoughts. For the counseling situation this


58
(TLT). The TLT is used to identify a person's subjective
thoughts or reactions to a topic. In the TLT subjects were
instructed to list the thoughts they had while listening to
a taped message on cognitive restructuring (CR).
Subsequently, they also rated whether their thoughts were
in favor of using CR, opposed to using CR, or neither in
favor of nor opposed to it. A general index of CFCR was
calculated for each subject by subtracting the number of
unfavorable thoughts from the number of favorable thoughts
listed and then dividing by the total number of relevant
thoughts yielding a ratio score (see Cacioppo and Petty,
1981) Higher CFCR scores indicated more positive or
favorable thoughts toward CR, while lower or negative CFCR
scores indicated less favorable thoughts or opposition to
the use of the CR technique. It was expected that under
conditions of high issue involvement individuals with high
ego-involvement would have lower CFCR than individuals with
low ego-involvement. It was also expected that under
conditions of low issue involvement individuals with high
ego-involvement would have greater CFCR than individuals
with low ego-involvement.
Cognitive responses toward the CR technique were also
operationalized in terms of outcome expectations and
behavioral intention ratings. The outcome expectation
asked subjects to indicate, "To what extent do you think
the cognitive restructuring technique would be beneficial


APPENDIX B
PERSONAL PROBLEM INVENTORY
WHAT TO DO: Read each sentence carefully. Write down on
each line whatever number is correct for you.
5 very much like me
4 somewhat like me
3 neutral
2 somewhat unlike me
1 very unlike me
1. Most people stand up for themselves more than
I do.
2. At times I have not made or gone on dates
because of my shyness.
3. When I am eating out and the food I am served
is not cooked the way I like it, I complain
to the person serving it.
4. I am careful not to hurt other people's
feelings, even when I feel hurt.
5. If a person serving in a store has gone to a
lot of trouble to show me something which I
do not really like, I have a hard time saying
"no."
6. When I am asked to do something, I always
want to know why.
7. There are times when I look for a good strong
argument.
8. I try as hard to get ahead in life as most
people like me do.
9. To be honest, people often get the better of
me.
10. I enjoy meeting and talking with people for
the first time.
104


44
information about either themselves (high involvement) or
another person (low involvement). She found that when
favorable information was provided, high involvement
subjects showed more attitude change than low involvement
subjects. When unfavorable information was provided, the
reverse was true.
In sum, research has produced contradictory findings
on the effects of level of involvement. To some degree
this has resulted from differing definitions of
involvement, but the contrary results have persisted even
when involvement was similarly defined. In hopes of
reconciling these opposing results, involvement will now be
examined in relation to cognitive processing.
Involvement and Cognitive Processing
From the ELM perspective Petty and Cacioppo (1979a)
propose that increasing involvement with an issue increases
one's motivation to cognitively process issue-relevant
information and can lead to either increased or decreased
persuasion. The research presented in the previous section
clearly demonstrates that high involvement can lead to
either enhanced or inhibited persuasion. Petty and
Cacioppo further postulate that whether persuasion is
enhanced or inhibited depends upon the individuals initial
opinion. If the persuasive message is contradictory to the
individual's initial attitudes, it is likely that the
individual is motivated and able to generate


77
rehearsal. Procedures for collection and evaluation of
data were standardized in order to prevent bias between
groups.
Analysis of data compared the four involvement groups
and argument quality. A 2 x 2 x 2 ANOVA for each cognitive
response variable determined the effects of ego-
involvement, issue involvement, argument quality, and
interactions.


69
Carr (1975) found RAS scores to be significantly correlated
(range -0.60 to -0.75) with interpersonal fears as measured
by the social factors from the Wolpe-Lang (1964) Fear
Survey. Other studies by Morgan (1974) and Hollandsworth
(1976) have found lower correlations between RAS scores and
a social fear survey. In Morgan's study of college
students the correlations ranged from 0.172 to -0.239 and
in Hollandsworth's study a correlation of 0.436 was found.
The combined data from these studies does appear to provide
some support for the construct validity of the RAS.
Concurrent validity of the RAS has also been
established in several studies. Rathus (1973) reports a
correlation of 0.705 between RAS scores and observer
ratings of verbal behavior in response to five questions
asking for assertive behavior. Frankel (1977) reports
significant correlations between the RAS and the Conflict
Resolution Inventory and between the RAS and the Assertion
Inventory. MacDonald (1975/1974) also supports the
concurrent validity of the RAS in her findings of a
moderate relationship between RAS scores and behavioral
measures of assertiveness. In addition, the RAS has been
used in treatment outcome studies and has been shown to be
a sensitive index of pretreatment to post treatment change
(Rathus, 1973; Blanchard, Turner, Eschette, and Coury,
1977) .


