Instructional use of the Internet

MISSING IMAGE

Material Information

Title:
Instructional use of the Internet stages of concern among faculty at the University of Florida
Physical Description:
xi, 111 leaves : ill. ; 29 cm.
Language:
English
Creator:
Toms, Sue Anne
Publication Date:

Subjects

Subjects / Keywords:
Instruction and Curriculum thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Instruction and Curriculum -- UF   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1997.
Bibliography:
Includes bibliographical references (leaves 86-91).
Statement of Responsibility:
by Sue Anne Toms.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 028074507
oclc - 37925289
System ID:
AA00014241:00001


This item is only available as the following downloads:


Full Text


















INSTRUCTIONAL USE OF THE INTERNET:
STAGES OF CONCERN AMONG FACULTY
AT THE UNIVERSITY OF FLORIDA














By

SUE ANNE TOMS













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


1997














This work is dedicated to the memory of

Ray Edward Toms (1922-1969) and of

Tawat Chotigeat (1921-1975).














ACKNOWLEDGMENTS


Any project of this magnitude is not executed as a solo

effort, but rather it represents the contribution of many to

its success. To those below, I offer my most heartfelt

appreciation:

Dr. Clemens L. Hallman, whom I have known since 1986

and who shared his enthusiasm about this project early on in

the process, for his continual support and encouragement.

Dr. Sebastian L. Foti, who always had the time to

listen and share his thoughts, even when he knew me as no

more than a stranger when I first arrived on campus, and who

demonstrated his unfailing loyalty throughout the course of

my studies.

Dr. James J. Algina for his kind guidance and endless

patience as I attempted to gain some mastery of statistical

analysis which includes skills similar to those of learning

a foreign language--holding many rules in mind

simultaneously.

Dr. Richard D. Downie for his assurance and the wisdom

of his many years on this campus and in the field of

international education.


iii








Dr. Constance L. Shehan for showing early on her

willingness to encourage this undertaking and for offering

the resources of the University Center for Excellence in

Teaching in support of this project.

Tosporn Chotigeat for not wanting me to go but not

wanting to hold me back either and for understanding why I

wanted to do this.

Ruth Evelyn Bishop Toms King for always telling me that

I could do anything that I set my mind to.















TABLE OF CONTENTS


ACKNOWLEDGMENTS..................................... iii

LIST OF TABLES...................................... viii

ABSTRACT ............................................ ix

CHAPTERS

1 INTRODUCTION.................................... 1

Statement of the Problem...... ................. 1
Need for the Study.............................. 2
Current Changes in Institutions
of Higher Education......................... 2
Teaching During These Changes................ 4
The Institutional Context .................... 7
Research Questions............................... 10
Limitations and Assumptions .................... 10
Definition of Terms............................. 11

2 REVIEW OF THE LITERATURE........................ 14

Adoption and Diffusion of Innovations........... 15
Beginning Teacher Concerns..................... 16
The Concerns-Based Adoption Model............... 18
Stages of Concern.............................. 19
Non-concern................................. 20
Self................................... .... .21
Task ....................................... 21
Impact ..................................... 21
Levels of Use ................................ 22
Innovation Configuration..................... 24
Intervention Taxonomy and
Intervention Anatomy........................ 25
Change Facilitator Stages of Concern and
Questionnaire... .......................... 25
Models Incorporating the Concerns-Based
Adoption Model.............................. 26
The Stages of Concern Questionnaire............ 29








Development of the Stages of Concern
Questionnaire ............................. 30
Psychometric Properties...................... 31
The Demographic Data Page ...................... 33
Instructional Use of the Internet
in Higher Education........................... 37
Summary......................................... 41

3 RESEARCH DESIGN................................ 43

Study Design/Procedures........................ 43
Variables.................................... 44
Sample......................................... 44
Instrumentation................................. 46
Development of the Questionnaire............. 46
Reliability/Validity......................... 47
Description of the Questionnaire............. 48
Data Collection................................ 56
Data Analysis................................... 59
Summary ........................................ 60

4 RESULTS AND DISCUSSION.......................... 62

Research Questions.............................. 62
Variables....................................... 63
Findings....................................... 64
Descriptive Statistics ....................... 64
Correlational Relationships.................. 70
Regression Analysis........................... 73
Research Question 1.......................... 73
Research Question 2........................... 75
Research Question 3.......................... 76

5 SUMMARY AND CONCLUSIONS......................... 78

Summary ........................................ 78
Findings....................................... 82
Implications ................................... 83
Recommendations for Further Research........... 84

REFERENCES.......................................... 86

APPENDICES

A STUDENT SENATE PRESIDENT'S LETTER TO ALL UF
FACULTY....................................... 93









B PERMISSION TO USE STAGES OF CONCERN
QUESTIONNAIRE .................................. 97

C INSTITUTIONAL REVIEW BOARD APPROVAL ........... 99

D STAGES OF CONCERN QUESTIONNAIRE................ 101

E TEXT OF INITIAL MAILING ....................... 106

F LETTER OF TRANSMITTAL: FIRST MAILING OF
QUESTIONNAIRE ................................ 108

G LETTER OF TRANSMITTAL: FOLLOW-UP MAILING OF
QUESTIONNAIRE ................................ 110

BIOGRAPHICAL SKETCH.................................. 111


vii















LIST OF TABLES


1 Demographic profile of the respondents....... 61

2 Peak stage of concern........................ 66

3 Levels of use of the Internet................. 67

4 How much teaching is modified based on
student learning............................. 68

5 Rank, gender, and national origin of
respondents................................. 69

6 Intercorrelations between peak stage of
concern and other variables................. 71

7 Summary of multiple regression for
variables predicting peak stage of
concern..................................... 74















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

INSTRUCTIONAL USE OF THE INTERNET:
STAGES OF CONCERN AMONG FACULTY
AT THE UNIVERSITY OF FLORIDA

By

Sue Anne Toms

August 1997


Chairman: Clemens L. Hallman
Major Department: Instruction and Curriculum

This descriptive correlational study explored the

patterns in the stages of concern of the faculty at the

University of Florida regarding the innovation adoption of

the Internet for instructional purposes. Recent

technological developments in computers and

telecommunications, especially in access to the storage and

transfer of vast quantities of information, have included

many possibilities for direct application to university

classrooms.

Three research questions were posed. What are the

relationships of the level of Internet use for instructional

purposes and the level of Internet use for all other

purposes to the sequence of stages of concern? Are there








significant differences in the peak stages of concern of the

faculty members grouped by the extent to which they modify

their instructional practices based on how or what students

learn? Are there significant differences in the peak

stages of concern among faculty members grouped by rank,

gender, age, or national origin.

The Concerns-Based Adoption Model provided the

theoretical framework for the study. Made up of several

dimensions, CBAM's seven stages of concern represent a

developmental sequence through which an individual passes

when confronted with change or innovation. The stages range

from non-concern (Awareness), to self (Information,

Personal), task (Management), and impact concerns

(Consequence, Collaboration, and Refocusing).

During Spring 1997 the Stages of Concern Questionnaire

was mailed to 1,650 faculty members within ten colleges at

the University of Florida. The final sample contained 540

responses, a return rate of 33%. The data were analyzed

using correlational and regression techniques.

Findings included significant correlations between the

peak or most intense stage of concern and level of use of

the Internet for instructional purposes, level of use of the

Internet for all other purposes, and attention to how

students learn. However, the multiple regression model

produced only two significant predictors of peak stage of

x








concern: level of use of the Internet for instructional

purposes and gender.














CHAPTER 1
INTRODUCTION


The history of modern education is littered with
the trash of technology left behind by unrealistic
purchases, naive users, and vendors working on a
quota system. (Polley, 1977, cited in Albright,
1996, p. 2)


Statement of the Problem

This quantitative study explored the patterns in the

needs thoughts, feelings, and perceptions of the University

of Florida faculty as manifested in their stages of concern

regarding the innovation adoption of the Internet for

instructional purposes. The sources of the problem are

several. The first involves the trend of traditional brick-

and-mortar colleges and universities investigating novel

modes of instructional delivery facilitated by recent

technological developments in computers and

telecommunications. Another stems from the customary

approach to faculty development within these organizations,

one which seldom considers at the level of the individual

the psychological orientation which occurs with expectations

for incorporation of these technological changes into their

daily teaching lives. Finally, the need originates in the










current tensions at universities across the nation between

incoming students' increasing levels of technological

sophistication and their demands for access to computers and

networked resources to supplement and enhance their own

learning. These tensions place unprecedented strains on

ever tighter budgets.


Need for the Study


Current Chanaes in Institutions of Hiaher Education

The innovation central to this study, use of the

Internet for instructional purposes, stands poised to

complement substantially the primacy of the model of higher

learning which has prevailed for many centuries in the

Western world. The rise of monasteries as centers of

learning produced the classroom model as we now know it.

Classrooms where an expert gathered with students and talked

to them to transmit knowledge represent the oral, pre-print

tradition (Warnock, 1996). Same time/same place for teacher

and student was the requirement in this transmission model

of teaching and learning.

The advent of the printing press brought about

repositories of knowledge in the form of libraries where

information could be stored through time on the substrate of

paper. Thus, students were able to learn from the written










works of past (even deceased) scholars, but they still had

to adhere to the 'same place' requirement, gathering where

these libraries were located. Knowledge transmission

continued to dominate.

In the present age, both time and place are losing

their importance. The availability and distribution of

information and with it knowledge are changing drastically.

The Internet represents a new medium for the storage and

transfer of information, both without the customary

substrate of paper (Warnock, 1996). Sources of information

distributed throughout networks accessible more widely than

ever before are becoming the norm.

Full realization of all that this means to the world of

students, scholars, and researchers will not be achieved for

some time to come. Nevertheless, the present represents a

key moment in the evolution of the spread of knowledge, the

likes of which have not been seen for many hundreds of

years. This is not to say that the current paradigm will be

replaced any time soon nor that it will ever disappear. One

fact is unarguable, however; new features will be added. As

the president of Harvard recently mused, "The world of

learning will not become lost in cyberspace, I suspect, any

more than it has drowned in books" (Rudenstine, 1997, p.

A48).










Teaching During These Chanaes

The process of "becoming a teacher is complex,

stressful, intimate, and largely covert" (Fuller & Brown,

1975, p. 75). College professors participate in this

process and find all of the above amplified. Furthermore,

the process is compounded for those who are expected to keep

abreast of changes in the teaching of their disciplines and

who are interested in incorporating what adoption of

instructional innovation offers.

The individuals in the front lines of instructional

innovation or change at post-secondary institutions, that is

the teaching faculty, are seldom trained in any systematic

way to become members of the professorate. They may have

taught as graduate students, but they usually have chosen a

career in the academic world due to an affinity for their

content area or for research. Those who feel a commitment

to teaching well and express an interest in improving their

teaching skills can typically avail themselves of the campus

resources at hand. Among the options at the University of

Florida is the University Center for Excellence in Teaching.

UCET's primary mission is to assure access to the

resources needed for teaching by both new and established

faculty. Its major functions include supplying new faculty

with "orientation, instruction, and consultation about

teaching and learning" (Shehan, 1994, p. 1). To this end,








5

its Advisory Board members act as consultants or partners to

new faculty to facilitate development of their teaching

styles and philosophies (Shehan, 1994).

In addition, another charge of UCET is to facilitateae

the continuing development of tenured faculty as teachers"

(Shehan, 1995, p. 4) The center offers information about

developments related to teaching at the higher education

level through its Conversations about Teaching series and

the Focus on Teaching annual workshops. Finally, UCET

contributes to the preparation of graduate students for

their roles as future educators (Shehan, 1995, p. 4).