34
college students, while others were just told that the
editorials were a journalism class project. Personal
relevance was also manipulated in this study by telling the
students that certain college regulations would change next
year (high relevance) or ten years from now (low
relevance). The editorial message used with all subjects
consisted of five strong arguments for the institution of
senior comprehensive exams as a requirement for graduation.
Attitude measure results revealed a main effect for warning
and a warning X relevance interaction. The interaction
showed that warning significantly decreased attitude
agreement only under the high relevance condition. Since
the warning reduced agreement even though the arguments
used were strong, this suggests that it induced biased
rather than objective processing. As this effect was
stronger for high than low relevance conditions the warning
did not appear to function as a simple rejection cue but
rather as a motivator for counterargument and resistance to
change.
The common thread that runs through the operation of
both types of forewarning effects is the presence of some
organized structure of prior knowledge within the recipient
of the persuasive attempt. In the social psychological
literature a person's organized structure of knowledge is
referred to as a schema (Landman & Manis, 1983; Markus &
Zajonc, 1985). Markus and Zajonc (1985) state that the


7
processing tends to be biased toward the maintenance of the
guiding schema and may override the objective processing of
externally provided communications (Ross, Leper, and
Hubbard, 1975; Markus and Sentis, 1982). George A. Kelly's
(1955) psychology of personal constructs proposes, as the
fundamental postulate of the theory, that "a person's
processes are psychologically channelized by the ways in
which he anticipates events" (Kelly, 1955, p.46). A person
uses his/her personal construct system to interpret and
"anticipate" the world. Kelly's Organization Collorary
suggests that some constructs are more important and have
more implications than others. More important or
superordinate constructs may serve as schemas for the
anticipation of incoming information. They also are more
resistant to change (Hinkle, 1966/1965). Thus, if a person
has developed a schema or superordinate construct for the
dimension of assertiveness, that person's processing of
information concerning counseling techniques for treating
assertiveness problems may be biased.
Therefore, in this study two types of involvement will
be distinguished. The first is issue involvement. This
would correspond to the involvement seen in individuals
seeking counseling. It concerns the presence or absence of
perceived problems in assertiveness. High issue
involvement should lead to greater argument elaboration.
The second, which will be called ego-involvement, will


67
revised version of the Rathus Assertiveness Schedule
(Rathus, 1973). The SRAS is a 30-item self-report
assertiveness inventory, which provides a global rating of
perceived trait assertiveness. Respondents mark each item
in terms of how characteristic the behavior is of the
individual from 5 (very much like me) to 1 (very unlike
me). A total score is obtained by summing item scores
after correcting for reversed scoring weights.
In an investigation of the test-retest reliability of
the SRAS, thirty undergraduate psychology students were
administered the SRAS twice with a 7 day interval between
adminstrations. A Pearson product-moment correlation
yielded a coefficient of .91 (p <.0001).
McCormick (1984) provides evidence for a satisfactory
degree of equivalence between the Simple Rathus
Assertiveness Schedule (SRAS) and the Rathus Assertiveness
Schedule (RAS). He reports a mean interim correlation of
0.79 between the two versions of the test with all
correlations reaching statistical significance. The
correlation between the total scores of the two tests was
0.94. He also reports a correlation of 0.90 between total
odd and even item scores for both versions. Finally, in
his sample of 116 undergraduate students he found an almost
identical distribution of scores on both tests. Mean
scores and standard deviations for the two versions were
M=96.12, SD=23.85 (SRAS) and M=96.29, SD=24.94 (RAS).


32
(r = .22) need for cognition group. This would be expected
since high need for cognition subjects should be more
likely to deduce their attitudes from a careful
consideration of the central arguments in the proposal.
It is now apparent that numerous variables can affect
people's motivation and/or ability to process persuasive
arguments in a relatively objective manner. As noted
earlier, however, variables can also affect cognitive
processing in a relatively biased way.
Biased Processing
Studies of the impact of forewarning and prior
knowledge have provided evidence within the ELM framework
for the relatively biased cognitive processing of
persuasive messages. Forewarning has been seen to affect
an individual's motivation to process arguments in a biased
manner, whereas prior knowledge generally affects an
individual's ability to engage in biased processing.
Forewarning can be further distinguished as either warning
of message content or warning of persuasive intent.
In a study of the forewarning of message content, Petty
and Cacioppo (1977) manipulated argument quality,
forewarning, and also issue-relevant thinking. Details and
results of this study were presented earlier in the section
on ELM research in support of the cognitive response
approach. Briefly, if subjects engage in issue-relevant
thinking, they resisted the message whether they were


APPENDIX E
COVERT REHEARSAL TECHNIQUE
As a final part of this study, we would like for you to
take a minute to imagine a situation that required you to
talk in front of a large class. Imagine that you are
giving a ten minute presentation on the thing that you
value most in life. What we would like for you to do is
to take the next thirty seconds to conjure up the clearest
possible image that you can of that situation. Imagine a
picture of yourself, the room, the other students, your
professor, etc. Do this now.
Now that you have the situation in mind we would like
to know how clear it is for you. People often differ in
how sharp their mental pictures are, and we would like for
you to rate how clear your mental image is on the scale
below. Do this now.
1.
Very fuzzy and unclear
2.
Somewhat fuzzy
3.
Midpoint: not real clear but not
real
fuzzy
4.
Fairly clear
5.
Very sharp and clear.
Now
that you have indicated how clear
this
image
you, we would like to have you conjure up the image again.
This time concentrate on the things that you are saying to
yourself or thinking in the situation. In other words, as
you imagine yourself in this situation, what kind of
thoughts run through your head? During the next three
minutes we would like for you to write down every thought
that runs through your mind in this situation. As you did
before ignore spelling, grammar, and punctuation. Just
write down one thought per line on the following page.
Please turn the page now and begin.
112


33
forewarned or not. These results support the notion that
it is not so much the warning but rather the accessing of
attitude-supportive cognitions and the subsequent
processing of the persuasive message in light of these
cognitions which facilitates resistance. It appears that
the forewarning served to motivate subjects to begin
thinking about their beliefs and the information and
thoughts that they already had to support those beliefs.
In line with this reasoning, a content forewarning is more
effective when there is some time delay between the warning
and message to allow thinking (Hass & Grady, 1975; Petty &
Cacioppo, 1977).
Hass and Grady (1975) also report that a forewarning of
persuasive intent is just as effective when it immediately
precedes a message as when it comes several minutes before
a message. This suggests that an intent warning functions
differently from a content warning. They propose that an
announcement of an intent to persuade may arouse a
psychological state of "reactance" that motivates a person
to defend their freedom to hold a particular attitude (cf.
Brehm, 1972). Such a state could produce biased
processing. In a study designed to explore whether
warnings of persuasive intent produced biased processing
Petty and Cacioppo (1979b) told students that they would be
evaluating radio editorials. Some students were given a
warning that the editorials were intended to persuade


92
relatively primitive affective and associational processes.
It may be that those subjects who thought more about the
message associated it more with counseling and the common
notion that counseling is beneficial. Because a main
effect for issue involvement was not significant for the
CFCR measure, it does not appear that high issue
involvement subjects saw the cognitive restructuring
technique itself as being better than did the low issue
involvement subjects. In other words, greater favorability
toward the technique was related to more thinking in
general, but not to more favorable thinking per se.
This theoretical explanation further suggests some
methodological factors that may have affected the results
of the study. It may be that the high and low levels of
issue involvement in this study do not correspond to the
high and low levels of issue involvement of previous ELM
research reported in the social psychological literature.
In previous research involvement manipulations were
designed to induce relatively pure forms of the central and
peripheral routes to persuasion. For example, in the
Petty, Cacioppo, and Goldman (1981) study high involvement
subjects were told that the message they were to evaluate
had relatively important consequences for them (i.e., if
they did not pass the senior comprehensive exam that was to
be instituted the next year, they would not graduate). In
contrast to this, low involvement subjects were led to