Two examples of the application of instructional uses

of the Internet advocated by UCET are found in a recent

issue of the center's newsletter in an article on teaching

large classes. The author offers eight suggestions, among

them encouraging others to developlp a homepage so that

students can get important information from there rather

than calling you" and encourageae the use of e-mail as a

way for students to get their questions answered. That way

you can respond at your convenience and save time"

(Schwartz, 1996, p. 2).

Although prominent among the topics recently offered in

UCET activities are diversity in the classroom and dealing

with individual differences in learning styles, Loucks and










Hall (1977) make the valid point that when teachers or

faculty are trained in some new instructional innovation,

they are often not treated as diverse individuals themselves

(p. 18). Faculty development is seldom carried out on the

basis of the needs and concerns of the faculty. For staff

developers whose clients teach across a myriad of

disciplines, in class groups of from 7 to 700, to students

who are fresh on the university scene or those who may have

been there nearly a decade, making instructional

interventions relevant is indeed a challenge.

This study speaks directly to this need. It yields

concerns data collected at the level of the individual

regarding adoption of the innovation of using the Internet

for instructional purposes using the Stages of Concern

Questionnaire, part of the Concerns-Based Adoption Model.

The model was devised to depict "the process of change in

terms of the individuals involved" (Loucks & Hall, 1977, p.

18). Hall and Loucks (1978) list the six basic assumptions

that underlie the model: (1) rather than an event, change is

a process; (2) facilitating change involves the individual

as the "primary target"; (3) change occurs at a very

personal level; (4) the change process occurs along a

pathway marked by stages in both thoughts and skills; (5)

interventions by staff developers are most effective when

they focus on the client and follow a "diagnostic/










prescriptive model"; and (6) the effectiveness of staff

development will depend on constant diagnostic input and on

systematic adaptation of the intervention as well as of the

innovation itself (pp. 37-38).

Within the CBAM structure are found basic frames of

reference, of which the concerns dimension is one. Concerns

have been defined as "perceptions, feelings, and

motivations" (Hall & Loucks, 1978, p. 39) which beset an

individual faced with change. Interventions based on

individuals' concerns not only facilitate the process of

adoption of an innovation, they also increase the

effectiveness and efficiency of the adoption and decrease

the trauma of the change for the individual. In short,

information on individuals' stages of concerns provides a

powerful tool for improving staff development interventions.


The Institutional Context

The final source of the problem at hand involves the

context within which the study was set. Two principal

factors constitute the pertinence of the study's outcomes to

the conditions regarding adoption of the Internet for

instructional purposes at the University of Florida.

The first of these factors revolves around the student-

driven movement to assure the accustomed access to and use

of the Internet without students incurring additional costs.








8

Until spring semester 1996 all UF students (even members of

the general public) had access to computer labs on campus

and to the full range of applications and networking offered

without charge. Starting with that semester, however,

undergraduate students were assessed a charge of $20 for

maintenance of an account which included basic Internet

access (they still could use word-processing, spreadsheet

and database applications free of charge). At that point,

the administration made a distinction between what was

considered essential computer access and what was considered

elective, a dichotomy with which the student body could not

agree.

The uproar caused by the imposition of the fee finally

resulted in a November 19, 1996, letter from the UF Student

Senate President to all faculty including the full text of a

resolution passed unanimously by the Senate expressing the

students' view: "the use of computers as an essential,

rather than elective, element of fulfilling educational

goals" (C. E. Dorworth, communication, November 19, 1996).

In addition, the letter went on to ask that all faculty

include the following in the syllabus for each course they

teach:










It is the formal policy of this class that in
order to fully and properly fulfill the
requirements of this course some use of and
proficiency in the use of computers, including
access to and use of the Internet (e-mail and
World Wide Web), will be required. (C. E.
Dorworth, communication, November 19, 1996)

Clearly, adoption of the innovation of the Internet for

learning purposes among the students represented by their

Student Senate had out paced the ability of the

administration to provide access to it within the given

budgetary constraints. (For full text of letter and

resolution, see Appendix A.)

The tensions fueled by student demands on limited

resources were not unique to the University of Florida

campus; they were also apparent at the level of the (state)

Board of Regents during the fall of 1996 and into 1997.

Despite student protest and after debating proposals as high

as $100, at their meeting on January 24, 1997, the Board

approved a recommendation that the legislature vote to

assess a technology fee of $50 per student throughout the

State University System (Malernee, 1997). This represents a

potential inflow of approximately $2,000,000 in the

technology budget for the University of Florida. Making a

contribution to a well-informed decision on allocation of

this money to avoid the "trash of technology left behind by

.naive users" (Polley, 1977, cited in Albright, 1996,










p. 2), as mentioned at the outset, represents the second

factor in the significance of the outcomes and the single

greatest need for this study.


Research Ouestions

What are the concerns regarding adoption of the

Internet for instructional purposes among the faculty at the

University of Florida? To explore this question, the

following were answered:

(1) What are the relationships of the level of Internet

use for instructional purposes and the level of Internet use

for all other purposes to the sequence of stages of concern?

(2) Are there significant differences in the peak

stages of concern of the faculty members grouped by the

extent to which they modify their instructional practices

based on how or what students learn?

(3) Are there significant differences in the peak

stages of concern among faculty members grouped by rank,

gender, age, or national origin?


Limitations and Assumptions

The findings offer a one-time view of a dynamic

process--the respondents' moves through the developmental

stages of concern about an innovation. The subjects were

instructional faculty affiliated with an academic unit

within the ten college-level units which offer undergraduate










education at a large southeastern U.S. research university.

Generalizibility is limited to those who completed the

survey.

The data were collected on a survey undertaken by the

University Center for Excellence in Teaching. The quantity

and quality of the responses may be influenced by this fact.

Those who have had experience with and benefitted from UCET-

sponsored activities may have been more likely to reply.

Since response to the questionnaire was voluntary,

response rates may have been higher among those with a

higher-than-representative degree of interest in and/or

experience with use of the Internet for instructional

purposes.

The results reflected the respondents' honest concerns

and are valid to the extent that the respondents share the

definitions of the terms used with those of the researcher.


Definition of Terms

For the purposes of this study, the following

definitions have been used:

The Internet is an international network linking

smaller networks; it includes electronic mail and file

transfer capabilities, as well as access to information

(text, graphics, audio, video) throughout those networks

(Oblinger, 1992).








12

Instructional Durnose refers to a use (either required

or optional) to support classroom routines, e.g., course

announcements, distributing hand-outs, sending or receiving

assignments, and research for term projects.

Concern is the mental construct represented by an

individual's feelings, preoccupations, thoughts, and

considerations directed at a specific task to be

accomplished or an issue to be resolved. This definition

follows that developed by Hall et al. (1979).

Stages of concern constitute a sequential set through

which a person passes when confronted with change.

Stages of Concern Ouestionnaire (SoCO) is an instrument

with thirty-five Likert-type items yielding scores on seven

subscales to measure the intensities of concern of educators

about the adoption of an educational innovation (Hall et

al., 1979).

Peak stage of concern is that stage with the highest

score and thus the most intense concern. A given individual

may exhibit varying levels of concern on the multiple items

measuring within the same stage, but an overall score for

each stage will show the stage of greatest intensity. In

the event of equal scores on two or more stages, the stage

furthest forward on the continuum, and thus in the process

of adoption innovation, was used.










Concerns Based Adoption Model (CBAM) is a conceptual

framework for understanding how individuals in educational

settings react to change. It posits that change is a

process, that it occurs at the level of the individual and

should therefore be measured at that level, and that

individuals confronted with change pass through a

developmental sequence of stages of concern as they adopt

that change. The stages of concern are one dimension of the

model (Hall et al., 1973).

Faculty refers to those persons employed at the

University of Florida during spring semester 1997 who are

remunerated as faculty within one of the following ten

college-level units: Architecture, Business Administration,

Education, Engineering, Fine Arts, Health and Human

Performance, Health Professions, the Institute of Food and

Agricultural Sciences, Journalism and Communications, and

Liberal Arts.

Foreign faculty are those of the above group who are

not native-born American citizens. They may be naturalized

American citizens, permanent residents, or holders of a visa

which permits them to work in the United States.















CHAPTER 2
REVIEW OF THE LITERATURE


The literature review for this study opens with a

summary of the findings of the two antecedents to the

theoretical framework central to it--research on the

adoption and diffusion of innovations and on beginning

teachers' concerns. It then turns to the Concerns-Based

Adoption Model, explaining its various dimensions and

tracing the development of the first one, Stages of Concern.

Examples of models which have incorporated the CBAM are

given. The review continues with a close look at the survey

instrument which measures the first dimension, the Stages of

Concern Questionnaire, including how it was developed and

its psychometric properties. This is followed by a review

of the literature concerning the variables captured on the

demographic page which makes up part of the instrument used

for this study. Finally, the research on current practices

of using the Internet for instructional purposes in higher

education settings is reviewed.










Adoption and Diffusion of Innovations

Within the vast body of research and theory on

innovation adoption, of direct relevance to this study is

the work of Everett Rogers. By synthesizing the findings

and theories from over 500 publications in his 1962 work

Diffusion of Innovations, he made a lasting contribution to

this area and significantly added to what is known about

adoption of an innovation.

The basic elements of diffusion of any innovation are

(a) the innovation, (b) its communication from one

individual to another, ( c) the social system, and (d) a

period of time (Rogers, 1962, p. 12). While diffusion

operates at the level of the social system or across

systems, adoption as a process does so at the level of the

individual. Rogers (1962) defines adoption as "the mental

process through which an individual passes from first

hearing about an innovation to final adoption" (p. 17).

Rogers segmented the adoption process into five stages:

awareness, interest, evaluation, trial, and adoption.

Through these five stages, the individual moves from a point

of random exposure to the innovation, to actively seeking

information about it, weighing its possible benefits, trying

it out on a limited basis, and finally adopting it.

Although rejection may truncate the process at any stage,








16

the outcome of the fourth stage (trial) is crucial to moving

to the final stage (pp. 81-86).

Rogers also typified adopters by the sequence in which

they adopt an innovation and their salient characteristics.

He termed those at the lead in the process "innovators" who

are venturesome; the next group are called "early adopters"

and are respected by others; the following "early majority"

are deliberate; the "late majority" are skeptics; and the

final "laggards" are traditional (Rogers, 1962, pp. 168-

171).


Beainnina Teacher Concerns

Central to Frances Fuller's research (1969) was an

exploration of the mental processes of an individual who is

confronted with change manifested in the demands of learning

to practice a future profession. Her pioneering studies on

undergraduate education majors dealing with the mismatch

between (a) their own motivation to learn and (b) the

content and sequence of the academic courses that teacher

educators had deemed appropriate and necessary for them

(1969) have made a lasting contribution.

To improve the match between learner goals and teacher

preparation programs, more information was needed about the

concerns of future teachers. An analysis of previous

studies about beginning teachers' concerns revealed










surprisingly consistent patterns, although missing were

concerns over the topics covered in the typical

undergraduate teacher-training curriculum (Fuller, 1969, pp.

208-210).

During counseling sessions with student teachers in

Fuller's first study, two categories of concerns emerged:

those with the self and those with pupils. Furthermore, the

preponderance of topics in the two categories was not

scattered about randomly during the eleven weeks of

counseling sessions, but rather a developmental pattern from

self to pupils was uncovered. In her second study, written

responses to open-ended concerns statements with student

teachers lent validity to the self-other dichotomy and its

sequential attribute. By reviewing findings of previous

studies through this dichotomouss' lens, as well as by

looking at (then) contemporary unpublished data from

colleagues, Fuller found further support for the two

categories and their developmental nature (1969).

Attempts to categorize the "amorphous and vague"

responses about concerns from education majors before their

student teaching experience, i.e., during their sophomore

and junior years, proved more elusive (Fuller, 1969, p.

219). As a result, a pre-teaching phase with non-concern

was added at the front of the two-phase developmental model










of teacher concerns conceptualized as self-->other. Among

the implications for further research, Fuller questions the

validity of generalization to other groups, including

college professors, and the feasibility of developing an

instrument for measuring concerns, thus opening the way to

her successors.

The timelessness of Fuller's findings is demonstrated

by the appearance of articles decades later exploring the

development of professors' careers as teachers. Kugel

(1993) proposes three major stages that professors move

through as they hone their teaching abilities. The first

centers on the self, the second on the subject matter at

hand, and the third on the student. The parallel between

such a model and the self-->other model of Fuller is

evident.


The Concerns-Based Adoption Model


While the legacy left by Rogers's work was much

knowledge in understanding innovation diffusion and adoption

at the level of social interaction and that of Fuller was a

glance into the motivation, thoughts, feelings, and

perceptions of beginning teachers, the Concerns-Based

Adoption Model or CBAM developed by Gene Hall and his

associates had as its mission "the study of what happens to

the individual classroom teacher and professor involved in










change" (Hall, 1976, p. 22). The CBAM, in conjunction with

the Procedures for Adopting Educational Innovations Project

at The Research and Development Center for Teacher

Education, University of Texas, addresses innovation

adoption within an educational setting not as an event, but

rather as a long-term process from the perspectives of the

individual, the organization, and the innovation (Hall et

al., 1975; Loucks & Hall, 1977). Goals of the model include

describing developmental changes in individuals as they move

through the process and effective intervention strategies

which can facilitate that movement. The basic diagnostic

frame of reference in the model includes the dimensions of

(a) the concerns that users have prior to and during

adoption of an innovation (Stages of Concern), (b) the way

in which the innovation is actually used (Levels of Use),

and ( c) how the innovation itself adapts to the exigencies

of the adopters (Innovation Configuration). In addition,

the fourth component prescribes intervention parameters to

facilitate the change process (Intervention Taxonomy and

Intervention Anatomy) (Hord & Loucks, 1980).


Stages of Concern

The CBAM project identified seven stages of concern

about an innovation (SoC). These stages expand on those

first proposed by Fuller and are grouped into self, task,










and impact concerns. Concerns are considered to be

manifestations of what an individual involved in the change

process is feeling, a construct referring "to the

categorization of expressions stated by the user related to

his use of the innovation" (Hall et al., 1973, p. 14).

Originally, the seven stages of concern were described as

(0) unaware, (1) awareness, (2) exploration, (3) early

trial, (4) limited impact, (5) maximum benefit, and (6)

renewal (Hall et al., 1973). Whereas the original works had

identified the concerns of preservice teachers, Hall (1976)

presents findings derived from a study that included

inservice teachers and professors in college settings.

Additional research resulted in defining the concerns more

precisely, developing a measurement of the stages of

concern, and testing the concerns across innovations (Hall &

Rutherford, 1976; Hall et al., 1977). Additional

applications of the concerns model to the change process are

reported among university administrators, state education

agency officials, and those working in the private sector

(Hall, 1979, 1985). Refined definitions of the stages of

concern (from Hall et al., 1979 [original concept from Hall

et al., 1973]) are given as

Non-concern--0 Awareness. Little concern about or

involvement with the innovation is indicated.










Self--l Informational. A general awareness of the

innovation and interest in learning more detail about it is

indicated. The person seems to be unworried about

himself/herself in relation to the innovation. S/he is

interested in substantive aspects of the innovation in a

selfless manner such as general characteristics, effects,

and requirements for use.

2 Personal. Individual is uncertain about the demands

of the innovation, his/her inadequacy to meet those

demands, and his/her role with the innovation. This

includes analysis of his/her role in relation to the reward

structure of the organization, decision making, and

consideration of potential conflicts with existing

structures or personal commitment. Financial or status

implications of the program for self and colleagues may also

be reflected.

Task--3 Manaaement. Attention is focused on the

processes and tasks of using the innovation and the best use

of information and resources. Issues related to efficiency,

organizing, managing, scheduling, and time demands are

utmost.

Impact--4 Conseauence. Attention focuses on impact of

the innovation on student in his/her immediate sphere of

influence. The focus is on relevance of the innovation for

students, evaluation of student outcomes, including








22

performance and competencies, and changes needed to increase

student outcomes.

5 Collaboration. The focus is on coordination and

cooperation with others regarding use of the innovation.

6 Refocusing. The focus is on exploration of more

universal benefits from the innovation, including the

possibility of major changes or replacement with a more

powerful alternative. Individual has definite ideas about

alternatives to the proposed or existing form of the

innovation.


Levels of Use

The second diagnostic component of the CBAM, levels of

use of the innovation, explains actual behaviors of the

innovation user without regard to such affective variables

as attitudes, motivation, or needs (those which are central

to the stages of concern) (Hall et al., 1975, p. 52). In

the broadest sense, stages of concern move from self to task

to impact whereas levels of use follow an orienting to

managing to integrating path (Loucks & Hall, 1977). The

eight levels of use describe the behaviors of the user (as

opposed to the feelings of the stages of concern) ranging

from non-use to orientation, preparation, mechanical use,

routine, refinement, integration, and renewal (Hall et al.,








23
1975; Loucks & Hall, 1977). These levels have been compared

to those passed through by a person learning to drive an

automobile (Hall & Hord, 1987, pp. 17-18). Assessment of

individuals' stages of concern and levels of use is

considered essential to choosing appropriate modes of

intervention in order to maximize the effectiveness and

efficiency of the change process and minimize the trauma of

that process on the individual. Interventions must be

coordinated with adopters' levels of use to be of maximum

benefit (Hall et al., 1975, p. 56). In addition, multiple

cycles of use make up the innovation adoption process and

intervention support must fall across these cycles.

Hall and Hord (1984) discuss the complex relationship

between stages of concern and levels of use, stating that at

the two extremes there seems to be a positive, linear

relationship, i.e. an individual whose peak stage of concern

falls at or near either extreme of the seven-unit SoC

continuum will tend to also be at or near that same extreme

level of use on the eight-unit LoU continuum. The

relationship between the two dimensions for those who fall

in the middle on one or both, nevertheless, is much less

clear. The overall hypothesized relationship follows an

elongated S-curve where "at early points in the change

process, use tends to 'drive' concerns, and at later points,










aroused concerns push Levels of Use" (Hall & Hord, 1984, p.

338).


Innovation Configuration

The third diagnostic dimension of the CBAM model

involves clarifying and operationalizing what use of the

innovation in its intended setting means (Hall & Loucks,

1981). The Innovation Configuration acknowledges an

interaction between the innovation user and the innovation

itself. By means of a checklist, it assesses user patterns

from among the components a user may or may not choose to

adopt. Each individual has a unique 'fingerprint' which

shows the pattern of that person's encounter with the

innovation. It will include the components selected, their

organization, and their variations (Yessayan, 1991).

Implications for change facilitators are clear; once

acceptable components and variations have been identified,

guiding users along the path of adoption is greatly

simplified. In addition, adopters evaluate their progress

and gauge the distance left to complete the change process.

Rutherford outlines the categories of components to an

innovation configuration: those which are crucial to the

innovation (musts), those which are desirable, and those

which are suitable but not required (1986, pp. 10-13).











Intervention Taxonomy (IT) and Intervention Anatomy (IA)

Later research on the CBAM has resulted in a final two-

part dimension which must also be mentioned. With an eye

toward more effective interventions for the successful

adoption of an innovation, the Intervention Taxonomy and

Intervention Anatomy were developed. The IT facilitates

comprehensive intervention planning by specifying six levels

for school improvement and staff development. The IA

addresses the internal components of potential interventions

and yields information which, when used with the IT,

contributes to more effectively addressing the individuals'

needs (Hord & Loucks, 1980).


Chance Facilitator Stages of Concern and Ouestionnaire

A more recent addition to the body of theory and

instruments to grow out of the original CBAM is the Change

Facilitator Stages of Concern dimension and its

corresponding CFSoC Questionnaire. Administration of the

original SoCQ to those charged with managing or facilitating

change proved inappropriate. For example, educational

supervisors are typically not concerned about use of the

innovation as adopters (teachers) are but rather with

realizing their role in helping the adopters use the








26
innovation. As a result, the CFSoCQ was developed (Hall et

al., 1991).

This instrument was used at the beginning and end of a

one-month residential training program for secondary school

science department heads in the Philippines during the

summer of 1991. Since the administrative structure of

educational authority places responsibility for

instructional leadership with these individuals, their role

is that of change facilitator within their own schools

(Matthews, 1993). The change facilitator concerns data

collected at the beginning was used to aid in the design of

the staff development program. Post-program data on the

change facilitator participants' stages of concern showed

significant progress through the developmental stages.


Models Incorporatina the Concerns-Based Adoption Model

Building on CBAM for the enhancement of staff

development interventions, McCarthy (1982) has included the

stages of concern in her novel approach to teacher

inservice. martini with modes of perception from thinking

to sensing/feeling along one axis, and adding modes of

processing from doing to watching along another axis, she

derives the four major learning styles. She then dissects

each learning style quadrant into two halves to portray










right-brain/left-brain dominance, and finally superimposes

the seven stages of concern (the first six inside the now

circular diagram and the seventh--refocusing--leaving the

circle). Her trademarked system called 4Mat accounts for

the affective stages of concern and guides a staff developer

to meeting the "needs of all four major learning styles,

while using right and left mode techniques" (McCarthy, 1982,

p. 20).

The stages of concern of the CBAM have also been

proposed in a model for faculty development specifically

aimed at technological innovations (Wedman & Strathe, 1985).

The authors advocate diverse strategies of faculty

development derived from three dimensions: the concerns

dimension (which they have simplified into four levels--

information, exploration, utilization, and

collaboration/innovation), the organization dimension

(individual, groups, departments, college), and the faculty

context dimension (instructional, creative, management,

personal). The result is a framework which may serve as an

alternative to the traditional "spray and pray" (Wedman &

Strathe, 1985, p. 19) approach to faculty development.

The usefulness of applying data from the first two

diagnostic dimensions of the Concerns-Based Adoption Model

to the design of inservice training is demonstrated in a










series of articles by Cicchelli and Baecher. Their 1987

work reports on 18 senior high school teachers undergoing

inservice regarding the use of microcomputers in the

classroom. Data gleaned from the stages of concern

questionnaire and levels of use interviews (of three users

and three non-users) served to guide the design and

implementation of 15 hours of training over three days.

Post-training administration of the SoCQ showed "a

significant change in teachers' concerns towards

microcomputers" (p. 85). In a similar study also involving

the implementation of computer technology at the high school

level, six teachers filled out the SoCQ to supply

information used in a nine-hour inservice program. When

administered the SoCQ afterward, the teachers showed a

difference of at least 10 percentile points in each stage

score (1990).