concerning assertiveness on a modified version of the Role
Repertory Grid. It was predicted that there would be
greater distinction of argument quality under high ego-
involvement than under low ego-involvement. In addition,
it was predicted that under conditions of high issue
involvement, high ego-involvement would exert a greater
influence on the weak than strong arguments. Under low
issue involvement, high ego-involvement would exert an
equal influence on both strong and weak arguments.
Results of the 2x2x2 Analyses of Variance on
outcome expectation attitude ratings, behavioral intention
scores, a general cognitive favorability index and covert
rehearsal scores did not support the predictions. The
analyses revealed significantly higher attitude ratings for
the high issue involved group than the low issue involved
group. They also revealed significantly greater cognitive
favorability for strong than weak argument quality, and a
similar trend for attitude rating and argument quality. No
significant results were found for behavioral intention and
covert rehearsal measures.
These results suggest that message variables strongly
determine acceptance of counseling messages. The results
of this study were also discussed in relation to the
contributions of the Elaboration Likelihood Model to
counseling research, and future directions were suggested.
vi 1


93
believe that the proposal would have no personal
consequences for them. These two groups of subjects
represent relatively extreme levels of involvement. In the
present study, and in the counseling situation in general,
the issues of concern are applicable to almost everyone.
That is to say, most topics discussed in counseling (for
example, assertion, problems in personal relationships) are
by definition highly involving since such issues are only
presented if they are personally problematic concerns.
Therefore, the levels of issue involvement in the present
study might best be described as moderately high and
moderately low. Further support for a moderate level of
involvement came from the nonsignificant difference on
Rathus scores between high and low issue involvement
conditions. This could explain the current failure to
duplicate the interaction between issue involvement and
argument quality found in previous ELM research. This same
argument could also be extended to the ego-involvement
variable.
Another methodology question concerns the
operationalization of issue involvement in counseling-
related research. Past investigations have attempted to
manipulate issue involvement in a number of ways. These
attempts have included manipulation of clients' "perceived
need" or requests for help (Dixon & Claiborn, 1981; Heppner
& Dixon, 1978) and motivations for counseling (Heppner &


100
shows promise in providing a general framework from which
to understand counseling as a social influence process, the
results of the present study suggest that the application
of the ELM to the counseling setting may not be as fruitful
as basic ELM researchers believe. The nature of counseling
itself may place limitations on the ELM's usefulness in
understanding the social influence process as it occurs in
the therapy room.


66
correlation of 0.62 for a bi-polar implications grid.
While reliability for total variance scores is generally
low, this may be due to its sensitivity to construct system
change as the individual and/or situation changes. Total
variance scores have been shown to effectively discriminate
thought-disordered schizophrenics from normals and other
psychiatric groups (Bannister, 1962; Bannister and
Fransella, 1966).
Rather than using the total variance score, this study
uses the proportion of variance accounted for by the single
relevant construct. The psychometric properties of this
score are not yet established. It seems reasonable to say
that despite the instability of the overall variance in the
system, the proportion of variance accounted for by any one
construct might remain relatively constant. To test this
assertion 30 students from an undergraduate psychology
course were administered the RG twice with a 7 day interval
between administrations. Proportion scores for the
relevant construct were used to calculate a Pearson
product-moment correlation. This test-retest correlation
was .66, (p <.0001).
Personal Problems Inventory
The Personal Problems Inventory (PPI; see Appendix B)
was used to assess issue involvement concerning assertion.
It consisted of the Simple Rathus Assertiveness Schedule
(SRAS) (McCormick, 1984). The SRAS is an easier-to-read,


2
above counselor characteristics. Little attention has been
paid to the influence of counseling message variables or
client characteristics, particularly to those client
characteristics which may be associated with resistance to
change.
Research within the area of social psychology on
attitude change has followed a similar route of
development. Early work on attitude change explored
communicator variables (such as prestige, expertise, and
credibility), message content, and some focus on audience
features. However, with the rapid development of a
cognitive social psychology in the 1970s, attitude change
research expanded to include the consideration of
consistency, dissonance, and attributional processes in the
communication recipient (Jones, 1985). With this expansion
message variables and recipient characteristics were
established as major variables in attitude change research
(Heppner and Heesacker, 1982). Future research on the
interpersonal influence process in counseling needs to more
fully integrate the findings from this research in
cognitive social psychology.
One social psychological theory that has relevance for
counseling research is the Elaboration Likelihood Model
(ELM: Petty and Cacioppo, 1981; Petty and Cacioppo, 1986).
Petty and Cacioppo refer to elaboration as the extent to
which a person thinks about the issue-relevant arguments


87
Self-doubting statements. The analysis of variance
revealed no significant findings for the SD score.
Additional Analyses
Correlations computed between eight of the design's
variables using Pearson product-moment correlations yielded
several significant relationships. The correlations most
germane to the present study included those between the
need for cognition variable and the other variables in the
study. These analyses yielded only one significant
correlation between the need for cognition score and the
score on the Simple Rathus Assertiveness Scale (SRAS), r =
.372, pC.0001. This result suggests that as subjects
increase in need for cognition, they see themselves as more
assertive. Perhaps even more interesting is the
nonsignificant correlation between the need for cognition
variable and the total number of relevant thoughts
generated by the subjects. This nonsignificant correlation
does not support Cacioppo and Petty's (1984) notion that
individuals high in need for cognition would tend to think
more extensively about messages presented to them.
Correlations are presented in Table 6.


I certify that I
opinion it conforms
presentation and is
as a dissertation for
have read this study and that in my
to acceptable standards of scholarly
fully adequate, in scope and quality,
the degrqe of Doctor of Philosophy.
V
£11
g J,
Associate
imeyer, Chairme
Professor of Psychology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Associate Professor of Psychology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.
Mar^t Alicki
Assistant Professor of Psychology


Table 5. Means with Standard Deviaions in Parentheses for Attitude Rating and
Behavioral Intention Score for the Three-Way Intervention of Issue
Involvement, Ego-Involvement, and Argument Quality
Variables
Independent
Dependent
Issue Involement
Ego-Involvement Argument
Quality
n Attitude Rating Behavioral Intention
Score
High
High
Strong
11
4.73 (1.19)
4.36(1.57)
High
High
Weak
12
4.67(1.78)
4.25(2.05)
High
Low
Strong
13
4.77(1.48)
4.46(1.66)
High
Low
Weak
14
4.21(1.58)
4.36(1.45)
Low
High
Strong
15
4.40(1.35)
4.40(1.72)
Low
High
Weak
16
3.69(1.49)
3.94(1.65)
Low
Low
Strong
11
3.82(1.08)
3.91(1.04)
Low
Low
Weak
12
3.00(1.41)
3.33(1.97)
oo


51
On the other hand, client factors may also serve to
influence cognitive processing in a biased manner. An
individual may be so extremely involved in an issue that he
or she avoids thinking about it altogether (e.g., the
denial of a person with alcohol dependence). Processing
may also become biased in the service of one's own ego or
for self-protection (e.g., Greenwald, 1980). If a person
organizes his or her identity around being a certain type
of individual (e.g., an easy to get along with, always
ready to please, nonassertive person), he or she would
probably be less objective in considering therapeutic
messages designed to change that behavior. This would
especially be the case if this identity had provided the
individual with rewarding secondary gains in the past.
Such a client may be well prepared to generate cognition in
support of his or her present attitudes and behaviors and
actively counterargue any therapeutic communications aimed
at changing that basic identity. Thus, this biased
processing would lead to greater resistance to change.
Hypotheses
As outlined in the preceding sections of this review,
cognitive processing of persuasive messages may follow
either a central information-processing route (a careful
consideration of the message content) or a peripheral
information-processing route (a cue or decision rule basis
for persuasion). Within the more central route, cognitive


46
communication than to the counterattitudinal one. Message
direction showed no significant effects on cognitive
responses under low involvement conditions.
In the second experiment Petty and Cacioppo (1979a)
again varied involvement, but they presented only a
counterattitudinal message with either strong or weak
arguments. The message in this experiment concerned the
institution of senior comprehensive exams. The ELM
predicts that increased involvement will lead to more
issue-relevant thinking about the arguments presented in
the message. For the message with strong arguments, which
are difficult to counterargue, increased involvement should
be associated with more persuasion. For the message with
weak and easy-to-counterargue arguments, increased
involvement will be associated with decreased persuasion.
While high involvement may initially motivate an individual
to reject a counterattitudinal message, the enhanced
cognitive processing of the message should enable the
virtues and flaws of the arguments to be objectively
recognized. Results of the cognitive response measures
supported these hypotheses. Under high involvement
conditions subjects produced more favorable cognitions and
fewer counterarguments to the strong than to the weak
arguments. Argument quality produced no significant
effects on cognitive responses under low involvement
conditions. Furthermore, high involvement increased the