Cicchelli (1990) enlisted stages of concern theory when

confronted anew with Fuller's conundrum of some two decades

before about the best content and sequence of courses for

teacher education programs. The Fordham University Pre-

Service Program enrolls both fourth-year undergraduate and

fifth-year graduate students in a year of

professional/technical teacher training. The difference

between the two groups is in the additional training in










liberal arts which the fifth-year students have had. The

SoCQ was administered to 24 fourth-year and 17 fifth-year

students at the beginning and again at the end of the two-

semester preservice program. Liberal arts training as a

part of teacher preparation programs would be supported if

the fifth-year students were found to be further along in

the developmental sequence of stages of concern at the

beginning or at the end of their year of professional

teacher training, i.e., if they demonstrated lower self

concerns and/or higher impact concerns. This, however, was

not the case when looking at the two groups. At least on

the personal dimension of concerns about teaching, the

additional training in liberal arts cannot be linked to a

more advanced starting point on or greater movement along

the SoC wave-like pattern either before or after a one-year

program of professional/technical education training

(Cicchelli, 1990, p. 45).


The Stages of Concern Ouestionnaire

The stages of concern are assessed on a questionnaire

of thirty-five items where respondents indicate the

relevance of the statements to their present concerns

regarding an innovation. Use of the SoCQ results in

"standardized, individualized data that can be aggregated








30

and used to facilitate, monitor, plan, and communicate about

a change process" (Hall & Hord, 1987, p. 333). This

questionnaire was developed and copyrighted in 1974 by The

Research and Development Center for Teacher Education at The

University of Texas at Austin as part of the Procedures for

Adopting Educational Innovations/CBAM Project (Hall et al.,

1979).


Development of the Stages of Concern Ouestionnaire

The conceptual basis for developing a questionnaire of

concerns about adoption of an innovation grew out of the

extensive work of Fuller and others during the 1960s. To

develop this "quick-scoring, paper-pencil instrument" (Hall

& Rutherford, 1976, p. 228), Hall and his associates

elicited responses to an open-ended concerns statement,

forced rankings, Likert scales, adjective checklists, and

interviews; and from a total of 544 Q-sorted items, a group

of 400 were judged as central (by a minimum of six out of

ten judges) to the measurement of the seven concerns derived

from the definitions espoused in the original 1973 CBAM

paper (Hall et al., 1973, 1979, p. 9). From these, editing

and eliminating redundant items left 195.

Hall and Rutherford (1976) report on the administration

of the 195-item pilot instrument to elementary teachers










involved with teaming and college faculty involved with

instructional modules (each group's members stratified by

their length of experience with the innovation in hopes of

capturing information from people all along the full

continuum of stages). Data from the 366 respondents were

factor analyzed, and the first seven resulting factors were

found to be congruent with the outcome of the previous Q-

sort. The resultant 35 items were produced when the factors

were rotated in the direction of the originally defined

stages (factor loadings exceeding 0.5) (Hall & Rutherford,

1976). Between 1974 and 1976, the shortened instrument was

administered to educators involved with eleven different

innovations for results in both cross-sectional and

longitudinal studies (Hall et al., 1979).


Psychometric Properties

The reliability of the SoCQ is interpreted from the

reported test-retest correlations for each of the seven

stages (n = 132, two weeks elapsed) of from 0.65 to 0.86

(four of the seven correlations were above 0.80) (Hall et

al., 1979). The alpha coefficients of internal reliability

of the stages for the same group range from 0.80 to 0.93

(Hall & Rutherford, 1976). Data from a stratified sample of

teachers and professors numbering 830, however, show alpha








32

coefficients for the seven stages of from 0.64 to 0.83 (six

of the seven correlations above 0.70)(Hall et al., 1979).

Indications of the validity of concerns theory and the

instrument include a report of the correlations between

individual items and the stage which they supposedly measure

(72% of the items had higher correlations with their

assigned stage than with other stages), lower correlations

between distant stages than between adjacent ones, results

of factor analyses which indicate independence of the

individual stage subscales, and multi-method procedures

using the open-ended concerns statements, surveys, and

interviews--all resulting in multiple Rs greater than .56 (p

< .05) for four of the seven stages--and longitudinal

studies demonstrating the wave-like pattern of movement

through the stages (Hall et al., 1979).

The 1979 manual for scoring and interpreting the SoCQ

calls to the attention of the potential user of the

instrument a few of its limitations. First among these is

the purpose of its use; having been devised for diagnostic

functions, Hall et al. (1979) point out that any use which

might involve "screening or evaluation" or even judgement

about personality would be inappropriate (p. 57). Secondly,

the age group on which the instrument was normed was adult,

and their occupational affiliations were with educational










institutions; use of the SoCQ with persons outside these

groups would invalidate the claims of reliability and

validity. Finally, the quality of the results of the

analyses is dependent on the good will of the respondents as

they fill out the questionnaire.


The Demographic Data Page

In addition to the thirty-five items designed to gather

data on the concerns of the individual respondents to the

SoCQ, a demographic data page is also included. The

information collected from the responses to these items

serves a two-fold purpose: description of the survey sample

and comparison of it to the population and input for

statistical analyses.

Results of research reported by Hall et al. (1979)

report "no outstanding relationships between standard

demographic variables and concerns data" (p. 52). One of

his associates, William Rutherford, concurs by mentioning

age, gender, and years of teaching experience as lacking any

"consistent relationship" between them and concerns stages

scores (1986, p. 7).

In their research article on student engagement of 90

developmental reading students at El Camino College, Marsh

and Penn (1988) also report no relationship between stages










of concern and demographic variables. Numerous other

studies report no significant differences in SoC scores

according to demographic variables (Shoemaker, 1990; Falvo,

1990; Hickox, 1994; Lewis, 1994).

Exception to the lack of findings related to age is

taken by Lee-Kang (1993), who found that undergraduate

textiles, clothing, and merchandising program departments

with a greater number of males over sixty were associated

with a lesser degree of adoption of the instructional use of

computers.

A study by Wells and Anderson (1995) which looked into

the relationship between gender (and various levels of

computer knowledge) and SoC scores before, during, and after

a course called Computers and Telecommunications in

Education for graduate students majoring in education at

West Virginia University found that gender was associated

with significant differences at the final administration of

the questionnaire on Stage 0 Awareness and at the midpoint

administration for Stage 5 Collaboration and Stage 6

Refocusing. However, since details about the sample size

and gender breakdown are not reported, these findings are

difficult to evaluate.

A review of the research on the variables of rank and

tenure also finds claims on both sides of the innovation










adoption. King (1990) claims that "junior faculty work on

instructional innovation at their peril" (p. 297), yet Lee-

Kang in the 1993 study cited above found that more assistant

professors in a program increased the likelihood of

implementation of instructional computer use at higher

levels.

In contrast to the standard demographic variables,

those which make up the constellation termed the "state of

the user system" by Hall et al. (1979) seem to be more

crucial to a full understanding of the stages of concern

framework. Among these, the most prominent is the second

diagnostic dimension of the CBAM, levels of use.

Also comprising the set of variables defining the

'state of the user system' are the length of time the user

has had experience with the innovation and whether the user

has had any formal training in the innovation. Both

experience and training have been found to be significantly

related to SoC in a number of studies. Aneke (1996) found

that high school teachers in Virginia involved with the High

Schools That Work reform had SoC that correlated positively

with the amount of experience with and number of hours of

training in the innovation. McQuain (1995) polled teachers

at thirteen high schools and a community college in Virginia

about the innovation of Technical Preparation (Tech Prep)










programs and found both experience and training to be

significantly related to SoC. Chandler (1994) researched

the innovation of institutional effectiveness criteria in

North Carolina community colleges and found significant

differences in respondents grouped by level of (a)

experience and (b) training.

While exploring how innovative instructional computer

users in the School of Humanities and Sciences at Stanford

differed from their colleagues, Leong-Childs (1989) found

that how professors conceptualized their role as teachers

was important in explaining these patterns. Those who paid

more attention to how their students learn (whom she deemed

transactive teachers) had a significantly higher probability

of using computers in innovative instructional ways than

those who primarily attended to what their students learn

(transmissive teachers).

Finally, Shoemaker (1990) explored the relationship

between differences on SoC scores regarding innovative

computer use and native language and culture among foreign-

born faculty members in ten foreign language departments at

the Defense Language Institute. The ten language groups

differed significantly on four of the seven SoC, and he thus

concluded that culture (defined by native language and










country of birth) plays a significant role in types and

intensity of concerns about the innovation of instructional

computer use at DLI.


Instructional Use of the Internet in Hiaher Education

From its originally conceived role as crucial to

national security, the Internet has undergone several

transformations (Dankel, 1996). Its defense role led to

greater dispersion among a select community of university

scholars and researchers for sharing and collaborating on

government-funded projects. Further development permitted

greater spread throughout academic institutions and greatly

increased numbers of users. Under the Internet umbrella and

spurning on its growth is the World Wide Web, developed in

1989 at CERN, the European Laboratory for Particle Physics.

The release in November, 1992, of a new tool, the Mosaic web

browser, made searching much easier and thus opened the door

for users with less computer acumen (Dankel, 1996).

Subsequent additions of user-friendly browsers have

continued the trend.

In summing up the year of 1993 and what happened in

networking at institutions of higher education, Clement and

Abrahams (1994) also give a taxonomy of instructional uses

of the Internet at that point in time:










In postsecondary settings, trends are for gradually
increasing use of networking to support education, in
contrast with research, where networking is well
established and near-essential in many disciplines. An
underlying trend that leads to ongoing change is the
pervasive invasion of campus networks into the college
culture at many universities. We see collaborative
courses being taught by faculty, often between
campuses; we see professors offering network bulletin
boards as course discussion areas, offering tutorial
chat lines and e-mail addresses at the equivalent of
"office hours"; assignment delivery via networks; and
network-based projects and simulations used as course
laboratory areas. The frequency of these applications,
though increasing, is still small in relation to the
total of campus opportunities. (p. 107)

Green's (1996) exhaustive, annual campus computing

survey of two- and four-year public and private colleges and

universities gives the reader a comprehensive view. The

data collected during fall 1995 from the 645 respondents

(survey sent to the primary academic computing official at

each of the 1,500 institutions in the sample), including 42

public research universities which most resemble the

institution of this study, show that approximately 75% of

all faculty have access to the Internet, although only 6% of

all college courses make use of resources on the World Wide

Web for instructional support (Green, 1996, p. 15). Further

indication of this small but growing trend is evidenced by

71.5% of the respondents rating Internet resources for

instruction as a "very important" priority and 50.3% also

seeing web pages for individual classes or whole courses at








39

the same level of importance in the coming 2-3 years (p. 2).

Nevertheless, only 15% of universities reported having a

plan in place for incorporation of instructional resources

drawn from the Internet/World Wide Web (p. 8). Finally,

97.4% of the respondents agreed fully or agreed somewhat

that over the next 2-3 years the Internet/World Wide Web

will be an "important source for content and instructional

resources" (p. 18).

Maddux's 1994 article dealing with seven different

prospects and the problems of adopting the Internet as an

instructional innovation cautions the reader about

statistics such as those above, particularly with regard to

'access.' While a figure like 75% of all faculty with

Internet 'access' may seem to pave the way for fairly rapid

incorporation of its resources into instruction, Maddux

points out that what is often termed 'access' is mere

physical availability. He says that instructional access to

the benefits of an information technology such as the

Internet is something much more complex. Another of his

stated problems involves support for adoption of the

Internet as an innovation. Support must be of two kinds:

technical and curriculum. It is precisely this curriculum

support which is the typically missing link between










availability of a new technology and access to its

instructional potential.

Discipline-specific accounts of experiences with the

development of instructional uses of the Internet abound.

Some, such as Smith's 1995 article, provide insights into

new ways of communicating in both directions between

instructor and student as well as student and student in a

particular discipline (physics in this case), but

applications are easily imagined across disciplines. Others

explore how teaching college-level writing is changing and

is changed by use of the Internet, and still others present

specific case studies including the new skills which are

added and the wider range of materials such as hypertext

documents, all despite the typically "chilly embrace

humanities departments have given computers" (Rouzie, 1995,

p. 4).

Some accounts are more than personal histories, as

teachers look for empirical evidence of differences in

student learning as a result of integrating use of the

Internet into their courses. A professor of sociology at

California State University at Northridge randomly divided

his Fall 1996 class in statistics, assigning one group to

his accustomed format and teaching the other through an on-

line version of the course. Although subjective comments










about the experience filled the spectrum, the results on

both the mid-term and the final examinations (the only times

when the on-line group ever showed up) revealed that "the

wired students outscored their traditional counterparts by

an average of 20 per cent" (McCollum, 1997, p. A23).


Suarv

This section has sought to trace the research pertinent

to the study at hand which relates directly to the adoption

of an innovation within an educational setting. A brief

overview of the landmark work of Rogers (1962) was followed

by an analysis of the effects of change at the micro level,

that of the individual. The work of Fuller (1969) which

served as a basis for the Concerns-Based Adoption Model of

Hall et al. (1973) was reviewed. The parts of the CBAM were

described in detail, and examples of the incorporation of

CBAM into other models also accounting for individual

preferences was discussed. Applications of diagnostic

frameworks of the CBAM were also presented. Development of

the Stages of Concern Questionnaire and its psychometric

properties were outlined. Special attention was given to

the demographic page which typically accompanies it and to

studies which looked at the possible relationships between

the variables captured on the demographic page and the








42

stages of concern. Finally, a brief overview of the recent

history of the Internet and of its growing contribution to

the changes underway in instruction at institutions of

higher education was given.














CHAPTER 3
RESEARCH DESIGN


Study Design/Procedures

Of primary interest in this descriptive correlational

study are the concerns regarding the innovation adoption of

the Internet for instructional purposes among the faculty at

the University of Florida. To explore these concerns, the

following questions were posed:

(1) What are the relationships of the levels of

Internet use for instructional purposes and the levels of

Internet use for all other purposes to the sequence of

stages of concern?

(2) Are there significant differences in the peak

stages of concern of the faculty members grouped by the

extent to which they modify their instructional practices

based on how or what students learn?

(3) Are there significant differences in the peak

stages of concern among faculty members grouped by rank,

gender, age, or national origin?










Variables

The dependent variable in the study was the peak or

most intense of the seven stages of concern about adoption

of the Internet for instructional purposes. Derived from

the Stages of Concern Questionnaire, each stage score ranges

from 0 to 35 on Stage 0-Awareness, Stage 1-Personal, Stage

2-Information, Stage 3-Management, Stage 4-Consequence,

Stage 5-Collaboration, and Stage 6-Refocusing.

The independent variables were level of Internet use

for instructional purposes, level of Internet use for all

other purposes, level of attention paid to how students

learn, level of attention paid to what students learn, rank,

gender, age, and national origin.


Sample

The population for this study consisted of the 1,650

persons employed at the University of Florida during spring

semester 1997 who are remunerated as faculty within the

following ten college-level units: Architecture, Business

Administration, Education, Engineering, Fine Arts, Health

and Human Performance, Health Professions, the Institute of

Food and Agricultural Sciences, Journalism and

Communications, and Liberal Arts and Sciences. The Office

of the Associate Vice-President for Academic Affairs

supplied the labels used to distribute the questionnaire to










the nine colleges, and the IFAS mailing facility supplied

them for that unit.

Since a time lag typically occurs between a) when a

faculty member leaves or a new one is appointed and starts

to work and b) when the database is updated to reflect the

changes, the information from these labels vary slightly

from the true number of faculty within an academic unit at

any given time.

The Office of Academic Affairs supplies labels pre-

addressed with department names and the number of faculty at

a given destination according to the latest database

figures. When mail is delivered bearing these labels, it is

put into faculty mailboxes without regard to the names of

the currently employed members. To provide a more complete

profile of the respondents and to control for questionnaires

destined to those other than the intended, additional

demographic data were collected on what percent-time the

faculty member was teaching during the semester of the

survey, whether the faculty member was working part- or

full-time, academic rank and type of faculty line

(tenure/non-tenure).

The mailing labels supplied by the mailing facility of

the Institute of Food and Agricultural Sciences were

personally addressed. These labels were affixed to the top








46

of the letters of transmittal, and the surveys were grouped

into envelopes by academic unit.


Instrumentation

The instrument used in this mailed survey was the 35-

item Stages of Concern Questionnaire, developed by The

Research and Development Center for Teacher Education at The

University of Texas at Austin as part of the Procedures for

Adopting Educational Innovations/CBAM Project (Hall et al.,

1979). Permission to use the copyrighted questionnaire for

the purpose of this study was requested on October 31, 1996,

and granted on November 1, 1996 (see letter in Appendix B).

The Institutional Review Board of the University of

Florida reviewed a draft of the entire instrument and the

cover letter submitted on December 5, 1996; approval was

granted on December 12, 1996 (see letter in Appendix C).

Development of the questionnaire

The conceptual basis for developing a questionnaire of

concerns about adoption of an innovation grew out of the

extensive work of Frances Fuller during the 1960s and Gene

Hall and his associates during the 1970s. From responses

elicited by an open-ended concerns survey, a total of 544

items was narrowed to a group of 400 judged as central to

measuring concerns. From these, editing and eliminating

redundant items left 195. After administering this 195-item








47

survey and analyzing the data from 359 respondents, over 60%

of the variance in the items was attributed to seven

factors, a finding resulting from item correlation and

factor analyses. The seven stages of concern represent the

outcome of this work (Hall et al., 1979).

Throughout the subsequent testing of the instrument and

its eventual shortening to the current 35 items,

administration of the survey was on populations of teachers

and college faculty involved in adoption of a number of

educational innovations. For example, in 1974 (n = 359) the

innovation at the higher education level was the use of

instructional modules, while that for elementary teachers

was teaming (Hall et al., 1979).

Reliability *lir.

Internal reliability (alpha) coefficients for the

stages range from .64 to .83 (n = 830, a stratified sample

of professors and teachers). Test-retest correlations (n =

132, two weeks elapsed) are from .65 to .86 (Hall et al.,

1979, p. 11).

Indications of the validity of the instrument include

the correlations between individual items and the stage

which they supposedly measure (72% of the items had higher

correlations with their assigned stage than with other

stages), lower correlations between distant stages than

between adjacent ones, results of factor analyses which








48

indicate independence of the individual stage subscales, and

multi-method procedures using the survey and interviews

resulting in multiple Rs greater than .56 (p < .05) for four

of the stages (Hall et al., 1979).

Description of the Ouestionnaire

Respondents choose from eight alternatives on a Likert-

type scale: (0) irrelevant, or one of seven levels of

concern (level 1 is defined as "Not true of me now"; levels

2, 3, and 4 represent gradations of "Somewhat true of me

now"; and levels 5, 6, and 7 represent gradations of "Very

true of me now" [Hall et al., 1979, p. 69]). Each of the

seven stage scores is the sum of the five individual item

scores measuring that stage of concern, constituting a range

of from 0 to 35. A description of each stage and the five

items which make up the score for that stage are as follows

(adapted from Hall et al., 1979; original concept from Hall

et al., 1973):

0 Awareness. Little concern about or involvement with

the innovation is indicated.

I don't even know what using the Internet for

instructional purposes would be.

I am not concerned about use of the Internet for

instructional purposes.

I am completely occupied with other things.








49

Although I don't know about instructional Internet use,

I am concerned about issues in this area.

At this time, I am not interested in learning about

using the Internet for instructional purposes.

1 Informational. A general awareness of the innovation

and interest in learning more detail about it is indicated.

The person seems to be unworried about himself/herself in

relation to the innovation. S/he is interested in

substantive aspects of the innovation in a selfless manner

such as general characteristics, effects, and requirements

for use.

I have a very limited knowledge of instructional uses

of the Internet.

I would like to discuss the possibility of using the

Internet for instructional purposes.

I would like to know what resources are available if we

decide to adopt instructional use of the Internet.

I would like to know what instructional use of the

Internet will require in the immediate future.

I would like to know how instructional use of the

Internet is better than what we are doing now.

2 Personal. Individual is uncertain about the demands

of the innovation, his/her inadequacy to meet those demands,

and his/her role with the innovation. This includes

analysis of his/her role in relation to the reward










structure of the organization, decision making, and

consideration of potential conflicts with existing

structures or personal commitment. Financial or status

implications of the program for self and colleagues may also

be reflected.

I would like to know the effect of using the Internet

for instructional purposes on my professional status.

I would like to know who will make the decisions

regarding use of the Internet for instruction.

I would like to know how my teaching or administration

is supposed to change.

I need more information on time and energy commitments

required by instructional Internet use.

I would like to know how my role will change when I am

using the Internet for instructional purposes.

3 Management. Attention is focused on the processes

and tasks of using the innovation and the best use of

information and resources. Issues related to efficiency,

organizing, managing, scheduling, and time demands are

utmost.

I am concerned about not having enough time to organize

myself each day.

I am concerned about conflict between my interests and

my responsibilities.










I am concerned about my inability to manage all that

instructional Internet use requires.

I am concerned about time spent working with non-

academic problems related to using the Internet for

instructional purposes.

Coordinating tasks/people takes too much of my time.

4 Consequence. Attention focuses on impact of the

innovation on student in his/her immediate sphere of

influence. The focus is on relevance of the innovation for

students, evaluation of student outcomes, including

performance and competencies, and changes needed to increase

student outcomes.

I am concerned about students' attitudes toward

instructional use of the Internet.

I am concerned about how instructional use of the

Internet affects students.

I am concerned about evaluating my impact on students.

I would like to excite my students about their part in

using the Internet for instructional purposes.

I would like to use feedback from students to change

use of the Internet for instruction.

5 Collaboration. The focus is on coordination and

cooperation with others regarding use of the innovation.










I would like to help other faculty in their

instructional Internet use.

I would like to develop working relationships with both

our faculty and outside faculty using the Internet for

instructional purposes.

I would like to familiarize other departments or

persons with our progress in using the Internet.

I would like to coordinate my effort with others to

maximize the effects of instructional Internet use.

I would like to know what other faculty are doing in

this area.

6 Refocusing. The focus is on exploration of more

universal benefits from the innovation, including the

possibility of major changes or replacement with a more

powerful alternative. Individual has definite ideas about

alternatives to the proposed or existing form of the

innovation.

I now know of some other approaches that might work

better.

I am concerned about revising my instructional use of

the Internet.

I would like to revise the instructional approach to

use of the Internet.










I would like to modify our instructional use of the

Internet based on the experiences of our students.

I would like to determine how to supplement, enhance,

or replace instructional use of the Internet.

In addition to the responses on the Stages of Concern

Questionnaire, data from demographic variables were also

collected. The percent time teaching during the semester of

inquiry, working part- vs. full-time, and the type of

faculty line held (tenure or non-tenure) were collected to

control for distribution of the questionnaires to those who

do not make up the population of primary interest to the

study.