70
Assertion Survey
The Assertion Survey (AS; see Appendix C) is a
questionnaire developed by the researcher to assess outcome
expectations. It consisted of eight 7-point Likert-type
items. Outcome expectation was assessed by the item, "To
what extent do you think the cognitive restructuring
technique would be beneficial to you, personally?"
Behavioral intention was assessed by the item, "How likely
are you to use the cognitive restructuring technique in the
future in situations requiring assertion?" As a check on
the argument quality manipulation, subjects answered a
seven-point Likert item asking them to rate the quality of
the arguments in the intervention: very poor arguments (1)
to very good arguments (7). In addition, as a general
check on the issue involvement manipulation, subjects
answered another Likert item asking them the extent to
which the message has implications for them personally:
not at all relevant to me (1) to very relevant to me (7).
Ancillary items assessed other aspects of the message
including voice quality, speaker qualification, and ratio
of delivery.
Thought Listing Techniques
The Thought Listing Techniques used in this study were
variations of the "thought-listing procedure" developed by
Brock (1967) and Greenwald (1968). The procedure has been
refined to become a suitable self-report technique for


75
again allowing for three minutes to complete the task.
Finally, subjects were instructed to remove their
headphones, turn in their materials to the researcher, and
receive their experimental credit from the researcher.
All subjects participating in the study will be given
a subject number code which will identify their responses
while maintaining their anonymity. All subjects also
received either experimental credit or extra class credit
for their participation in the study.
Analyses of Data
The analyses compared the impact of high and low
levels of ego-involvement, issue involvement, and message
quality on attitudes toward cognitive restructuring.
Dependent variables included an index of cognitive
favorability in response to the taped messages, outcome
expectation attitude ratings, behavioral intention ratings,
and post-coding of covert rehearsal thoughts. A series of
2x2x2 analyses of variance were performed to determine
the effects of ego-involvement, issue involvement, argument
quality, and interactions on these variables. Duncan's
Multiple Comparison Procedure was used to evaluate the
effects predicted in the research hypotheses.
Summary
This study was designed to test and compare the
varying levels of ego-involvement, issue involvement, and
argument quality upon a subject's cognitive responses to


27
message consisting of weak arguments should produce less
agreement when consideration is high rather than low. By
manipulating argument quality along with another variable,
it is possible to assess whether that variable enhances or
reduces argument processing in a relatively objective or
biased manner. In relatively objective processing if the
variable increases processing, subjects' attitudes and
thoughts should be more clearly distinguished when the
variable is present rather than absent. In the case of
biased processing a variable will produce varying levels of
effects depending upon the direction of the bias and
argument strength. A variable that biases thinking in a
positive direction should generally have a greater
influence on a strong than a weak message, because it will
be more difficult for an individual to generate favorable
thoughts to weak than strong arguments. On the other hand,
a variable that produces a negative bias should have a
greater effect on a weak than a strong message. This
should occur because it will generally be more difficult
for an individual to generate counterarguments to strong
than weak arguments. Research utilizing the argument
quality manipulation has provided empirical evidence for
the effect of numerous variables upon the extent of
cognitive processing in both relatively objective and
biased manners. A sampling of this research will be
presented next.


39
past psychological research similar variables have been
alternately referred to as "issue involvement" (Kiesler,
Collins, and Miller, 1969), "personal involvement" (Sherif,
Kelly, Rodgers, Sarup, and Tittler, 1973), and "ego-
involvement" (Greenwald, 1980). This has occurred because
involvement can be judged in a variety of ways, such as the
degree to which an issue has "intrinsic importance" (Sherif
and Hovland, 1961), the centrality of the issue to a
person's self, and the number of personal consequences of
the issue.
Involvement became a central variable in research on
persuasion with the advent of social judgment theory
(Sherif and Hovland, 1961; Sherif, Sherif, and Nebergall,
1965). Social judgement theory views persuasion as a two-
stage process. First, one makes a judgment about the
position of a persuasive message in relation to one's own
position. Second, attitude change or no change occurs
depending upon the assimilation or contrast effects evoked
by the judged discrepancy between the message and one's own
position. Central to the theory are two assumptions: (1)
an individual's own stand on an issue serves as an internal
anchor for judging messages, and (2) the more "involved" an
individual is in the issue, the stronger the anchoring
effects of the initial opinion. Greater involvement thus
leads to more resistance to persuasion, because people were
postulated to hold broader "latitudes of rejection" as