Levels of use of the Internet for instructional

purposes and levels of use of the Internet for all other

purposes were collected to address research question 1 and

to test theory from the Concerns-Based Adoption Model.

Stages of concern are the first dimension of the CBAM, and

levels of use make up the second dimension. In the CBAM

eight different levels of use are proposed. More pro-active

in nature, the levels of use sequence resembles that of the

stages of concern quite closely, as do the description of

the states and the decisions made by the user. The eight

levels of use (Hord, et al., 1987, p. 55) and their

corresponding stages of concern are:










Level of Use Stage of Concern

Level 0 Non-Use Stage 0 Awareness

Level 1 Orientation Stage 1 Information

Level 2 Preparation Stage 2 Preparation

Level 3 Mechanical Use
Stage 3 Management
Level 4a Routine
Stage 4 Consequence
Level 4b Refinement

Level 5 Integration Stage 5 Collaboration

Level 6 Renewal Stage 6 Refocusing

Levels of use are typically determined during a focused

interview with the help of an extremely detailed and

explicit chart. Since the size of the sample for this study

necessitated gathering the data on a survey, a taxonomy

appropriate for such a technique had to be found. Dreyfus

and Dreyfus (1986) offer the five-stage model for skill

acquisition which has been used here. A category for the

non-user was added to their five levels as outlined below.

The corresponding levels of use from the CBAM are also

shown:

CBAM Level of Use Survey Terms

Level 0 Non-Use

Level 1 Orientation Non-user

Level 2 Preparation

Level 3 Mechanical Novice










Level 4a Routine Advanced beginner

Level 4b Refinement Competent user

Level 5 Integration Proficient user

Level 6 Renewal Expert

Finally, items inquiring about the number of years at

the institution, length of time involved with instructional

use of the Internet, prior formal training in the Internet

for instructional purposes, and recent or concurrent

involvement in some other major innovation or program were

taken from the original demographic variable sheet suggested

by Hall, et al. (1979) and contribute to the full assessment

of the "state of the user system" (p. 52).

An estimate of the completion time of the 35-item SoCQ

was ten to fifteen minutes. The additional items did not

add substantially to this range.

The questionnaire was distributed to a panel of six

experts, all members of the same group to whom it would

eventually be distributed, for comment on overall clarity of

the items and suggestions for improvement. The final

version reflected the incorporation of their suggestions.

In one case the order of the demographic variable items was

changed, and response to other comments led to naming 'the

Internet for instructional purposes' rather than relying on

'the innovation' throughout (survey found in Appendix D).











Data Collection

The first announcement of the study was mailed on

January 10, 1997, on a bright pink post card. Its primary

objective was to alert the faculty to the coming survey.

The text of the notice is found in Appendix E.

The questionnaire itself was duplicated front and back

on high-quality pink paper and was accompanied by a letter

of transmittal on top. The letter of transmittal was on the

letterhead of the University Center for Excellence in

Teaching and was signed by the director (see Appendix E).

The first page under the letter of transmittal was dedicated

to instructions on filling out the concerns questionnaire.

The format and wording were taken directly from Hall et al.,

1979. Specific definitions of 'Internet' and 'instructional

purpose' were included, as well as examples of the latter.

The survey began on the back of the instructions page.

The questions were numbered consecutively, and the final

page included an area for the respondent to write comments.

The return address was placed on the instruction page,

repeated again at the end of the questionnaire below the

comments area, and printed three times on the final

(outside) page. Respondents were requested to fold, close,

and return the questionnaires via campus mail.










To prepare the questionnaires for mailing, each

academic unit was assigned a three-digit code number, and

that same number was placed on each questionnaire destined

for that academic unit. The number of faculty indicated on

the label determined the number of questionnaires placed in

a large manilla envelope destined for that academic unit. A

cover sheet was included on top requesting the cooperation

of the recipient in distributing the questionnaires to the

faculty for whom they were intended.

The questionnaire itself was sent through campus mail

on Friday of the third week of spring semester, i.e.,

January 24, 1997. The timing of the mailing sought to

maximize the attention it might get among those on an

academic schedule with an uneven workload distribution

throughout the typical sixteen-week semester.

Simultaneous with the mailing, the questionnaire was

posted on the World Wide Web at the UCET site. Care was

taken in the design of the survey for the web site so that

responses would be comparable irrespective of which medium a

respondent used. Although all those who replied had a paper

copy of the questionnaire available to them, they did not

need any information from the paper copy to be able to

answer the form over the Internet. Links to the full text

of the letter of transmittal and to the instructions for










filling out the Stages of Concern Questionnaire were

included. At the end of the form the respondent activated a

"Send Form" button which transmitted the results to an email

address anonymously.

The Internet address for the site was included in the

cover letter and repeated on the questionnaire instruction

page. The electronic form had one additional demographic

item at the beginning; it requested the respondent's

departmental affiliation (either the three-digit code from

the paper copy or the name of the academic unit), since this

information was lost from the paper copy by those choosing

to submit their responses electronically.

The requested return date of the instruments was

Wednesday, February 5, 1997, or twelve days after they were

mailed out.

A follow-up mailing following the same procedure was

undertaken. The letter of transmittal included a polite

reminder and a request to complete the questionnaire. Those

who had already done so were asked to disregard the request

(see Appendix F for follow-up letter of transmittal). The

mailing of the follow-up went out on Friday, February 21,

1997, and return was requested by Wednesday, March 5, 1997,

again an interval of twelve days. This second deadline fell








59

in the middle of the week immediately prior to the week-long

spring break.


Data Analysis

Responses to the survey from the three mailings totaled

540 of a possible 1,650 in the population. Of these, 50 (or

9%) were received electronically. The sample thus

represents a 33% response rate. This compares favorably

with the results found during a review of the literature for

similar surveys on equivalent populations (response rate

percentages in the high twenties were the rule although one

was found at 32). Furthermore, surveys within the last year

on subsets of this population, one undertaken within the

Institute of Food and Agricultural Sciences only and another

campus-wide of new faculty after their first year at the

University of Florida, had response rates of 20% and 30%

respectively.

The survey data were entered, verified, and saved on a

3.5" disk from where they were subsequently stored on the

mainframe of the University of Florida Center for

Instructional and Research Computing Activities. Analyses

of the data were carried out using SAS run interactively

with the UNIX operating system on the mainframe.








60

Demographic information on the respondents is shown in

Table 1. Comparison of the sample to all faculty as

described in the University of Florida Affirmative Action

Plan-1996 shows that females are represented only slightly

higher in the sample than in the population (26.2% of the

sample vs 25.7% of all faculty) and that those on tenure-

track faculty lines responded in a higher proportion (85.3%

of the sample vs 74.0% of all faculty occupy faculty lines).

In order to address the research questions, a

correlational analysis and multiple regression were used.

The alpha level to test for statistical significance was set

at .05.


Suarv

This section has given an overview of the study

undertaken, including the design of the study, the

variables, the instrument, and the procedures followed.

Specifically, the sample is described; the instrument is

identified and described, and its development is traced;

methods of data collection are outlined; demographic

variables are reported for the respondents; and analyses of

the data are set out.










Table 1


Demographic Profile of the Respondents


Time (N = 531)

Part-time
Full-time


Gender (N = 530)


Male
Female


30 6
501 94


379 72
151 28


Rank (N = 538)


Instructor/Lecturer
Post-doc
Assistant Professor
Associate Professor
Professor
Other


Track (N = 534)


Tenure
Non-tenure


456 85
78 15




457 86
74 14


National Origin (N = 531)

American
Foreign


_f %














CHAPTER 4
RESULTS AND DISCUSSION


This sections opens with a restatement of the problem

and of the research questions of interest in the study.

These are followed by the variables chosen for the study and

then a section on the findings. The data are first analyzed

descriptively one at a time. Then the bivariate

relationships are examined. Finally, the results of the

multiple regression analysis of the data are given,

specifically in relation to the research questions set out

at the beginning of the study. Also included is a

discussion of the meaning of each of the significant

findings and its relationship to findings of previous

studies.


Research Ouestions

Central to this descriptive correlational study were

the concerns regarding the innovation adoption of the

Internet for instructional purposes among the faculty at the

University of Florida. To explore these concerns, the

following questions were posed:

(1) What are the relationships of the levels of










Internet use for instructional purposes and the levels of

Internet use for all other purposes to the sequence of

stages of concern?

(2) Are there significant differences in the peak

stages of concern of the faculty members grouped by the

extent to which they modify their instructional practices

based on how or what students learn?

(3) Are there significant differences in the peak

stages of concern among faculty members grouped by rank,

gender, age, or national origin?


Variables

The dependent variable in the study was the peak or

most intense stage of the seven stages of concern about

adoption of the Internet for instructional purposes.

Derived from the Stages of Concern Questionnaire, each stage

score ranges from 0 to 35 on Stage 0-Awareness, Stage 1-

Personal, Stage 2-Information, Stage 3-Management, Stage 4-

Consequence, Stage 5-Collaboration, and Stage 6-Refocusing.

Although all respondents to the questionnaire have scores on

all seven stages of concern, only the peak stage of concern

of each individual was of interest in this study. When two

or more stage of concern scores shared an equal intensity

(i.e., an equal score within the 0-to-35 range), the stage

at the further (furthest) point along the developmental










continuum from Stage 0 to Stage 6 was chosen as the peak

stage of concern.

The independent variables were level of Internet use

for instructional purposes, level of Internet use for all

other purposes, level of attention paid to how students

learn, level of attention paid to what students learn, rank,

gender, age, and national origin.

In order to address the research questions, the alpha

level was set at .05, and a correlational analysis and

regression technique were used. Borg and Gall (1989)

mention that when dealing with "the degree of relationship

among various combinations of three or more variables"

multiple regression is a technique offering "considerable

versatility" (p. 601) in correlational studies.


Findings

Descriptive Statistics

The greatest portion of the sample, 132 or about one-

fourth of the respondents, was found to exhibit their peak

stage of concern regarding adoption of the educational

innovation of use of the Internet for instructional purposes

at the first of the self stages, Information. Another 119

(22%) of the sample were found to be at Stage 4 -

Consequence, the first of the impact stages. The mean peak










stage of concern for the sample was 2.90 (SD = 1.94) (high

Stage 2 Personal). (See Table 2 for a complete summary.)

Table 3 displays the findings on the levels of use of

the sample participants with regard to the Internet. Over

half of the sample self-reported a level of use of the

Internet for instructional purposes of non-use or novice (n

= 349). Levels of use of the Internet for all other

purposes, however, were higher. Over two-thirds of the

sample use the Internet for non-instructional purposes at

the self-reported levels of novice, advanced beginner, or

competent user. The mean for instructional use of the

Internet fell at level 1 novice (M = 1.24, SE = 1.33),

whereas that for its use for all other purposes fell at

level 2 advanced beginner (M = 2.28, SD = 1.30).

Responses to the survey items asking how much

participants modify their teaching on the basis of how and

what students learn are shown in Table 4. About 65% (n =

311) of the participants reported that they alter their

teaching on the basis of how students learn to an average

degree or quite a bit. Likewise 310 members of the sample

or 67% reported one of the same two categories when asked

about how much they change their teaching according to what

students learn.

A profile of the sample respondents with regard to

their rank, gender, and national origin is given in Table 5.










Table 2

Peak Staae of Concern


Stage (- = 540) _%


0 Awareness 47 9

1 Information 132 24

2 Personal 77 14

3 Management 53 10

4 Consequence 119 22

5 Collaboration 26 5

6 Refocusing 86 16

Note. M = 2.90.

SD = 1.94.










Table 3

Levels of Use of the Internet


Level _A


Instructional purposes (N = 533)

0 Non-user 207 39

1 Novice 142 27

2 Advanced beginner 81 15

3 Competent user 67 13

4 Proficient user 23 4

5 Expert 13 2

Non-instructional purposes (N = 537)

0 Non-user 45 8

1 Novice 115 21

2 Advanced beginner 150 28

3 Competent user 122 23

4 Proficient user 84 16

5 Expert 21 4


Note Instructional purposes: = 1.24.

SD = 1.33.

Non-instructional purposes: I = 2.28.

S= 1.30.










Table 4

How Much Teachine Is Modified Based on


Student Learnina


Degree of Modification f %


How students learn (N = 476)

0 Not at all 15 3

1 Moderately 79 17

2 Average 139 29

3 Quite a bit 172 36

4 A great deal 71 15

What students learn (N = 471)

0 Not at all 11 2

1 Moderately 70 15

2 Average 127 27

3 Quite a bit 188 40

4 A great deal 75 16


Note. How:

SW =

What: M =

SD =


2.43.

1.03.

2.52.

1.00.













Table 5

Rank. Gender and National Origin of Respondents


Rank (N = 538) _



Instructor/Lecturer 34 6

Post-doc 2 0

Assistant Professor 106 20

Associate Professor 138 26

Professor 226 42

Other 32 6



Gender (I = 530)

Male 379 72

Female 151 28



National Origin (I = 531)

American-born 457 86

Foreign-born 74 14










One-fifth of the sample were Assistant Professors, one-

fourth (26%) Associate Professors, and just under one-half

(42%) were Professors. The sample was about three-fourths

male (72%). The age range of the respondents was from 24 to

83 years old; the mean age was 48.58 and the standard

deviation was 9.90. The participants were 86% native-born

American citizens.


Correlational Relationships

Table 6 contains a correlation matrix showing the zero-

order or Pearson product-moment correlations between all the

variables in the study. For the purposes of this part of

the analysis of the data, the respondents who held a rank

other than Assistant Professor, Associate Professor, or

Professor were excluded. Lee et al. (1989) state that "no

adjustment may be necessary when a very small proportion of

cases is excluded" (p. 40)--the case here where 68 of 538,

or 12%, were excluded. In addition, the four possible

answer choices for the question of whether the respondent

was an international faculty member were reduced to a

dichotomous variable: American- or foreign-born.

Davis's taxonomy (1971) categorizes the strength of

association measured by correlations in the following

manner:









Table 6

Intercorrelations Between Peak Stage of Concern and Other Variables


Peak L'iip L'net How What Rank Sex Age Orig


Peak
L'iip 33 --
L'net 0.17 0.65 --
How 0.0 0.10 -0.05
What 0.02 0.07 0.03 0.57

Rank -0.05 -0.13 -0.14 -0.18 -0.09
Sex 0.04 -0.15 -0.24 0.17 0.11 -0.32
Age -0.07 -0.16 -0.25 -0.04 0.01 0.68 -1 --
Orig 0.04 0.02 0.07 0.04 0.09 -0.02 -0.03 0.01


Note. Peak = peak stage of concern; L'iip = level of instructional Internet use; L'net =

non-instructional Internet use level; How = attention to how students learn; What =

attention to what students learn; Orig = national origin (American- or foreign-born).

Single underline denotes p < 0.05; double underline denotes p < 0.01.










Correlation Value Classification

.70 and above Very strong

.50 to .69 Substantial

.30 to .49 Moderate

.10 to .29 Low

.01 to .09 Negligible



When the independent variables are viewed one at a time

for a relationship to the dependent variable peak stage of

concern, three are found to be significantly correlated with

the outcome variable. Moderately correlated with peak stage

of concern is level of use of the Internet for instructional

purposes (r = 0.33, p < .01). Level of use of the Internet

for all other purposes is also significantly related to peak

stage of concern, but this variable demonstrates only a low

level of strength of association (r = 0.17, R < .01)

according to Davis's terminology. Finally, degree to which

respondents reported that they modify their teaching based

on how students learn also showed a low level of strength of

association (r = 0.10, p < .05).

Further bivariate analysis of the data included viewing

the scatter diagrams to assure that a linear relationship

was "a reasonable approximation" (Agresti & Finlay, 1986, p.

274) of the assumed form of the relationships between peak










stage of concern and the independent variables. In each

case, this assumption was confirmed.


Repression Analysis

Finally, a multiple regression with all of the

variables of interest in the study was carried out to

determine the contribution of each to the prediction of peak

stage of concern when all others were controlled. Results

are given in Table 7.


Research Ouestion 1

What are the relationships of the levels of Internet

use for instructional purposes and the levels of Internet

use for all other purposes to the sequence of stages of

concern?

Both levels of use of the Internet for instructional

purposes and levels of use of the Internet for all other

purposes are significantly correlated with peak stage of

concern. However, when controlling for other variables,

only level of use of the Internet for instructional purposes

has a significant strength of association with peak stage of

concern (L = 7.591, p < 0.0001). All other variables being

equal, a respondent who was one level higher in use of the

Internet for instructional purposes exhibited a peak stage

of concern 0.62 stage higher.










Table 7

Summary of MultiPle Regression for Variables Predictina Peak
Staae of Concern


Variable SE B


Level of use of the Internet
for instructional purposes 0.62* 0.08

Level of use of the Internet
for all other purposes -0.08 0.09

Degree of modification to teaching
based on how students learn 0.12 0.11

Degree of modification to teaching
based on what students learn -0.12 0.11

Rank 0.15 0.15

Gender 0.54* 0.21

Age -0.01 0.01

National Origin -0.02 0.24


Note. D < .05.










The findings of the significant contribution of level

of use of the educational adoption being investigated in

this study, use of the Internet for instructional purposes,

confirm the theory of the Concerns-Based Adoption Model,

which states that stages of concern about the innovation and

Levels of Use of the same "will move in nearly a one-to-one

correspondence" (Loucks & Hall, 1977, p. 20).


Research Ouestion 2

Are there significant differences in the peak stages of

concern of the faculty members grouped by the extent to

which they modify their instructional practices based on how

or what students learn?

The degree to which participants reported that they

modify their teaching based on how students learn proved to

be significantly correlated to peak stage of concern,

although the degree to which they claimed to modify their

instruction based on what students learn did not.

This finding would lend support to the results of the

1989 study by Leong-Childs, who found that Stanford

University School of Humanities and Sciences professors who

were innovative instructional computer users differed

significantly in the amount of attention they pay to how

their students learn.










Nevertheless, when all other variables of the current

study were controlled, neither degree to which respondents

modify their teaching based on how students learn nor the

same based on what students learn contributed significantly

to the prediction of peak stage of concern (1 = 1.165 for

attention to how students learn and t = -1.125 for attention

to what).


Research Ouestion 3

Are there significant differences in the peak stages of

concern among faculty members grouped by rank, gender, age,

or national origin?

Rank was not significantly correlated to peak stage of

concern; and when all other variables were controlled, it

did not have a significant strength of association with peak

stage of concern (i = 1.024).

Although gender was not significantly correlated to

peak stage of concern, when all other variables were

controlled it contributed significantly to the prediction of

peak stage of concern (1 = 2.543, p = 0.0114). When all

other variables were controlled, women respondents were

predicted to be .54 stage higher on their peak stage of

concern.

This finding for the respondents in the sample would

seem to refute the results of research reported by Hall et










al. (1979) reporting "no outstanding relationships between

standard demographic variables and concerns data" (p. 52)

and concurring with Rutherford (1986) that gender (and other

variables) lack any "consistent relationship" (p. 7) with

stage of concern scores.

Age was not significantly correlated to peak stage of

concern; and when all other variables were controlled, it

did not have a significant strength of association with peak

stage of concern (I = .0832). This would validate the

findings of Hall et al. (1979) and Rutherford (1986).

National origin was not significantly correlated to

peak stage of concern; and when all other variables were

controlled, it did not make a significant contribution to

the prediction of peak stage of concern (L = -0.071).

Although Shoemaker (1990) found significant differences in

stage of concern scores about adoption of instructional use

of computers in foreign language teaching among the five

foreign culture groups (as defined by native language and

country of birth) he investigated, his study dealt solely

with foreign culture groups. Since Shoemaker's study did

not include Americans as a cultural group and the foreign-

born respondents of this study were not further subdivided,

no direct parallel can be drawn.














CHAPTER 5
SUMMARY AND CONCLUSIONS


This section opens with a summary of the study

undertaken. It then moves on to reiterate the findings and

to state the conclusions which can be reasonably drawn from

those findings. Implications for the conclusions are given,

and finally recommendations for further research are

outlined.


Summer

This exploratory, descriptive correlational study set

out to measure the stages of concern of the faculty of the

University of Florida regarding adoption of the innovation

of use of the Internet for instructional purposes.

The current paradigm shift underway as institutions of

higher education break out of the constraints of the last

several hundred years of a model of same time, same place

learning has incited exploration of new modes of

instructional delivery. Recent advances in computers and

telecommunications technology have been crucial to the role

the Internet will play in this shift. Greater capacity for

storage and faster retrieval and transfer of information








79
have meant greater potential for this medium than could have

been conceived of only a short time ago.

Teachers, not unlike others confronted with change or

innovation, react in a number of ways to the current times.

Some avail themselves of training and informal meetings to

learn about what others have done or to share what they have

accomplished. A mid-April 1997 talk entitled "Impact of

Multimedia Lectures" given by a faculty member of the

Department of Psychology at the University of Florida and

sponsored by the University Center for Excellence in

Teaching and the Office of Instructional Resources resulted

in higher attendance than any such voluntary meeting to

date. Over 60 faculty members, administrators, and graduate

teaching assistants gathered to hear about the experience of

one of their colleagues and to learn from it.

The topic addressed at this lecture represents quite a

different orientation from that of workshops being given as

few as two years ago, where during a one-day conference

called Data Day '95 University of Florida officials

organized sessions to inform faculty and graduate students

about Internet resources that faculty may not be aware of"

(Passarella, 1995, p. 8). At that time a member of the

organizing committee admitted to not anticipating a

sophisticated level of experience among those attending.








80

Faculty concerns over instructional use of the Internet

have become even more pertinent during the 1996-1997

academic year. As changes in access to computers and

networked resources in student computer laboratories across

campus have been put into effect, students have resisted

paying an additional fee. The administration's distinction

between essential and elective computer access has been one

with which students could not agree.

The theoretical framework for the study was the

Concerns-Based Adoption Model, which has been developed to

help facilitate change at the level of the individual

involved with innovation adoption within educational

settings. The model was developed by Gene Hall and his

associates at The Research and Development Center for

Teacher Education of the University of Texas at Austin as

part of the Procedures for Adopting Educational

Innovations/CBAM Project. The model has multiple dimensions

which serve to guide change facilitators and others for

successful implementation of educational change.

The first dimension of the CBAM involves the stages of

concern, further described as seven psychological constructs

through which an individual passes when confronted with

change. Concerns are defined as an individual's motivation,

thoughts, feelings, and perceptions.








81

The Stages of Concern Questionnaire has been developed

to measure an individual's concerns regarding an educational

innovation. Interventions designed to resolve the most

intense stages) of concern are thought to promote more

efficient and effective implementation of the innovation

being proposed. Moreover, arousal of concerns at higher

stages is unlikely until resolution has been achieved at

lower stages of concern. The 35-item questionnaire offers a

pencil-and-paper measure of an individual's concerns and

supplies important information for planning training

interventions.

Specifically, the research questions the study set out

to answer were:

(1) What are the relationships of the level of Internet

use for instructional purposes and the level of Internet use

for all other purposes to the sequence of stages of concern?

(2) Are there significant differences in the peak

stages of concern of the faculty members grouped by the

extent to which they modify their instructional practices

based on how or what students learn?

(3) Are there significant differences in the peak

stages of concern among faculty members grouped by rank,

gender, age, or national origin?

In order to investigate these questions, a mailed

survey using the Stages of Concern Questionnaire was carried










out of 1,650 faculty members in ten colleges within the

University of Florida. A total of 540 responses (33%)

provided the data for analyzing the results.


Findings

With respect to the first research question, both

levels of use of the Internet for instructional purposes and

levels of use for all other purposes were significantly

correlated with peak stage of concern among the members of

the sample. Nevertheless, when controlling for other

variables of interest in the study, only level of use of the

Internet for instructional purposes contributes

significantly to the prediction of peak stage of concern.

Typically, the respondent who was one level of use higher

was .62 stage further along in the developmental continuum

known as stages of concern.

With regard to the second research question, only the

extent to which respondents reported that they change their

teaching based on how students learn was significantly

correlated to peak stage of concern. When the other

variables were controlled, neither the extent to which

respondents reported that they change their teaching based

on how students learn nor the same based on what students

learn showed a significant contribution to prediction of

peak stage of concern.










Finally, among the variables of rank, gender, age, and

national origin, none was significantly correlated to the

outcome variable. Furthermore, only gender exhibited a

significant contribution to prediction of peak stage of

concern when all other variables were controlled.


Implications

Since instructional use of the Internet is similar to

instructional use of other computer technology in that it

often has a preceding use which may involve research or even

recreation before transfer of the power and potential of the

technology to classroom applications, the finding in this

study that level of use of the Internet for instructional

purposes will significantly contribute to predicting peak

stage of concern is not surprising. Promoting use of the

Internet for any purpose among university faculty may well

spur the user on to recognizing possible teaching

applications. In order to increase awareness of what is

available, an issue as simple as faculty member access to

network connectivity in the office first, and updated

equipment with sufficient processing speed and memory

capacity second, may well be key among the conditions that

would promote instructional applications of this medium.

Mention is made of spending future allocations from the

technology fee generated by the monies collected from










students on connecting every classroom to the Internet.

However, this may well be 'buying the saddle before buying

the horse.' Attention to assuring that every faculty office

is connected first and supplied with sufficient resources to

benefit from something so simple and low-tech as email first

and then from what is available on the Internet may indeed

be a more informed route.

Since gender is correlated negatively with rank (-0.32,

p < .01) as well as with age (-0.18, p < .01) in the sample,

we know that women tend to be represented more than would be

expected at the lower ranks and among the younger faculty

members. In addition, gender maintained a significant

contribution to the prediction of peak stage of concern when

all other variables were controlled. This would mean that

controlling for other variables women of the sample tend to

be further along the developmental continuum represented by

the seven stages of concern (and probably levels of use)

regarding adopting the innovation of use of the Internet for

instructional purposes. The findings place these young

women at the lower ranks in key positions to become change

agents for their colleagues.


Recommendations for Further Research

If level of use of the Internet for instructional

purposes is associated with reaching a given stage of










concern regarding adoption of this educational innovation

and, in turn, level of use of the Internet for all other

purposes is correlated significantly with its use in the

classroom, then further research into the components of its

use in both categories is needed. Another topic of interest

would be the progression of the interest in and use of the

Internet from non-instructional uses such as research or

recreation to interest in its teaching and learning

applications.

Simply gathering information about the numbers and

locations of faculty with networked connectivity which

permits Internet access in the office and about those who

have the appropriate equipment to make the most of the

educational potential of the medium would be of great value

in setting systematic goals for implementation of wider

faculty use of the Internet for instructional purposes

across the campus.

Whether driven by student demand, through gradual

coercion from sources in the administration, or from

colleagues who are innovators, instructional use of the

Internet is an innovation which holds promise of being

central to the changing patterns in delivery of instruction

by institutions of higher education in the United States and

throughout the world for some time to come.















REFERENCES


Agresti, A., & Finlay, B. (1986). Statistical methods
for the social sciences (2nd ed.). San Francisco: Dellen.

Albright, M. (1996, February 5). Instructional
technology and higher education: Rewards, rights and
responsibilities. Keynote address of the Southern Regional
Faculty and Instructional Development Consortium, Baton
Rouge, LA. (ERIC Document Reproduction Service No.
ED 392 412)

Aneke, N. (1996). Teachers' stages of concern about a
school-wide educational reform (Doctoral dissertation,
Virginia Polytechnic Institute and State University, 1996).
Dissertation Abstracts International. 57(3), 934A.
(University Microfilms No. 9624162)

Borg, W., & Gall, M. (1989). Educational research (5th
ed.). New York: Longman.

Cicchelli, T. (1990). Concerns theory in liberal arts
related to preservice education of undergraduate and
graduate students. Educational Research Ouarterly, 14(4),
41-48.

Cicchelli, T., & Baecher, R. (1987). The use of
concerns theory in inservice training for computer
education. Computers in Education. 11(2), 85-93.


Cicchelli, T., & Baecher, R. (1990). Theory and
practice: Implementina computer technology in a secondary
school. (ERIC Document Reproduction Service No. ED 336 059)

Clement, J., & Abrahams, J. (1994). Networking in 1993.
Educational Media and Technology Yearbook 1994. 20, 106-119.










Chandler, J. (1994). Stages of concern expressed by
selected North Carolina community college faculty and staff
about the adoption of institutional effectiveness criteria
(Doctoral dissertation, North Carolina State University,
1994). (University Microfilms No. 9425455) Dissertation
Abstracts International. 55(4), 843A. (University Microfilms
No. 9425455)

Dankel, D., Jr. (1996). The use of the World Wide Web
in distance education. Paper presented at Computer Expo
1996, University of Florida, Gainesville, FL.

Davis, J. (1971). Elementary survey analysis.
Englewood, NJ: Prentice Hall.

Dreyfus, H., & Dreyfus, S. (1986). Mind over machine:
The power of human intuition and expertise in the era of the
computer. New York: The Free Press.

Falvo, J. (1990). An analysis of language teachers'
concerns regarding new technologies: The case of Minnesota
teachers of Spanish assessing satellite video technology to
deliver authentic cultural materials (Doctoral dissertation,
The University of Nebraska-Lincoln, 1990). Dissertation
Abstracts International. 52(3), 827A.

Fuller, F. (1969). Concerns of teachers: A
developmental conceptualization. American Educational
Research Journal. 6(2), 207-226.

Fuller, F., & Brown, 0. (1975). Becoming a teacher. In
K. Ryan (Ed.), Teacher education: The seventy-fourth
yearbook of the national society for the study of education.
Part II (pp. 25-52). Chicago: The University of Chicago
Press.

Green, K. (1996). Campus computing 1995. Encino, CA:
Campus Computing. (ERIC Document Reproduction Service No. ED
394 383)

Hall, G. (1976). The study of individual teacher and
professor concerns about innovations. Journal of Teacher
Education. 27(1), 22-23.

Hall, G. (1979). The concerns-based approach to
facilitating change. Educational Horizons. 57(4), 202-208.










Hall, G. (1985). A stages of concern approach to
teacher preparation (Report 3213). Austin, TX: The Research
and Development Center for Teacher Education. (ERIC Document
Reproduction Service No. ED 265 126)

Hall, G., George, A., & Rutherford, W. (1977).
Measuring states of concern about the innovation: A manual
for use of the states of concern questionnaire. (ERIC
Document Reproduction Service No. ED 147 342)

Hall, G., George, A., & Rutherford, W. (1979).
Measuring states of concern about the innovation: A manual
for use of the states of concern questionnaire. (R & D
Report No. 3032). Austin, TX: The Research and Development
Center for Teacher Education.

Hall, G., & Hord, S. (1987). Change in schools:
Facilitatina the process. Albany, NY: SUNY Press.

Hall, G., & Loucks, S. (1978). Teacher concerns as a
basis for facilitating and personalizing staff development.
Teachers Colleae Record. 80(1), 36-53.

Hall, G., & Loucks, S. (1981). Program definition and
adaptation: Implications for inservice. Journal of Research
and Development in Education. 14(2), 46-58.

Hall, G., & Loucks, S., Rutherford, W., & Newlove, B.
(1975). Levels of use of the innovation: A framework for
analyzing innovation adoption. Journal of Teacher Education.
26(1), 52-56.

Hall, G., Newlove, B., George, A., Rutherford, W., &
Hord, S. (1991). Measuring change facilitator states of
concern--A manual for the use of the CFSoC questionnaire.
Greeley, CO: Center for Research on Teaching and Learning,
University of Northern Colorado. (ERIC Document Reproduction
Service No. ED 353 307)

Hall, G., & Rutherford, W. (1976). Concerns of teachers
about implementing team teaching. Educational Leadership.
34(3), 227-233.

Hall, G., Wallace, R., Jr., & Dossett, W. (1973). A
developmental conceptualization of the adoption process
within educational institutions. Austin, TX: The Research
and Development Center for Teacher Education. (ERIC Document
Reproduction Service No. ED 095 126)










Hickox, C. (1994). Training for the Internet: Stages of
concern among academic library staff in the Amigos
Consortium (Doctoral dissertation, East Texas State
University, 1994). Dissertation Abstracts International.
55(11), 3383A. (University Microfilms No. 9510955)

Hord, S., & Loucks, S. (1980). A concerns-based model
for the delivery of inservice. (ERIC Document Reproduction
Service No. ED 206 620)

Hord, S., Rutherford, W., Huling-Austin, L., & Hall, G.
(1987). Taking charge of change. Alexandria, VA: Association
for Supervision and Curriculum Development.

King, K. (1990). Information technologies in support of
teaching and learning. Higher Education Management. 2(3),
294-298.

Kugel, P. (1993). How professors develop as teachers.
Studies in Higher Education. 18(3), 315-328.

Lee, E., Forthoffer, R., & Lorimer, R. (1989).
Analyzing Complex Survey Data. Newbury Park, CA: Sage.

Lee-Kang, D. (1993). Factors affecting the adoption of
instructional use of computers in undergraduate textiles,
clothing, and merchandising programs (Doctoral dissertation,
The Ohio State University, 1993). Dissertation Abstracts
International. 54(11), 4009A. (University Microfilms No.
9412003)

Leong-Childs, D. (1989). Professors' use of computers
for innovative instruction (Doctoral dissertation, Stanford
University, 1989). Dissertation Abstracts International.
50(7), 1961A. (University Microfilms No. 8925907)

Lewis, D. (1994). Concerns and characteristics
affecting the adoption of computer based instruction by
diabetes educators (Doctoral dissertation, West Virginia
University, 1994). Dissertation Abstracts International.
55(6), 1535A. (University Microfilms No. 9427972)

Loucks, S., & Hall, G. (1977). Assessing and
facilitating the implementation of innovations: A new
approach. Educational Technoloyv, 17(2), 18-21.

Maddux, C. (1994). The Internet: Educational prospects
and problems. Educational Technoloav, 34(7), 37-42.




Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EZT5NRG36_CFOWMJ INGEST_TIME 2013-11-15T22:29:49Z PACKAGE AA00014241_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES