Moral attribution and the evaluation of action


Material Information

Moral attribution and the evaluation of action
Physical Description:
xii, 192 leaves : ; 28 cm.
Forsyth, Donelson Ross
Publication Date:


Subjects / Keywords:
Attribution (Social psychology)   ( lcsh )
Judgment (Ethics)   ( lcsh )
Responsibility   ( lcsh )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )


Thesis--University of Florida.
Includes bibliographical references (leaves 187-191).
Statement of Responsibility:
by Donelson R. Forsyth.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000083376
notis - AAJ8710
oclc - 05205740
System ID:

This item is only available as the following downloads:

Full Text








Thanks are extended to those individuals who provided me with their

comments, criticisms, and time by serving as members of my doctoral

committee. These include Dr. Marvin Shaw, Dr. Joel Cohen, Dr. Larry

Severy, Dr. Robert Watson, Dr. Thomas Simon, and Dr. William Yost. Also,

I wish to thank my graduate student cohorts (including Doug Tuthill, Mark

Leary, Rody Miller, and Teddi Walden) for their help and assistance, as

well as the introductory psychology instructors for facilitating subject

recruitment. Lastly, I would like to thank my friend and mentor, Dr.

Barry R. Schlenker, for guiding my educational and research activities

during my graduate career.






ABSTRACT . . . .



The Attribution of Responsibility .
The Evaluation of Action . .
The Attribution of Morality . .
The Perception of Aggression .
Actions, Responsibility, and Morality:
Conclusions . .


S ix

. 3
. 9
. 15
. 16

. 18


The Cognitive-Developmental Approach. ... .22
Piaget's Cognitive-Developmentalism. 24
Kohlberg's Six Stages of Moral Reasoning .... 27
Applications and Limitations of the
Cognitive-Developmental Approach ... 29
Psychological Processes and Philosophical Theory. .. 36
The Ethics of Aesthetics and Utility ...... .38
The Ethics of Idealism and Rule-
Universalism. . ... .40
Applications and Limitations of the
Ethical Ideology Approach . ... 48


. 52

Scale Development . .
Step One: Initial Factor and Item
Analysis . .
Step Two: Final Revision of the EPQ. .
Step Three: Scale Characteristics. .
Hypotheses and Methods . .
Predictions . .
Subjects . .
Procedure . .



Hypotheses .. . .... 76
Hypothesis I . . 76
Hypothesis II . . 76
Hypothesis III . . 77
Method . . . 77
Results. . . .. 79
Hypothesis I . . 79
Consequences and the Attribution of
Morality . . .. 88
Experiment I: Summary of Results .. 92


Hypotheses . . 96
Hypothesis I . . 97
Hypothesis II . . 98
Method . . . 99
Results. ... . . 101
Attribution of Responsibility .. 101
Evaluation of the Outcome . .. 106
Attribution of Morality. . .. 113
Responsibility and Morality. . 116
Experiment II: Summary of Results . 116


Hypotheses . . 120
Hypothesis I . . 123
Hypothesis II. . . 123
Hypothesis III . . 123
Methods. . . .. 124
Results. . . 126
The Consistency of the Action . .. 126
Attribution of Responsibility .. 135
Attribution of Morality . ... 136
Experiment III: Summary of Results . 155



Part 1: Items Included in the Original
Ethics Position Questionnaire (EPQI) .. 168
Part 2: Items for the Initial EPQ Scale Revision. 171


B EXPERIMENT I . .. .. .176

Part 1: Situations Used in Experiment I ...... 176
Part 2: Dependent Measures Used in Experiment I 178


Part 1: Situations Used in Experiment II. ... 181
Part 2: Dependent Measures Used in Experiment II. .. 183

D EXPERIMENT III . .... .. 184

Part 1: Consequences Used in Experiment I .. 184
Part 2: Dependent Measures Used in Experiment I 185

REFERENCE NOTES . .... .. 186

REFERENCES. . . . 187




1-1 Heider's Levels of Responsibility .

3-1 Summary of the Factor Analysis of the Initial Set
of Items Used in Generating the Revised EPQ .

3-2 The Ethics Position Questionnaire (EPQ) .

3-3 The Idealism-Pragmatism Scale . .

3-4 The Absolutism-Relativism Scale .

3-5 The Generality Scale . .

3-6 The Reliabilities of the Ethics Position
Questionnaire (EPQ) . .

3-7 Summary of the Scale Characteristics of the Ethics
Position Questionnaire (EPQ) . .

3-8 Random Orders of Materials . .

4-1 The Effects of Scenario and Rule-Universalism
on Attributions of Morality . .

4-2 The Effects of Ethical Ideology and Scenario on
Perceptions of Actor's Motives . .

4-3 The Effects of Scenario and Idealism on Males
and Females Attributions of Morality .

4-4 The Effects of Scenario and Universalism on Males
and Females Attribution of Morality .

4-5 The Effects of Scenario on Males and Females Judgments
of Whether the Action Should Have Been Done and the
Attribution of Responsibility . .

4-6 Relationship between the Evaluation of Positive and
Negative Consequences and the Attribution of Morality.

4-7 The Effects of Scenario, Outcome Quality, and Ethical
Ideology on Evaluations of Consequences .


S 7

. 55

. 58

S. 63

. 64

. 65

. 66

. 67

. 74

. 81

. 83

. 85


. 87

S. 89

S. 93

5-1 Summary of Analysis of Variance. . ... 102

5-2 Effects of Level of Responsibility and Outcome
Quality on Attribution of Responsibility, Outcome
Evaluations, and the Attribution of Morality .. 103

5-3 Effects of Level of Responsibility and Outcome
Quality on Males and Females Attribution of
Responsibility and Evaluations of Outcomes .. 105

5-4 Effects of Idealism and Level of Responsibility on
lMales'and Females' Outcome Evaluations .... 1.08

5-5 Effects of Idealism, Outcome Quality, and Level of
Responsibility on the Evaluation of the Outcome
and Attribution of Morality. . ... 109

5-6 The Effects of Ethical Ideology, Outcome Quality,
and Level of Responsibility on Outcome Evaluation. ... 112

5-7 The Effects of Ethical Ideology, Outcome Quality,
and Level of Responsibility on the Attribution
of Morality. . .. . 115

6-1 Summary of Analysis of Variance. . ... 127

6-2 The Effects of Standard Consistency and Outcome
Quality on Evaluations of an Action's Compatibility
with Moral Norms . .... 131

6-3 The Effects of Standard Consistency and Outcome
Quality on Males and Females Evaluations of an
Action's Compatibility with Moral Norms. ... 131

6-4 The Effects of Outcome Magnitude and Outcome
Quality on Evaluations of an Action's Compatibility
with Moral Norms . . 133

6-5 The Effects of Standard Consistency and Outcome
Quality on Evaluations of an Action's Compatibility
with Four Moral Norms. ... .. . 133

6-6 The Effects of Outcome Magnitude and Standard
Consistency on Evaluations of an Action's Compatibility
with Four Moral Norms. . . .. 134

6-7 The Effects of Outcome Quality and Standard
Consistency on Attribution of Responsibility ...... 134

6-8 The Effects of Standard Consistency and Outcome
Quality on the Attribution of Morality ... 137

6-9 The Effects of Outcome Magnitude and Outcome
Quality on Attribution of Morality . .... .137

6-10 The Effects of Standard Consistency, Outcome Quality,
and the Attribution of Morality. . .. .138

6-11 The Effects of Standard Consistency and Ethical
Ideology on Attributions of Morality . 143

6-12 The Effects of Ethical Ideology, Outcome Magnitude,
Outcome Quality, and Consistency with Standards of
Morality on Attributions of Morality . 146

6-13 The Effects of Consistency with Standards on Males
and Females Attributions of Morality . .. 154



Figure Page

2-1 The Taxonomy of Ethical Ideologies. . ... 42

6-1 The Effectsof Standard Consistency, Outcome Quality,
and Outcome Magnitude on the Attributions of Morality 140

6-2 The Effects of Standard Consistency and Ethical Ideology
on Attributions of Morality . 144

6-3 The Effects of Ethical Ideology, Outcome Magnitude,
Outcome Quality, and Consistency with Standards of
Morality on Attributions of Morality. . ... 149

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



Donelson R. Forsyth

August, 1978

Chairperson: Barry R. Schlenker
Major Department: Psychology

A series of experiments was conducted to investigate an attribution-

based theory of moral judgment which emphasizes the impact of three factors

on moral evaluations: the perceived quality of the actions committed by a

person, the responsibility of the person for the actions, and the ethical

ideology of the attributor. These factors were investigated in three

experiments in which subjects made moral evaluations about described

actors and actions. The first factor, the evaluation of action, was found

to be primarily influenced by 1) the consequences of an action and 2) its

compatibility with norms of morality. In general, the evaluation of an

act was more favorable the more positive its consequences and the more

consistent it was with moral norms. However, if the outcomes were

extremely positive or extremely negative, highly positive outcomes served

to lessen blame for violating a moral norm and enhanced praise for con-

forming to a norm, while negative outcomes led to a reduction in

attributed morality after both violations of and conformity to norms.

The second factor, actor's responsibility, also influenced moral

attribution. If the actor was merely associated with or only accidentally

caused an outcome, then the individual was not blamed if the consequences

were negative nor commended when the consequences were positive. However,

if the actor produced an outcome which he or she foresaw, intended, or

produced under environmental pressure, positive outcomes led to greater

attribution of morality than negative outcomes. Thus, attribution of

responsibility was found to be inversely related to attribution of

morality after negative outcomes, but directly related to moral attri-

bution after positive outcomes.

Lastly, moral evaluations varied dependent upon the ethical ideology

endorsed by the individual. Following the classification suggested by

Schlenker and Forsyth, the Ethics Position Questionnaire (EPQ) was

developed to measure two factors relevant to moral evaluation: degree

of idealism and acceptance of universal moral principles. The two scales

which comprised the EPQ were generated using factor analysis and

traditional scaling techniques, and analysis of the scales indicated

both were psychometrically adequate. When dichotamized and crossed, the

EPQ yields a typology consisting of four different ethical ideologies --

idealistic-absolutism, pragmatic-absolutism, idealistic-relativism, and

pragmatic-relativism. The attributions of individuals classified as

representative of each of these ideologies suggested the idealistic-

absolutists attributed the least morality when judging an actor whose

behavior produced any negative consequences. They were also the most

influenced by the normativeness of the behavior, evaluating non-normative

actions more negatively than all other attributors. Pragmatic-

absolutists, on the other hand, were generally the most favorable

in their attributions, perhaps because of their emphasis of the positive

consequences of action. For these attributors, the evaluation of

positive consequences and not the evaluation of negative consequences

was related to attributions of morality.

Idealistic and pragmatic-relativists were the most likely to allow

highly positive consequences mitigate blame for actions which were

inconsistent with a moral standard. This was particularly true for

pragmatic-relativists who, for example, attributed more morality to the

person who lied in order to produce a positive outcome than to the person

who told the truth but caused a negative outcome. Idealistic-relativists'

attributions of morality, unlike the attributions made by individuals

in all the other ethical ideologies, were correlated with their evaluations

of both positive and negative consequences, suggesting they take into

account both types of information when making attributions.



Interacting individuals must frequently determine the moral

acceptability of performed or planned actions. Since immoral behavior

incurs negative sanctions in many interpersonal situations, participants

must be able to predict how other people are going to evaluate the things

they do if they wish to avoid such sanctions. In addition to anticipat-

ing the reactions of others to personally performed behavior, the moral

quality of other people's actions must also be appraised. A pattern of

conflict-free interaction implies that individuals are able to both

restrict their behaviors so that they do not conflict too greatly with

society's conception of morality, as well as make ethical judgments

about other's actions. Since future interactions are often greatly

affected by any conclusions drawn about morality, judgments of ethicality

can influence a wide range of individual and interpersonal behavior.

In the past the study of moral judgment has been relegated by

psychologists to the realm of moral philosophy. More recently, however,

interest in the psychology of moral judgment has been revived largely

through research investigating individual variations in moral reasoning

and the determinants of responsibility allocation. Rather than providing

support for the idea of uniformity in moral reasoning, the research of

developmental psychologists such as Piaget (1932/1960) and Kohlberg

(1968) suggests that people make their judgments in different ways.

While not disagreeing with those philosophers and psychologists who note

the high degree of agreement between adult judges concerning the morality

of "clear" cases, these researchers suggest that this uniformity may

reflect similarity in conclusions rather than similarity in moral judg-

ment processes. Individuals may employ different standards, different

comparative processes, or wholely disparate criteria in their judgments

of the behaviors of others, and while these different processes often

result in agreement across judges, the moral reasoning underlying this

apparent agreement is quite different.

The second area of research that has stimulated interest in moral

judgment processes has focused on the factors which will lead a perceiver

to hold other individuals responsible for actions and their consequences.

The notion of responsibility is fundamental to a system of ethics since,

as the prominent philosopher of law H. L. A. Hart (1968) notes, it is the

mediating link between the individual and societal sanctions. An

explanation of responsibility allocation in everyday interactions has

been advanced by researchers working within attribution theory, which

is a broad social psychological framework that describes how individuals

make inferences about what characteristics other people possess. Although

the attribution theory approach is generally concerned with how people

determine when behavior reflects the existence of underlying traits,

theory and research suggest that responsibility determination lies at

the heart of the attribution process.

This treatise draws together both research investigating how

people differ in their approaches to making moral judgments and studies

of attributional processes to specify some of the major determinants of

moral judgment. Chapter One begins by placing the question into an

attribution theory framework which assumes that moral judgment funda-

mentally depends upon two integrally related processes: the determination

of responsibility for action and evaluation of the moral quality of the

act itself. Chapter Two reviews the theory and research of the cognitive

developmentalists Piaget and Kohlberg relevant to individual differences

in moral reasoning, as well as several alternative approaches to indi-

vidual variability. One such approach, suggested by Schlenker and

Forsyth (1977), was used as the model for a measure of individual ethical

ideology, and the characteristics of the measurement instrument which was

developed are discussed in Chapter Three. Three experiments, which

examined the predictive utility of the measure and several hypotheses

about the effects of responsibility, consequences, and ethical ideology

on moral attribution are described in Chapters Four, Five, and Six,

while Chapter Seven discusses the obtained results and suggests several


The Attribution of Responsibility

Complex forms of social behavior require that interacting individuals

can competently perceive and interpret cues concerning the qualities and

characteristics of others, can decipher and understand the nature of on-

going interaction, and can gather and weigh the available information

pertaining to the causes of environmental and interpersonal events.

Despite the seeming complexity of these tasks, most people can perform

them well enough, without any apparent directed effort or organization

of the cognitive processes involved. In explanation, Heider (1958)

proposed that individuals achieve an understanding of the world around

them by utilizing certain intuitive "common-sense psychology" principles;

principles which offer basic explanations of the events, individuals,

and entities they encounter. As Heider (p. 2) states, "the ordinary person

has a great and profound understanding of himself and of other people

which, though unformulated and only vaguely conceived, enables him to

interact with others in more or less adaptive ways."

Attribution theory, a social psychological analysis of the processes

involved in perceiving the causes of behaviors and events, focuses on

what information people employ to gain an understanding of their social

environment. In his highly influential The Psychology of Interpersonal

Relations (1958), Heider proposed that the perceiver's goal, like that

of the scientist, is to understand reality and to become able to predict

and control events in the environment. The individual in a social world

accomplishes this goal by achieving an understanding of the relatively

stable foundational properties of the environment that produce the more

dynamic and variable events. These attributions about inferred under-

lying causal sources enable the perceiver to systematically organize

observed behaviors and promotes extrapolation to other possible situations.

The attributor's world is thus reduced from a meaningless chaos of events

to an understandable unfolding of predictable events and outcomes.

Heider's attribution theory, in part derived from the insights of

Piaget (1932/1960), carefully examined the effects of how inferences made

about other interactants influence the course of social behavior. Heider

and later attribution theorists such as Jones (Jones & Davis, 1965) and

Kelley(1967) propose that one of the most important factors influencing


moral attribution is the degree to which the actor is seen as responsible

for producing the act under evaluation. Within attribution theory,

responsibility refers to the degree of perceived relationship between

an actor and an act; the extent to which the actor is seen as linked to

the act in a unit relationship (Heider, 1958). An individual who is

perceived to be responsible for some action is judged to be accountable

or answerable for that activity, or judged to be open to positive or

negative sanctions as a consequence of being linked to the event (cf.,

Heider, 1958; Shaw & Sulzer, 1964). All propose that individuals can

infer a great deal about a person's dispositions and characteristics by

observing that person's actions in various situations, but that any

attributions made are dependent to some extent on responsibility

attributions. For example, if a person is forced into performing some

act, then the nature of the action and the consequences are not, in a

strict sense, caused by the actor alone. The actor is therefore less

responsible for the act than if he freely chose to perform it, and so

little can be said about the actor's underlying disposition toward the

activity. What is required is a situation in which the person freely

decides upon what course of action to take and can determine what

consequences are to be sought after and achieved.

The extent to which the actor is seen as responsible.depends upon

the perceived influence of internal and external forces on the activity

or actor under scrutiny. For Heider (p. 56), "behavior can be ascribed

primarily to the person or to the environment; that is, behavior can be

accounted for by relatively stable traits of personality or by factors

within the environment." Focusing on the balance between internal and

external determinants of behavior, Heider suggested five levels of

responsibility attribution which differ by degree of intentionality,

association, and external justification. These five levels, as described

by Shaw and Sulzer (1964, pp. 39-40), are presented in Table 1-1. At the

first level, that of association, the actor is held responsible for any

actions which he is connected with in even the slightest way, no matter

how remote. At the level of causality, however, the actor is held

responsible only for effects that he has caused in a more direct sense.

At this level the degree to which the actor could foresee the consequences

of his actions is not taken into account by the attributor as it is at the

third level, foreseeability. The individual at this level is held to be

responsible for the produced event even though he may not have "set out

to achieve it, either as a means or an end" (Hart, 1968, p. 120). At the

fourth level, intentionality, attribution of responsibility depends not

only on the demonstration of foreseeability of consequences, but on

intentions as well. This is the most psychologically informative kind

of perceiver data, available when a person voluntarily performs a given

action in order to achieve a certain foreseeable goal of a certain

societally determined quality.

At the last level, justification, the person is held only

partially responsible for intentional actions due to the influence of

various factors. Although Heider and Shaw and Sulzer emphasize the

impact of external factors, internal factors can also reduce responsibility

at this level. Heider draws the distinction between internal factors

which determine if the actor "can" perform the action and those external

variables which influence whether or not the actor "may" perform the

action. Similarly, Steiner (1970) suggests that perceived freedom

Table 1-1

Heider's Levels of Responsibility

LEVEL I: ASSOCIATION (Global Association)

The person is held responsible for any effect that he
is connected with in any way. Responsibility is determined
by pseudocausal reasoning rather than by consideration of
objective causal connections. Thus, the individual may
be blamed for harmful acts committed by friends.

LEVEL II: CAUSALITY (Extended Commission)

The person is held responsible for any effect that he
produced by his actions, even though he definitely could
not have foreseen these consequences.


The person is held responsible for any foreseeable effect
that he produced by his actions even though the effect
was not a part of his goals or intentions. He is held
responsible for the lack of restraint that a wider cognitive
field would have produced.

LEVEL IV: INTENTIONALITY (Purposive Commission)

The person is held responsible for any effect that he
produced by his actions, foreseeing the outcome and
intending to produce the effect.

LEVEL V: JUSTIFICATION (Justified Commission)

The person is held only partly responsible for any effect
that he intentionally produced if the circumstances were
such that most persons would have felt and acted as he
did. That is, responsibility for the act is at least
shared by the coercive environment.

Note: Source is Shaw and Sulzer, 1964.

depends upon judgments of the actor's (a) ability to perform the action

and (b) the restrictiveness of environmental influences. For example,

if a man witnessing a drowning does not possess the ability to swim,

then his responsibility for the failure to save the victim's life would

be lessened. A similar reduction in responsibility would also be observed

if environmental factors, such as sharks or a strong undertow, prevented

successful intervention in the emergency.

Heider's model of attribution can be interpreted in two related

ways. He implies that individuals, depending upon developmental and

situational factors, may be level-specific in their attributions, and

second, that attribution of responsibility is greatest for actions

intentionally caused by the actor who foresees the relevant consequences,

but decreases with the introduction of such factors as unforeseeability

and external pressure. As an example of the first interpretation, in

some cases individuals may base their responsibility judgments on the

elements at the level of foreseeability rather than considering such

aspects as intentionality or coercive pressure (cf. Fishbein & Ajzen,

1973; Ross & Ditecco, 1975; Shaw & Sulzer, 1964). Ross and Ditecco

emphasize this interpretation, suggesting that moral judgment will be

greatly affected by the concept of responsibility adopted by the evaluator.

They note that four factors influence the sophistication of attributions

of responsibility made by adults, including specification by an external

source (as when a judge instructs a jury as to the degree of responsi-

bility needed to find a person guilty of first degree murder), motivations

(such as ego-defensiveness), semantics (e.g., the impact of labels such

as "traitor" versus "patriot"), and the context of the interaction (e.g.,

between short-term versus long-term friends).

The second interpretation of the levels suggests that they specify

the key elements in the responsibility attribution process. Empirical

investigations of Heider's levels of attribution of responsibility

attribution have generally supported this interpretation of the theory

by finding a curvilinear relationship between attribution of responsibility

and responsibility level (e.g., Shaw, 1968; Shaw, Briscoe, & Garcia-Esteve,

1968; Shaw, Floyd, & Gwin, 1971; Shaw & Reitan, 1969; Shaw & Sulzer, 1964).

For example, Shaw and Sulzer (1964) contrasted differences in attributional

tendencies for adults and second graders in the initial investigation of

Heider's levels. After presenting subjects with a series of short

scenarios that described different situations containing the minimum

variables necessary to elicit attribution at each of the five levels,

subjects were asked to rate the responsibility of the actor in each story.

Although the outcome of the action affected the children's attributions

more than the adults' attributions, both groups did increase responsi-

bility allocation over the first four levels. In addition, responsibility

dropped for the level five situations.

The Evaluation of Action

A young couple, before deciding to get married, live together for

over a year in a small apartment; three black men, barely old enough to

be out of high school, rob and kill a liquor store owner; a clerk, working

two jobs to put his younger brother through college, volunteers to serve

as a youth counselor in his spare time; a father viciously beats his

child for no apparent reason; a student cheats on an exam; a housewife

cancels plans to join an elite social club in order to have enough time

to do volunteer work at the local hospital; a burglary suspect lies to

his interrogator; a young man saves three children from death in a

burning house but dies himself while trying to rescue a fourth. All

these examples describe people performing actions which can be

appraised as either morally "good" or morally "bad." Attribution

theory, with its emphasis on the allocation of responsibility, describes

the factors which determine when a person's actions are used as

indicators of his dispositions, but the importance of the evaluation

of the action itself should not be overlooked. Clearly, how a person

is morally appraised is determined not only be degree of responsibility,

but by the moral evaluation of the action as well.

Heider attempts to incorporate action evaluation into his analysis

of moral judgment via the concept "ought." According to Heider, an ought

is defined as a normative expectancy which serves as a standard to which

all behavior can be compared. These oughts are "standards of what ought

to be done or experienced, standard independent of the individual's

wishes" (p. 219). Heiter cites Asch (1952, p. 357) who points out that

"action that fits the requirements we judge to be appropriate or right;

to fail to act appropriately we experience as violating a demand, or being

unjust." This usage of ought is similar to Hollingworth's (1949, p. 3),

who writes "the key to morals or ethics is the feeling of obligation, the

recognition of imperatives in thought and action." Behavior which is

irrelevant to this feeling of obligation "transforms mere behavior into

conduct." Thus, moral judgment depends fundamentally on evaluating conduct

in relation to some standard or set of expectations. If the action is

consistent with prescriptive standards then, by definition, it ought to

be done and is judged to be moral. If, on the other hand, the action is

one which ought not be done since it violates some proscriptive standard,

then the action is immoral.

Heider's concept of ought carries with it two assumptions. First,

oughts are taken to be trans-situationally applicable. For any particular

moral evaluator, oughts "refer to invariant standards, to 'laws of conduct'

which hold in spite of many variations in incidental or momentary factors"

(p. 230). While there are certain limits to these ought invariances, such

as when the circumstances of a situation are judged to be unique, in

general the same ought should be applicable to a range of relevant

situations. Second, oughts are assumed to have transpersonal validity.

As would be consistent with an intuitionist ethical philosophy, oughts

can be directly perceived; therefore, irrespective of personal ethical

values or culture, all people "should perceive the same ought require-

ments in a particular situation" (p. 222). Heider summarizes his

explanation of the concept of ought by describing it as "a cognized

wish or requirement of a suprapersonal objective order which has an

invariant reality, and whose validity therefore transcends the point of

view of any one person" (p. 222).

Limitations of Heider's 'ought'. While the concept of ought or

duty is recognized by many philosophers as an important element in the

moral sphere, its application in a psychological explanation of moral

judgment encounters several problems. First, the term has an ambiguous

meaning due to its frequent and long-standing use in philosophical and

lay discourse. Hollingworth (1949), for example, contends that the word

ought has as many as ten distinct meanings, and philosophically the term

seems to have almost as many different meanings as philosophers who use

it (cf. Aiken, 1952; Kant, 1873/1973; Prichard, 1952; Sellars, 1952).

Second, the approach may preclude the possibility of an individual

being a moral skeptic who does not rely on ought principles in making

moral judgments of others. Such an approach to ethics has been taken by

several philosophers such as Hume and Fletcher (Jones et al., 1962), who

argue that situational variations make general statements about what ought

to be done inapplicable. Although the existence of a prescriptive

philosophical position which rejects the assumption that inviolate

moral principles can be formulated does not "prove" that these oughts

are not utilized at the psychological level, the existence of individuals

who can be regarded as "common-sense moral skeptics" is an empirical


Lastly, simply stating that an action that is compatible with an

ought is judged to be moral while an action that is incompatible is

judged to be immoral does little to clarify the nature or origin of

oughts themselves. An ought is conceptually a universal norm which is

a part of the interpersonal regulatory system of all societies. How-

ever, the existence of such universal norms has yet to be supported

empirically. Also, Heider fails to provide answers to such puzzles as

"How does the ought arise?"; "What situations are unique enough to prevent

the application of ought requirements?" (e.g., Why is killing during war-

time condoned?); and "What universal features of human existence can

account for the invariant perception of social oughts?"

Oughts and actions, actors and achievement. A partial solution to

the problem of oughts as standards for interpersonal judgment has been

suggested by Kelley (1971). Kelley's main objective is to deal with moral

evaluation within the context of other types of attributional processes,

with particular focus on reality evaluation and achievement evaluation.

He points out that two characteristics which Heider ascribes to oughts

are consistency across time and modalities and consensus among persons.

These two features are the major defining qualities of an external

attribution (see Kelly, 1967), and so oughts are seen as being non-personal,

external entities that are objectively real. Ought perceptions thus

follow the same process as that of reality evaluation, which relies

on consistency, distinctiveness, and consensus information.

For example, an attributor may believe that murder ought not be

committed, that it is immoral. The individual may attribute this ought

and moral reaction to either an external source or an internal source.

If internal, the person could conclude that he is just the "kind of a

person who doesn't like to see people murdered." If external, however,

the person could conclude that murder is wrong, that it is the act itself

which determines his reaction rather than his own personal characteristics.

Kelley maintains that an external explanation is supported by the

perceiver's common-sense attributional data since (a) consensus among

others who judge that act is high, that is, they all condemn murder,

(b) the person's reaction is consistent across time and situations, and

(c) reaction to this particular class of behaviors is distinctive; that

is, the attributor does not react in a similar manner to other kinds of

interactions. Thus oughts become a "matter of fact, to be discriminated

and judged consistently and to be validated concensually, just as other

aspects of reality are so discriminated and validated" (Kelley, 1971,

p. 294). Actions which violate oughts are judged immoral, while actions

which are consistent with oughts are judged moral. Concensus among

members of a particular culture regarding an ought need not be perfect,

although this is the usual case with some types of behavior (e.g.,

murder, theft, rape, etc.).

Kelley is most concerned with the conditions which determine when

the individual will be judged responsible for a morally "good" action,

and not with the conditions which determine the moral evaluation of a

person who performs a "bad" action. In the context of the evaluations

of positive actions, the attributed externality of oughts has a great

impact. Kelley suggests that because oughts are seen as external

entities, they may be perceived as exogenous causal forces which require

the individual to act appropriately in the given situation (cf. Bem's

"mand," 1967). Thus, the attributor may not conclude that any special

characteristic in the observed individual need be inferred when action

is ought compatible since an external, interpersonal factor -- the

ought -- is the actual cause of the "good" behavior. For example, the

individual who refrains from murdering his neighbor is not seen as being

particularly moral since his behavior is clearly ought-required. As

Ross and Ditecco (1975) suggest, "there is an asymmetry in assignment

of credit or blame. Blame is often attributed for failure to adhere to

an ought standard, but mere adherence may receive little praise" (p. 93).

Although Kelley's analysis clarifies the origin of the assumption

that oughts are universal, his analysis ignores an important distinction

between judging a person to be moral and judging a person to be morally

commendable. Certainly examples can be cited in which the individual

who performs an action which is prescribed by an ought voluntarily,

intentionally, and with foreseeability of consequences is judged to be

moral. For example, if an attributor (P) embraces the ought "people

should contribute small sums of money to charity" and observes a person

voluntarily and intentionally give a small sum to charity, then it does

not seem improbable for P to conclude that the giver is moral. However,

to be seen as morally commendable, the observed individual would need to

give a large sum to charity to actually surpass the action required by

the norm.

In addition, although high agreement is often obtained among

individuals of the same cultural grouping concerning the morality of

certain kinds of actions, far less agreement is achieved when the moral

quality of other types of actions is discussed. People will, for example,

often agree that it is wrong to lie, steal, murder, or rape, but dis-

agree about whether abortion, mercy-killing, or premarital sex is


The Attribution of Morality

The essence of moral judgment is captured and formalized in the

dynamics of the courtroom. Judge and jury, who take the part of the

attributor, are exposed to information that pertains to the defendant's

(actor's) responsibility for some legally questionable action. The

prosecution is primarily concerned with establishing that 1) some action

has been done which runs counter to a judicial statute and hence is

illegal and 2) that the defendant is the person responsible for committing

the act and producing its consequences. The defense, on the other hand,

presents evidence which is designed to disprove either one or both of

these contentions and hence protect the defendant from the possibility

of sanction.

As with judicial proceedings, interpersonal attributions concerning

the morality of others depend upon the perceived relationship between

the actor and action plus the nature of the act itself. The work of both

Heider and Shaw demonstrates that the responsibility attributed to the

individual varies depending upon such factors as judged foreseeability

of consequences and the freedom and intentions ascribed to the actor.

However, even if the actor is viewed as responsible for the action and

its consequences, this does not necessarily imply that the individual's

morality is reflected in his actions. In many instances, the act

itself is evaluated as one which is either morally commendable or

morally reprehensible before attributions about the actor's ethicality

are made. For example, a student who is responsible for his failing

performance on a psychology test may be judged as ignorant of certain

facets of psychology. However, he would not be considered immoral unless

failing the examination was judged to be an immoral action.

The Perception of Aggression

The impact of both responsibility and action evaluation on the

attribution of morality is apparent in recent research (Feshbach, 1971;

Tedeschi, Smith & Brown, 1974) investigating when an action will be

labeled "aggression." While Tedeschi et al. conclude that the scientific

usage of the term aggression has proven problematic, they do suggest that

aggression is "important as a label used by naive observers in character-

izing actors" since the "labeling of behavior as aggressive has an

important function in the assignment of moral responsibility and

mediation of retribution" (p. 557). Tedeschi et al. propose that the

analysis of the common usage of the term reveals that aggressive actions

are those that are not only viewed as intended constraints on the

behavioral alternatives of another person, but also must be evaluated

by the perceiver as anti-normative and illegitimate. While other

factors (i.e., physical characteristics) influence the attribution of

aggression, only the three characteristics of intentionality, constraint,

and illegitimacy are necessary and sufficient conditions.

Similarly, Feshbach (1971, p. 289) considers how violence is

evaluated by members of society and emphasizes the situational nature

of violence evaluation; "violent acts with the same formal character-

istics -- for example, one man killing another -- are evaluated by most

people very differently depending on the context in which the act occurs."

He suggests several factors which affect evaluations, with the most

important being the degree to which the act conflicts with either the

formal (i.e., laws) or informal ("oughts") norms of society. A second

factor, the self-serving nature of the action, also influences the

categorization of a violent act, as does the actor's degree of respon-

sibility for the action under scrutiny. Feshbach suggests that the

degree of responsibility attributed to the actor is influenced by the

ascribed level of choice available (as indicated by the number of

alternatives open to the individual), the initiator of the interaction

(defensiveness or offensiveness of the action), the degree to which the

actor is aware of the nature and quality of the acts performed, the

stability of the actor in terms of ability to distinguish right from

wrong, intentionality, and degree of ego-control. Finally, miscellaneous

influences are suggested, including "normative considerations of fair

play, the degree and manner of the violence, the age and sex of the

victim, and more generally, the appropriateness of the target" (p. 290).

Although neither Tedeschi et al. nor Feshbach provide a clear

theoretical structure for dealing with the evaluation of violence, their

analyses do summarize the basic processes involved in moral attribution

by emphasizing both responsibility and act evaluation. Also, both note

the importance of the circumstances surrounding the action on its

evaluation since a great many factors can determine whether the act is

viewed as consistent or inconsistent with a moral standard. Lastly,

each denies that evaluative conclusions reached concerning actions will

be universally shared since the perspective of the attributor is a

crucial variable to consider. Feshbach points out that the attributor's

attitude towards the objective of the violence will have great effects

on the evaluation reached, and Tedeschi et al. conclude "that no action

can be identified as aggressive or violent without taking into account

the value system of the perceiver" (p. 557). Even within a single

culture, variations in personal values have been found to greatly

affect the moral evaluation of violent actions (e.g., Feshbach & Feshbach,

1969; Opton & Duckless, 1970). This problem raises the issue of the

impact of individual variations on moral judgment, which is the topic

dealt with in detail in the next chapter.

Actions, Responsibility, and Morality: Conclusions

A conceptualization of moral judgment based on attribution theory

and research proposes that moral attribution, defined to be the process

by which an individual formulates appraisals of the moral goodness or

badness of others, is primarily determined by the interfusion of

attribution of responsibility and action evaluation processes. In a

simple instance, the moral attribution process begins with an attributor

who considers information concerning the quality of the particular behavior

under scrutiny, and the actor who performed (or is perceived to be related

in some other way to) the behavior. While many factors could potentially

influence evaluations of actions, this analysis emphasizes how the conse-

quences of the action and the compatibility of the action with moral

standards influence action evaluation. All other things being equal, the

evaluation of an act should vary directly with the quality of the conse-

quences produced, such that actions which result in "good" outcomes are

more positively evaluated than ones which produce negative outcomes.

However, the evaluation of an action will also be influenced by the

compatibility of the act with moral standards or "oughts." As several

theorists suggest (e.g., Heider, 1958; Kelley, 1971; Kohlberg, 1971),

moral evaluations are greatly influenced by whether or not the perceiver

believes the action is consistent or inconsistent with society's norms of

morality. As Shaw's (1976, p. 250) conceptualization of norms suggests,

moral norms are interpersonal "rules of conduct established by the members

of the group to maintain behavioral consistency." These standards for

appropriate action "represent standardized generalizations concerning

expected behavior in matters that are of some importance to the group" and

they "refer to what should be done" in any given situation. At a general

level, then, the attributor's evaluation of an action depends upon (a)

the quality of the consequences of the action and (b) the act's consistency

with moral norms.

The second factor in the attribution of morality process involves

judgments of the actor's responsibility for the action. As attribution

of responsibility theorists (e.g., Heider, 1958; Shaw & Sulzer, 1964)

note, greater responsibility will be attributed to the actor to the

extent that internal, personal factors are perceived to have caused the

outcome rather than external, situational factors. The moral evaluation

of a previously unknown actor thus depends to a great extent on respon-

sibility and action ascriptions. The greater the responsibility of the

actor for the action, then the greater the perceptual correspondence

between the moral evaluation of the action and the moral evaluation of

the actor. While other factors, such as prior estimates of the actor's

morality and the ethical perspective adopted by the attributor may also

influence moral judgment, attributionally action evaluation and attribution

of responsibility are crucial determinants of the attribution of morality.



In the late 1950's, Lawrence Kohlberg began investigating some of

the factors that influence peoples' moral reactions to others (1959).

Not surprisingly, he found that when people were asked to make judgments

about others described in various ethical dilemmas that he had constructed,

judges disagreed about the morality of the actors. Some would condemn

the person as immoral, while others would commend the person for his

morality and sense of duty. More interesting, however, was Kohlberg's

conclusion that the reasoning processes underlying these judgments about

morality of others did not seem to be the same across all the judges.

In fact, there was apparently a wide range of variability in the approaches

people took in making moral judgments, with some individuals being

influenced more by certain variables, while other individuals focused

on a quite different set of factors.

Kohlberg's investigation of moral judgment, while certainly not the

first research into the process, did clearly indicate that individuals

may differ in the manner in which they make judgments of the morality of

others. Given the importance of understanding the impact of possible

individual variations on the attributional processes described in the

previous chapter, this chapter examines the question in detail. After

reviewing the cognitive-developmental approach to individual differences

as captured in the work of Piaget and Kohlberg, problems relating to

cultural biases, content orientation, and psychometric difficulties are

suggested. The cognitive-developmental approach is then contrasted with

Schlenker and Forsyth's (1977) examination of variations in individuals'

orientations toward ethical issues, and conclusions are drawn about the

adequacies of the two approaches.

The Cognitive-Developmental Approach

The cognitive-developmental approach to the dynamics of human

maturation has yielded one of the most influential theories of moral

development. The prominent theorizing of Piaget (1932/1960) and Kohlberg

(1959, 1964, 1968, 1969, 1971, 1976) shares a common focus on organism-

environment interaction that results in the development of increasingly

sophisticated cognitive structures in the organism. This interaction

leads to stages of cognitive development, which represent the restructuring

of the preceding cognitive schemata into increasingly complex structures

in a metastatic process. These stages represent qualitatively different

levels of cognitive processing, as thought is reorganized to increased

levels of cognitive integration and differentiation.

Moral judgment develops in a similar process, since the "age

developmental trends in moral judgment have a formal cognitive base

parallel to the structural base of cognitive development" (Kohlberg,

1969, p. 390). As with cognitive development, the process of moralization

moves through a series of discontinuous stages. Corollaries which

accompany the stage assumption include: (a) these stages are order-

invariant, so that the higher level stages cannot be reached except by

passing through the lower level stages; (b) while the order and number

of stages is invariant, the end-points of the developmental sequences

may differ for individuals, and the time required for passing through

the various stages may differ by individual and culture; and (c) the

movement from one stage to another is a gradual process, the result of

assimilation of incoming information to fit existing structure and the

accommodation of the cognitive structure to fit the incoming information.

Although a number of theories of morality are based on a stage approach

to development (cf. Bull, 1969; Bobroff, 1960; Kay, 1968; McDougall,

1908), only the two most prominent theories, those of Piaget and Kohlberg,

will be given detailed consideration.

Before examining the theories of Piaget and Kohlberg, however, their

relevance to understanding individual variations in making moral judgments

should be made explicit. Although the cognitive developmentalist is

concerned with the broader issue of the moralization of the individual,

the postulation of a developmental sequence suggests that individuals

can be at different stages, and that their moral judgments will necessarily

be influenced by the stage they have attained. Therefore, in order to

reach a complete understanding of the dynamics underlying individuals

moral judgments, the range of possible variations in the judgment process

must be explored. For example, Piaget suggests that certain key elements,

such as consequences and actor's intentionality determine moral judg-

ments. However, the impact of these variables depends upon the moral

development of the child; if he or she is not yet "morally mature,"

the consequence information may be more influential than intentions

information. Thus, the analysis of moralization yields a theoretically

defensibleclassification system which specifies important variations in

individuals' approaches to making moral judgments.

Piaget's Cognitive-Developmentalism

In his classic work, The Moral Judgment of the Child (1932/1960),

Piaget extends his studies of cognitive development to the area of moral

evaluations made concerning a variety of human behavior, including lies,

accidents, and thefts. Piaget's method consisted of presenting a child

with pairs of narratives that described two actions that culminated in a

given negative outcome. Children were asked to decide which of the two

perpetrators of the negative outcome was "naughtiest," and to explain

their choice. For example, to examine the effects of the severity of

the consequences of behavior on its moral evaluation, Piaget constructed

parallel stories, one describing how an actor accidentally produced a

large, negative outcome while attempting to perform some well-intentioned

action, while the actor in the comparison story produced a much smaller

negative outcome, but did so under mischievious circumstances.

Moral heteronomy. Drawing inferences from the children' responses

during these interviews, Piaget theorized that the evolution of moral

judgment in the child progresses through two major cognitive-developmental

stages. The first stage, that of heteronomy, is characterized by

obedience to others' orders and moral realism. Heteronomous thought, a

term which Piaget borrowed from Kant, occurs when the child believes that

it is right to obey the will of an adult and wrong to have self-will.

Moral realism refers to the child's tendency to "regard duty and the

value attaching to it as self-subsistent and independent of mind, as

imposing itself regardless of the circumstances in which the person may

find himself (Piaget, 1960, p. 111). The three major aspects of moral

realism include: (a) Good is defined as conformity to rules, which are

to be obeyed, not judged or interpreted; (b) the letter rather than the

spirit of the rule is to be obeyed, since the rule is not truly under-

stood nor internalized; and (c) the criterion for blame and punishment

is the degree of objective responsibility of the actor, which Piaget

defines as the evaluation of the "act, not in accordance with the motive

that prompted them, but in terms of their exact conformity to established

rules" (pp. 111-112).

This moral realism is the result of the "intersection of two kinds

of causes," egocentrism and the morality of constraint. Egocentrism, for

Piaget, is typified by the child's inability to decenter -- to shift his

or her attention from one aspect of an object or an event -- and by the

inability to perceive a situation from another's point of view. The

morality of constraint is characterized by unilateral respect, an

unexamined sense of duty, and by heteronomy. "The child accepts from

the adult a certain number of commands to which it [sic] must submit

whatever the circumstances; wrong is what fails to do so; the intention

plays a very small part in this conception, and responsibility is

entirely objective" (Piaget, 1960, p. 335).

Moral autonomy. Gradually, reorganization of the cognitive bases

of moral thought shifts moral judgment to a higher stage. This shift is

the result of a number of processes, particularly the reduction of

egocentrism and the replacement of the morality of constraint with the

morality of cooperation. The decrease in egocentrism is the result of

a growth in cognitive ability which allows the child to "take the role

of the other" in social interactions. The morality of cooperation, which

subsumes the ideals of mutual respect for others in reciprocity-governed

interaction, is generated through equalitarian interactions with peers

resulting in the internalization of rules and the reasons for their

importance. For Piaget's subjects, this shift occurred between the

years of seven and nine, and culminated in the attainment of the second

stage of moral development, autonomy. The child's behaviors are

governed by the adult morality of cooperation, so that the behaviors

are engaged in for their own sake and not because of contingent sanctions.

Rules are seen as necessary for the maintenance of cooperation, but may

be changed by mutual agreement. Subjective responsibility criteria

replace the objective, mutual respect usurps unilateral respect, moral

realism becomes moral relativism, and the concept of expiatory punish-

ment is replaced by reciprocal punishment.

The differences between heteronomous and autonomous stages of

moral thought are demonstrated in the following examples of two children'

responses after being told a pair of stories about how some cups were

broken; one story told of fifteen cups being broken unintentionally by

a child entering a room, and a second story depicted a single cup being

broken by a jam-stealing child (Piaget, 1960, pp. 125 & 130, respectively):

Constance (Age 7)
If you were their mother, which one would you punish most
---The one who broke the cups.
Is he the naughtiest?
Why did he break them?
---Because he wanted to get into the room.
And the other?
---Because he wanted to take the jam.
Which one of them would you punish most severely?
---The one who broke the fifteen cups.

Nus (Age 10)
Which of them is the naughtiest?
---The naughtiest is the one who wanted to take the jam.
Does it make any difference the other having broken more cups?
---No, because the one who broke the fifteen cups didn't
do it on purpose.

Kohlberg's Six Stages of Moral Reasoning

The theorizing and research of Kohlberg (1959, 1964; 1968; 1969;

1971; 1976) argues from a Piagetian-rooted cognitive-developmental

perspective that dynamic processes of cognitive growth result in the

progressive reorganization of perceptions, conceptual abilities, and

judgments of morality. Like Piaget, Kohlberg conducted interviews with

children, and he concluded that three distinct levels were reflected in

children' responses, with each level being comprised of two discernable

stages. At the preconventional level, obedience to authority is stressed

for hedonistic reasons such as maximization of personal gain or avoidance

of punishment. Stage 1 children, like Piaget's heteronomous thinkers,

accept without question imposed rules of conduct without interpreting

their fairness or bases. At Stage 2, however, focus is shifted towards

self-interest, as the child attempts to maximize his own personal gains

without consideration of the needs of others.

The hallmark of the second level, conventional thought, is conformity.

The child at this level is concerned with acting in accordance with the

expectations of society, and also attempts to support these rules. Stage

3 is characterized by attempts to gain credit by careful conformity to

stereotypical images of what are appropriate behaviors. Kohlberg notes

that intentionality affects responsibility delegations at this stage,

but that too much emphasis is sometimes placed on motives, with the result

that "nice" intentions are sometimes cited as mitigating excuses for a

careless action. Stage 4, on the other hand, is typified by an authority

and social order orientation. At this stage "right behavior consists of

doing one's duty, showing respect for authority, and maintaining the

given social order for its own sake" (Kohlberg, 1968, p. 26).

The final level, postconventional or autonomous, stresses inde-

pendence from concern with obedience to existing societal impositions.

Behavior is guided by inner principles, which are held apart from those

recommended by society. Stage 5 morality, which is apparently derived

from a social contract morality, stresses the difference between personal

values and legal rules of conduct. While laws are recognized as beneficial

for society-at-large, they are not held to be infallible and hence may be

altered. Finally, Stage 6 morality is humanistic, a morality of respect

for others as individuals, for the equality of all, and of justice as

equity; behavior is guided by individually derived moral principles

held by the individual to be comprehensive, logically consistent, and

universally applicable.

In sum, both Piaget and Kohlberg hypothesize that individuals

functioning at the various stages of moral development can be identified

by both the content of their rules and by the reasons they adopt them.

Piaget perceptively detects the tension between conformity requirements

and demands for individuality (cf. Kelley, 1971; Ziller, 1964), and

proposes that immature individuals resolve the dilemma by avoiding

individuality, while mature adults face the problem by embracing

equalitarian principles of authomomy, cooperation, and mutual respect.

Both the mature and immature are principled, but the principles of the

young are unexamined and blindly followed, while the principles of the

adult are internalized and understood. In addition, a single set of

principles or rules is accepted by the adult or "morally mature" members

of society, sharing similarity in content and motivational bases. For

Piaget, all mature individuals base their moralities on seeking autonomy

while maintaining cooperative and equitable relationships with others.

Although Kohlberg's analysis is more detailed than Piaget's, it shares

a similar stress on autonomy as an end-point of moral development and

the acceptance of rules for equalitarian reasons.

Applications and Limitations of the Cognitive-Developmental Approach

Although the research generated by the cognitive-developmental

perspective has been substantial, several problems are encountered when

the framework is used to explain individual variability in moral ideology.

Fundamentally, most of these criticisms revolve around two basic issues:

(a) mixing philosophical notions of ethics within a psychological theory

of moral judgment and (b) the methods used to measure stage attainment.

Philosophical foundations. The theories of both Piaget and Kohlberg

are based on an ethical philosophy that dates back to Aristotle and

finds expression in the ethics of such philosophers as Kant, Hobbes,

Bradley, and Frankena. In his Nicomachean Ethics (Thomson [trans.], 1953),

Aristotle stresses the idea that the highest good is what the individual

seeks "for its own sake and never for the sake of some other thing"

(p. 36). Morality can be achieved by being "self-sufficient," yet at

the same time conforming to the golden mean and avoiding activities

which conflict with accepted norms. Incorporating this notion of

morality into their scientific explanation of the psychology of moral

judgment, Piaget implicitly and Kohlberg explicitly assume that autonomy

and cooperation are the bases of "mature morality," making any morality

based on other considerations immoral or immature.

Simpson (1974, p. 87) argues that this assumption of both theories

undermines their scientific status, providing an example of the

"ascendency of the moral philosopher over the empirical scientist."

Focusing on Kohlberg's work, Simpson points out that (a) Kohlberg

collapses the often-noted science-description/ethics-prescription

distinction, (b) that the approach is not universal and content

irrelevant but restricted by its typically Western morality, and (c)

that Kohlberg's developmental research reflects a doctrine of social

evolutionism, ethnocentrism, and cultural bias. Simpson cites a number

of Kohlberg's published statements to lend support to her indictments.

These include Kohlberg's recommendation to individuals learning to

score responses to his scale that they read Frankena's (1963) book on

ethics; his observation that members of preliterate societies will not

be able to score better than Stage 3 in his schema; his frequent use of

"ought" and "should" in a nonpredictive, prescriptive sense; and his

acceptance of a framework that apparently indicates, because of its

structure, that females are less morally mature than males. Examples

of Kohlberg's more revealing statements, drawn from his paper entitled

"From is to ought: How to commit the Naturalistic Fallacy and get away

with it in the study of moral development" (1971), include:

There is a universal set of moral principles held by men in
various cultures, our stage 6. (These principles, we shall
argue, could logically and consistently be held by all men in
all societies; they would in fact be universal in all mankind
if the conditions for socio-moral development were optimal
for all individuals in all cultures.) At lower levels than
stage 5 or 6, morality is not held in a fully principled
form.(p. 173)

This [his research on moral development] suggests that
relativism, like much of philosophy, is the disease of
which it is the cure; the very questioning of the arbitrar-
iness of conventional morality presupposes a dim intuition
of non-arbitrary moral principles. (p. 180)

Our psychological theory as to why moral development is upward
and sequential is broadly the same as our philosophical
justification for claiming that a higher stage is more
moral than a lower stage.(pp. 180-181)

These assumptions of our psychological theory correspond
to parallel metaethical assumptions of a moral theory,
that is, to assumptions as to the basic nature and validity
of moral judgment. (p. 184)

Our claims for superiority for higher stages are not claims
for a system of grading the moral worth of individual persons,
but are claims for the greater adequacy of one form of moral
thinking over another. (p. 214)

Three primary modes of moral judgment, and the corresponding
types of ethical theory, deal with (a) duties and rights
(deontological), (b) ultimate aims or ends teleologicall),
and (c) personal worth and virtue (theory of approbation).
Our claims for the superiority, then, are claims for the
superiority of stage 6 judgments of duties and rights (or
of justice) over other systems of judgments of duties and
rights. (p. 214)

Kohlberg thus admits that his model of moral development is

prescriptive, that it ranks one kind of morality over other kinds of

morality. However, he feels that he is justified in his claims since

they are based on empirical evidence attesting to the invariance of

his stages and their trans-cultural universality. However, detailed

reviews of research investigating these assumptions (e.g., Graham,

1972; Kurtines & Grief, 1974; Simpson, 1974) agree that no conclusive

decisions can be reached on the basis of the present findings. Secondly,

his claim that "metaethical assumptions must be compatible with, if not

derived from, acceptable psychological theorizing on moral judgment"

(1971, p. 184) would be rejected by many ethical philosophers (cf. Alston,

1971; Peters, 1971). It does not logically follow that simply because

a certain kind of action is judged ethical such an action should be

judged ethical. As R. S. Peters (1971, p. 262) notes, this assumption

"would involve us in the most feeble form of the naturalistic fallacy"

since "nothing about what ought to be done follows from the empirical

fact we use a word in a certain way." Lastly, although Kohlberg (1971,

p. 217) argues that his deontological Stage 6 morality is a more moral

mode of judgment by "formalist metaethical criteria," consensus has

never been reached among ethical philosophers concerning what these

metaethical criteria are, let alone that Kohlberg's approach meets the

criteria better than other approaches to ethics.

Measurement issues. Many of the methodological problems encountered

by Piaget and Kohlberg stem from their use of an open-ended, semi-

structured interview technique for generating their basic stages of moral

development. Although claims are made for invariance and universality of

sequence, it is possible that the stages are largely determined by the

idiosyncrasies of the population used in the initial interviews. For

example, because Piaget focused on children's moral judgments, he

suggests that moral maturity is reached by the age of 10 when the child

moves into the autonomous stage. However, Kohlberg interviewed respondents

over a greater age range and argues that many adults never become com-

pletely morally mature. Thus, while Kohlberg's classificatory scheme

is more meaningful when measuring adults' judgmental processes, both

approaches suffer from generalizability problems. From Kohlberg (1959,

p. 89), "The number of types we came out with was eventually rather

arbitrary, and undoubtedly determined by the limits of variation of our

particular population."

A second set of difficulties revolves around problems of measuring

stage attainment. Only Kohlberg has attempted to systematically develop

a psychometrically acceptable technique in his Moral Judgment Scale

(MJS). The scale, which follows an interview format, involves presenting

a series of short stories describing moral dilemmas to respondents. The

dilemmas are "ones in which acts of obedience to legal-social rules or

to commands of authority conflict with the human needs of welfare of

other individuals" (Kohlberg, 1964, p. 12). Interviewees are asked to

evaluate the actions of the central person in the narrative, and to give

the reasons for their particular evaluation. An example of one such

dilemma (Kohlberg, 1964, pp. 18-19) follows:

In Europe, a woman was near death from a special kind
of cancer. There was one drug that the doctors thought might
save her. It was a form of radium that a druggist in the same
town had recently discovered. The drug was expensive to make,
but the druggist was charging ten times what the drug cost him
to make. He paid $200 for the radium and charged $2000 for a
small dose of the drug. The sick woman's husband, Heinz, went
to everyone he knew to borrow the money, but he could only get
together about $1000, which was half of what it cost. He told
the druggist that his wife was dying and asked him to sell it
cheaper or let him pay later. But the druggist said: "No, I
discovered the drug and I'm going to make money from it." So
Heinz got desperate and broke into the man's store to steal the
drug for his wife. Should the husband have done that?

As originally designed (Kohlberg, 1959), an interviewer was to classify

the 50 to 150 moral expostulations of the interviewee as representative

of one of the 180 cells (30 dimensions of morality by six possible

stages) of the classification system. The results could then be

summarized in terms of a "global" score of moral development or by a

"profile" analysis which reports the number of responses at each stage.

As a semi-structured interview technique, the MJS is open to

criticisms of interviewer biases and effects, verbal ability confounds,

respondent biases such as test-taking sets, and potential subjectivity

in scoring problems. Although few studies have directly examined the

test-retest reliability of the MJS, several indirect measures suggest

that it is low (e.g., Turiel, 1966; Rest et al., 1974), and Kurtines

and Grief (1974) complain that the internal consistency of the stories

has never been examined. Also, while those studies that have reported

interviewer reliabilities suggest that it reaches acceptable levels

(.78 to .85), the constant revision of scoring procedures by Kohlberg

means that studies separated by several years may not be using comparable

scoring procedures.

Several different steps have been taken to circumvent these

problems. One method involves asking subjects to give written responses

to Kohlberg's dilemmas without suggestions from an interviewer (e.g.,

Froming & Cooper, in press; Porter & Taylor, Note 1; Wilmoth & McFarland,

Note 2). These responses can then be classified by several raters and

the inter-rater reliability estimates obtained.

Alternatively, several researchers have attempted to develop

objective measures of Kohlberg's six stages. Rest and his colleagues

(Rest, 1973; Rest et al., 1969; Rest et al., 1974) have subjects indicate

a preference for (Rest et al., 1969) or the importance of (Rest, 1973;

Rest et al., 1974) statements representative of Kohlberg's stages.

Rest's Defining Issues Test (DIT) asks respondents to read six moral

dilemmas, each of which is followed by 12 statements relevant to the

situation. These statements were written so that they (a) contain the

distinguishing features of Stages 2, 3, 4, 5, and 6, (b) are short and

to the main point, and (c) are matched as to syntax, length, and

vocabulary complexity. For example, after the Heinz and the drug

dilemma, items recommended for consideration included "Whether or not

a community's laws are going to be upheld," "Isn't it only natural for

a loving husband to care so much for his wife that he'd steal?" or "What

values are going to be the basis for governing human interactions?"

(pp. 493-494). Subjects are requested to both rank each item as to its

importance as well as select and rank the four most important statements

"in deciding what ought to be done" (p. 494).

Rest's DIT successfully deals with certain psychometric and

methodological problems inherent in Kohlberg's measurement technique

and has been supported by some validation (Rest, Note 3). Rest recommends

the use of the P score measure of the DIT, which is the sum of a subject's

responses to those items indicative of principled morality (Stages 5 & 6).

Since the many alternative methods of examining DIT responses apparently

yield similar results (cf. Cooper, 1972; Rest et al., 1974), the P

score is considered the central index of responses to the DIT; high P

scores indicate a strong reliance on principles in making moral judgments,

and low scores suggesting the adoption of lower level, conventional

moral thought. In the initial assessment of the adequacy of the measure,

Rest et al. (1974) did find developmental trends such that levels 2, 3,

and 4 decrease with age while P scores increase, a significant corre-

lation of + .68 between P and Kohlberg's moral judgment scale (global

scoring method), and significant relationships between P and several

value content scales (Law and Order scale, negative correlation;

Libertarian Democracy Scale, positive correlation; Comprehension of

social-moral concepts scale, positive correlation).

Unfortunately, Rests' work is open to several criticisms. Rest's

model of morality, like that of Kohlberg's, relies on the comparison and

evaluation of responses dependent upon a culture-specific content. For

example, graduate students tested included seminarians and doctoral

students in political science and moral philosophy. Rest et al. (1974,

p. 495) note that "the latter group is probably the best estimate of

expert opinion," without clarifying what moral philosophy students are

expert at. Also, Rest fails to adequately interpret the theoretical

meaning of the P-score. Since the stages of Kohlberg's sequence are

qualitatively distinct and discontinuous, the interpretation of the

"moral maturity" index is ambiguous since it lumps together theoretically

different types of responses. Lastly, although Rest et al. report

substantial correlations between the DIT and the MJS, Froming (personal

communication) has found that this correlation is greatly reduced when

age of respondent is controlled.

Psychological Processes and Philosophical Theory

The philosophical study of ethics addresses the issues which underly

moral decision making. Moral philosophy recognizes that individuals do,

should, and must make moral judgments of the type "Person X or action Y

is not good." The basic question of "Why is X or Y seen as 'bad' rather

than 'good'?" necessarily involves answering further questions such as

"What are the meanings of ethical terms like 'good,' 'bad,' 'right,'

and 'wrong'?;" "Do good actions share some common characteristics which

determine the application of this label -- a set of characteristics

which do not apply to bad actions?"; and "What are the reasons which

support the assumption that the possession of these characteristics

makes a thing 'good' rather than 'bad'?"

The relevance of moral philosophy to a psychological analysis of

moral judgment must be clearly delineated in order to avoid the problem

of confusing the two disciplines. A great deal of moral philosophy is

concerned with making prescriptive statements concerning how actions

and individuals should be morally evaluated, and thus how people should

make moral judgments. A psychological analysis, on the other hand,

performs only a descriptive function by proposing and testing a

theoretical formulation of how individuals do make moral judgments.

Thus, philosophy addresses the question of what actions are moral and

which immoral, while a scientific theory is concerned with how individuals

make psychological judgments about morality.

However, stating that science is descriptive while moral philosophy

is prescriptive is to oversimplify. Science and ethics may be integrally

related when decisions concerning what actions should be taken are made.

For example, a political policy may be adopted based on the moral

principle that human rights are sacred and should not be violated by

any government. When this principle is accepted as the basis for action,

information regarding how inequities may best be reduced is required.

Scientific procedures then become useful in determining if the means

proposed as solutions to the problem are viable, what the short and long

term implications of implementing certain programs will be, and the

psychological, political, sociological, and economic reactions which may

accompany the implementation of the programs (Schlenker & Forsyth, 1977).

In addition, examination of the distinctions which ethical

philosophers consider important may provide insight into possible

individual differences in the moral judgment processes of non-philosophers.

For example, the disagreement between philosophers about whether or not

benevolent motives justify actions which produce negative consequences

may be evident at the psychological level as well; certain judges may

feel that motives are more important than consequences, while others may

feel that consequences are the primary data for moral judgments.

The Ethics of Aesthetics and Utility

The possibility that philosophical issues parallel psychological

processes was explored by Sharp (1898) in one of the earliest studies of

individual differences in moral judgment. Puzzled by the marked lack of

consensus that typifies philosophical treatments of moral deliberation,

he suggested that "the great majority of moralists have been and are

either hopelessly incompetent or careless or both, or that there exist

different types of moral judgment" (p. 198). Sharp distributed to

people (who had not done "any study of theoretical ethics," p. 210) ten

short moral dilemmas. All the vignettes contained information about the

circumstancessurrounding an action which produced a morally valuable

outcome, and ranged from choice of a career to stealing to allowing

another person to die. For example, one situation (Situation V, p.

202) was:

The following might have happened at the Johnstown
(Penn.) flood. A man found he had just time either to warn
his wife or two other women (not relatives). Both these
women have family ties, etc. so that looked upon purely
from an objective standpoint the death of anyone will
involve as great a loss to all concerned as the death of

After describing the central character's decision, Sharp asked subject

to make moral judgments about the depicted actions and explain the

reasoning underlying these judgments. Although the judgments made were,

on the average, evenly split between praise and blame, Sharp concludes

that the reasons subjects gave for their responses revealed three types

of moral ideologies. The first type, referred to as the aesthetic judges,

identifies moral worth with the "beauty of character" or motive revealed

by action. The second type suggests a utilitarian emphasis on conse-

quences. Members of the final group, who "were not the least numerous,

alternated in the use of each, employing now one, now another, or both

concurrently" (p. 222). Some examples of responses to Situation V


He should save his wife -- he had promised sacredly to
protect her. To the other women he did not owe this moral
obligation. (p. 231)

If he knew hundreds of people were to be swept away and
drowned, without time enough to warn both of the parties, I
should say it was his duty to save his wife. (p. 208)

To warn the two women. Same principle as [I described]
above, the greatest good to the greatest number, and least sum
of suffering. Warning his wife would bring more satisfaction
to himself, and less to others, but it would be selfish. He
is supposed to look upon all as on a plane of common humanity,
not as separated by artificial ties. (p. 232)

All other things being equal, the number of lives he was
able to save determines what is the proper impulse to follow,
and not his personal relation to the persons in danger. But
I would not condemn a man for saving his wife in preference to
two other women, and I doubt if anyone else would condemn him.
The numbers in this case are too nearly equal, though of course
the fundamental principle is the same. (p. 208)

Question of duty to your wife or to two persons not at
all connected with you; of the death of one or two persons.
I presume, looking at it from the good to the world and ignoring
self, one should save two lives rather than one, but the other
being his wife I should say save the wife. A man owes his first
duty to his family, after that to the world. (p. 233)

Sharp concludes that "some value conduct primarily for what it brings,

others for what it reveals," and that both approaches may be "taken

towards the facts of the world of matter and consciousness, and we may

accordingly ask whether they appear as determining factors in our judg-

ments of right and wrong" (p. 222).

The Ethics of Idealism and Rule-Universalism

In a more recent study, Schlenker and Forsyth (1977) had subjects

complete an "Ethical Position Questionnaire" which was comprised of items

tapping the major concerns of adherents to various ethical philosophies;

the specific focus of the items was the topic of psychological research.

In order to empirically group the items from the questionnaire into

coherent categories, subjects' responses were factor analyzed, revealing

two strong bi-polar factors. The first, which Schlenker and Forsyth

labelled idealism, included items relevant to the benefits and costs of

research. Subjects whose factor scores classified them as idealists

insisted that no harm, however small, was permissible in research, that

people's welfare was crucial, and that it was of primary importance that

a project might advance science. At the opposite end of the continuum

were the pragmatists, who felt that some degree of harm was permissible

and that it was not of primary importance for a scientific advance to

follow from all research; a balancing of benefits and costs seemed

paramount for these persons.

The second factor was labeled rule-universalism since it contained

items pertinent to the degree subjects felt that universal, relatively

rigid ethical codes could be developed. Some of the items with which

subjects could agree or disagree included "The application of a particular

code of ethics depends entirely upon the particular study; what is

appropriate in one study might be totally inappropriate in another,"

"Any code of research ethics should serve only as a series of flexible

rules and should not be expected to apply in any and all situations," and

"What is ethical varies from one situation and society to another." High

rule-universalism appears indicative of an ethical ideology which espouses

the importance of principles in making moral judgments, while low rule-

universalism is indicative of a relativistic ethical ideology.

When dichotamized, the idealism factor yields the two groups

idealism and pragmatism, while absolutism and relativism correspond to

the end-points of the rule-universalism continuum. By crossing the two

dimensions, Schlenker and Forsyth were able to generate the four group,

two-way classification of ethical ideologies pictured in Figure 2-1.

The four ethical types which comprise the taxonomy include idealistic-

absolutism, idealistic-relativism, pragmatic-absolutism, and pragmatic-

relativism. As Schlenker and Forsyth point out, each of these ideologies

is related to schools of thought in the philosophical study of ethics.

The statements which subjects labeled idealistic-absolutists agreed with

are consistent with a general approach to moral philosophy known as

deontology. This ethical philosophy rejects a rule or action's conse-

quences as a basis for moral evaluation and appeals to natural law and/or

rationality to determine ethical judgments. In a deontological ethical

philosophy, acts are to be judged as moral or immoral through their

comparison with some universal moral rule to which no exceptions can

be made. The philosopher Immanuel Kant, generally regarded as the










Figure 2-1

The Taxonomy of Ethical Ideologies

wL ,

foremost proponent of the deontological school, prescribed that one must

"Act only on that maxim which you can at the same time will to be a

universal law" (cited in Frankena, 1973, p. 30). For Kant, this general

universal principle or "categorical imperative would be that which repre-

sented an action as necessary of itself without reference to another end"

(Kant, 1873/1973, p. 75). Applied to interpersonal conduct, Kant (p. 82)

deduces that one must "act to treat humanity, whether in thine own person

or in that of any other, in every case as an end withal, never as a means


Deontologically, a moral principle can allow no exceptions, regard-

less of the consequences. For example, suppose that a physician finds

himself tempted to lie to a terminally ill patient about his chances for

recovery. A deontological outlook would judge the lie, despite its

benevolent goals, as immoral. While the lie may have ennumerable

positive consequences for relatives, the doctor, and the patient, it

"is an infringement on the rights of humanity" (Kant, 1962, p. 264). For

the dontologist, "A lie is a lie, and is in itself intrinsically base

whether it be told with good or bad intent" (Kant, 1962, p. 265).

Although no ethical ideology adopted by an individual may possess all

the characteristics of a purely deontological approach, the idealistic-

absolutists emphasis on maintaining consistency with moral principles

in order to obtain desired goals is similar to a deontological philosophy.

The statements endorsed by pragmatic-absolutists, on the other hand,

are more compatible with a teleological ethical philosophy. The

teleological approach proposes that the ultimate judgment of the morality

of an action or set of actions depends upon the consequences produced by

it. One is ethically bound to act in a way that produces "good" conse-

quences, with good being variously defined in terms of pleasure, happiness,

self-realization, fulfillment, and/or demand. The position has a long

history and Bentham's dictum, that action should produce the greatest

good for the greatest number, epitomizes the approach.

Although the utilitarian formulation is relatively simple when

all the consequences are good or all bad, it becomes more complicated

when good and bad consequences are produced by a single action. When

applied to the example of a doctor deciding whether to lie to an ill

patient, the teleological philosopher advises that the potential benefits

of the lie must be weighed against the potential costs of the lie. The

doctor must calculate the benefit/cost ratio, taking into account

positive factors (such as reducing the anxiety of the patient and family,

aiding the patient's recovery by lessening his worry, and sparing himself

the discomfort of telling someone they are going to die) and negative

factors (such as arousing distrust in doctors if the lie is found out,

actually reducing the patient's chance of recovery by hiding the serious-

ness of his condition from him, and complicating legal procedures such

as preparation of the will). Thus, while the judgmental processes and

behaviors of individuals classified as pragmatic-absolutists may not be

consistent with a teleological philosophy, the items endorsed by this

group did parallel a teleological approach to ethics. Like utilitarian

philosophers, these "naive teleologists" suggest that absolute moral

principles are important, but also agree that one must apply these rules


Lastly, all relativists endorse an ideology which is related to

skepticism, although they may be either pragmatic or idealistic in their

orientation. At the philosophical level, a skeptical ethical ideology

recognizes that there are many moral points of view, and all the varieties

of skepticism seek in one way or another to criticize those who attempt

to present specific ethical principles. Ethical egoism is an example of

a skeptical ethical philosophy that takes a pragmatic approach to

evaluating action. The ethical egoist argues that no moral standards

are valid except in reference to one's own behavior. The only moral

pronouncement that can be made is that one should act according to what

one feels is right, and not act in a way that one feels is wrong. When

universally applied, the pronouncement becomes "Everyone should always

act so as to promote his own interest," a position adopted by Hobbes as

an enlightened egoist (Davis, 1973). At one level, ethical egoism is

similar to teleology since egoists typically allow consequences to serve

as the basis for determining right and wrong. Egoists, however, avoid

the problem of transpersonal agreement in judgments since only their

own value judgments are crucial. Also, egoists stand in direct opposition

to utilitarians for they consider behaviors in light of the greatest

goodness to themselves rather than to others.

Fletcher's situation ethics (1966) provides an example of an

idealistic skepticism. While endorsing neither strict principledness nor

anarchy, Fletcher argues that morality should attempt to focus on "a

contextual appropriateness -- not the "good" or the "right" but the

fitting" (1973, p. 186). Fletcher idealistically suggests that this

will be possible if one's actions are based on agap6, or love of others.

Like these skeptical approaches to moral philosophy, the relativistic

ideology reveals a distrust in absolute moral principles, and argues

instead that each situation must be examined individually.

Although the comparison of each ethical ideology with a philo-

sophical counterpart does much to clarify the nature of each position,

the distinction between a philosophical analysis of a moral issue and a

psychological analysis of an individual's ethical ideology should not

be overlooked. Kant, as a deontologist, argued that actions are judged

by assessing their compatibility with moral standards. The idealistic-

absolutist, as a psychological "deontologist," also argues that moral

standards are important determinants of moral judgments. However, the

link between the "true" deontology of Kant and a psychological

"deontologist," is not perfect. For example, idealistic absolutists,

while they argue that principles and not consequences provide the basis

of morality, may never-the-less base their moral judgments on the conse-

quences of an action as well as on its normativeness. Thus, while it is

heuristic to note the parallels between deontology and idealistic-

absolutism, teleology and pragmatic-absolutism, and relativism and

skepticism, the comparison does not suggest that the psychological

ideologies do not differ from the corresponding moral philosophies.

Ethical ideology and moral judgment. In addition to empirically

distinguishing individuals who endorsed different ethical ideologies,

Schlenker and Forsyth's results also provided some indication of the

relationship between ethical stance and moral judgment. Subjects in the

study read a description of a psychology experiment in which deception

was employed, and rated the described study as to scientific value,

amount learned, degree of harm produced, and ethicality. The results

indicated that for pragmatists who endorse absolute principles

teleologistss), estimation of ethicality was positively correlated with

perceptions of benefits such as amount of information learned and the

experimenter's concern for science, but uncorrelated with cost consid-

erations such as amount of harm done subjects, predictions of possible

harm, and the foreseeability of harm. For the idealistic-absolutists

("deontologists"), moral judgments were negatively correlated with

judgments related to the harm done subjects, while positive considerations

were uncorrelated with judgments of morality. Thus, subjects' moral

judgments did not perfectly reflect the philosophical ideology they

espoused. One possible source of the differences in perspective between

the two ideologies may lie in how idealistic their expectations are. As

regards psychological research, a subject classified as an idealistic-

absolutist, although interested in the advancement of science, also

believes that the advancement may come about without any negative

ramifications such as harming participants in research projects.

Therefore, irrespective of the positive attributes of the research,

he or she may condemn the action since it falls below his or her

idealistic standards. The pragmatic-absolutist, on the other hand,

pragmatically assumes that some negative consequences will unavoidably

result from any psychological research project, so that the essence of

ethicality is perceived to lie in the positive consequences which offset

the ubiquitous but insignificant negative consequences. Interestingly,

the group whose moral judgments most completely covaried with both

benefits and costs were the idealists who believed in non-universal, fluid

rules (i.e., the relativists). Thus, Schlenker and Forsyth, using their

Ethical Positions Questionnaire as a measure of moral ideology, found

support for (a) the existence of variation in individuals' ethical

ideologies and (b) that these variations influence moral evaluations.

Applications and Limitations of the Ethical Ideology Approach

The problem of cultural bias and value judgment encountered in

dealing with Kohlberg's theorizing is avoided by Schlenker and Forsyth.

Given Kohlberg's emphasis on the importance of principled morality, his

typology must necessarily downgrade the reasoning underlying the

maturity of individuals who do not rely on deontological principles in

their moral judgments. A typology based on the Ethics Positions

Questionnaire measurement of idealism/rule-universalism does not "rank"

any morality as more mature than another, and recognizes individuals who

are skeptical about the existence of universal laws. Given the value-

ladenness of moral judgment, extra care seems to be required when attempt-

ing to examine its processes scientifically. The distinction between

philosophy and science is an important one since it determines how

questions are to be answered (i.e., either empirically or through

philosophically acceptable means), and the Schlenker and Forsyth

classification does not meld the two. In addition, the approach is

consistent with other recent research which suggests that certain

individuals may be "naive" philosophical relativists. For example,

Hogan (1970, 1973) argues that certain individuals' moral judgments are

less closely related to the compatibility of an action with moral

standards. He argues for the existence of persons who follow an

"ethics of personal conscience" who tend to be "innovative and form

creating" but may also be opportunistic and impulsive. Hogan contrasts

these individuals with those who follow the "ethics of responsibility."

These individuals are characterized by conventionality in their moral

judgments, tending to be reasonable, dependable, and emphasizing currently

accepted laws and political institutions. Hogan's (1973, p. 225) Survey

of Ethical Attitudes (SEA) measures these two philosophies of "responsi-

bility" and "personal conscience" by asking individuals if they agree

or disagree with such statements as "(a) All civil laws should be judged

against a higher moral law; (b) right and wrong can be meaningfully

defined only by the law; (c) an unjust law should not be obeyed; (d)

without law the life of man would be nasty, brutish, and short."

The idealism/rule-universalism typology is not without its limita-

tions, however. The items used to construct the two subscales of

idealism and rule universalism were selected empirically via factor

analysis. Thus, the questionnaire contained other items in addition to

those later selected for the scales, and these items could have influenced

subjects' scores. In addition, all the questions used by Schlenker and

Forsyth for the classification of individuals were in reference to

psychological experimentation. Although there is no inherent problem with

this emphasis since Schlenker and Forsyth were primarily interested in

individuals' judgments of the psychological experiment, their items

cannot necessarily be applied to judgments of other situations and the

categories suggested may be relevant only when people judge psychological


A second problem revolves around the issue of the characteristics

of the individuals who endorse different ethical ideologies. While the

consistency between the four ethical ideologies reported and the approaches

to ethics actually taken by philosophers supports the generality of the

typology, further clarification of the differences between judges who

endorse the various ideologies is needed to determine the impact of


ideology on moral attribution. For example, the manner in which individuals

in different groups differ in their actual moral judgments has not been

examined. The finding that idealistic-absolutists emphasized negative

consequences more than pragmatic-absolutists should mean that the latter

group is more positive in their moral judgment of an action that produces

mixed consequences. However, an in-depth analysis of the characteristics

of members of the ideologies suggested by Schlenker and Forsyth has not

yet been conducted.



While attribution theory hypothesizes that both responsibility

ascriptions and action evaluation are determinants of moral judgments,

research suggests that individual differences in ethical ideology may

systematically influence such attributions as well. Adopting the

classificatory framework explored by Schlenker and Forsyth (1977), the

attributor's idealism and belief in the universality of moral rules

should impact on attribution of morality. To provide an example, suppose

that two individuals witnessing an action that produces positive and

negative consequences both apply universal rules in making their moral

judgments; however, the first attributor is idealistic, while the second

is pragmatic. The second individual, since he or she emphasizes the

positive consequences of the action more than the negative, may evaluate

the act itself more positively. The idealistic individual's appraisal,

on the other hand, would be less positive because of the greater salience

of the negative consequences. Further, the idealistic individual may

also be more likely to hold the actor responsible for the negative conse-

quences since he or she assumes that bad outcomes are ultimately avoid-


This chapter describes the hypotheses and methods which were used

to investigate the impact of ethical ideology on the attribution of

morality. Because the original Ethics Position Questionnaire used by

Schlenker and Forsyth was specific to judgments of psychological

experiments, a revised version of the EPQ was developed which was

applicable to all interpersonal encounters which involve moral decisions.

After a description of the steps which were taken in this scale develop-

ment process, the hypotheses which guided the research are presented.

The chapter concludes with a description of the methods which were

followed in the three validational studies which investigated how an

actor's degree of responsibility, the quality of the consequences, and

the consistency of the action with norms of morality influence moral


Scale Development

The Ethics Position Questionnaire (EPQI) developed by Schlenker

and Forsyth (1977) was originally comprised of 68 items which dealt with

various issues relevant to the morality of psychological research. The

items were selected as indicants of the major issues which characterize

discussions of moral judgments and "ran the gamut from ones concerned

with the feasibility of universal ethical codes to ones concerned with

deception, harm to research participants, and the ability of science to

solve the world's problems" (p. 382). The EPQI was administered to 180

male and female college students, who responded by noting degree of

agreement or importance on seven and 9-point scales. From this total

group of statements the 27 items contained in Appendix A were selected

via a principle axes factor analysis with varimax rotations as indicative

of the two dimentions of ethical ideology; 13 for the rule universalism

factor (absolutism-relativism) and 14 for the idealism factor (idealism-


A revision of the scale was carried out in order to produce a

measurement device that would possess the following characteristics:

(1) applicability to all interpersonal encounters rather than just

psychology experiments, (2) high inter-item consistency on each subscale

but broad representativeness of the desired dimensions, (3) stability over

time, (4) orthogonality between the two subscales, and (5) little or no

relationship between the two scales and social desirability indicators.

As with the EPQI, the scale would enable the classification of individuals

according to ethical ideology, including idealistic-absolutists

("deontologists"), pragmatic-absolutists teleologiststs"), idealistic-

relativists ("idealistic skeptics") and pragmatic-relativists ("pragmatic


Step One: Initial Factor and Item Analysis

The first step taken in revising the EPQI involved generating a

large number of items which were subjectively judged to be indicants of

the two dimensions of ethical ideology. These 120 items included some

used in the EPQI but reworded to be less content specific, and new items

which seemed to reflect either universalism or idealism. On the basis

of face validity and relevance to the two dimensions, 55 items were

selected for further use. The 26 questions concerned with idealism and

29 items related to rule-universalism are contained in Appendix A.

As part of the initial scaling procedure 65 students from the

Introductory Psychology subject pool were asked to indicate their degree

of agreement with each item on 9-point Likert-type scales. A total

score for both idealism and rule-universalism was then computed, and

correlated with each individual item. These correlations (see Appendix

A) were used to provide a basis for a scale construction process

designed to create two subscales with adequate length and internal

consistency. An item was incorporated into the scale if the correlation

between it and the relevant subscale was high and the correlation between

the item and the second subscale was low.

Although high internal consistency was desired, it was also

important that the items retained in the final EPQ tapped all the

relevant domains of content. Therefore, factor analysis was used in

conjunction with item analysis to insure that the scales did not yield

a restricted range of applicability. Using principle components factor-

ing and orthogonal varimax rotation, 16 factors were extracted which

had eigenvalues greater than or equal to 1.0. These factors accounted

for 77.0% of the total variance and are described in Table 3-1. Using

this factor analysis, groups of items were selected for scaling which

sampled from as many of the different factors as possible to insure

generality of the final scale.

The items retained after the preliminary scaling can be seen in

Appendix A. Based on both the item and factor analysis, 14 items were

selected for the idealism scale and 13 for the rule-universalism scale.

An item was retained provided (a) the correlation between it and the

relevant scale was high, (b) the correlation between it and the second

subscale was low, and (c) the item was representative of one of the

domains of content suggested by the factor analysis. Scaling also

uncovered the importance of a third group of items. These items all

measured the degree to which the individual was willing to make moral


-3- (C
S.Ln A-

--T 04

UA ir 0o -T -r -3r r-

r\ "o N CM -> (
C'.1 (< E -T" -T" C

>n mo --n C(
-3" Ln -.T D 0N -T

- m -: -

0) a) C)
4 4- 4 4-1





S 4-J



In 4- )
c -


4-1 0)

- U 4-1


< <



,0o -r -T


o *-:- -~ en
r'- r- c > -r

mn co c0 o
-T CeN (N




-T r D r. rC 3

-. r-. .0 C .0 Cr
--T E E % 0

-T ft\ CM -T CM

w) a) a) 0 0) ) 0)
441 4> 4 4- 41 44 44
-ooo u

(N In a0 0 C)

r-~. '-0 In -3 -3 3

m ( o-T-

3 > X > Cn
0 0)
LLo L- t LJ

rr\( m CM

ln -

d) >

(a -


> 0



U 0)
(a C
LL 0





L. -

r- LA -

_r \.. -. "% T
...7 -- .-

,- O

u 1


N O OO 0- e'a
-.) -, -.0 L t -" ---

S cN L- -
E E E E E c

S 0 0 0 -

-1U C
tD -u

S -- -- 1


u --

L O F- U I OO -.(U.

N E C c 7
c .*
V -t a. q

Ov O >

S41 .- l. l_ 4 44

judgments of others, and contained the items "Others actions should not

be judged," "Although I use my moral principles to judge my own actions,

I don't use my principles in making judgments of others," and "I would

not condemn another who performs an action which violates one of my

moral rules if the doer thinks that he is acting morally." Because

these three items were highly intercorrelated but did not seem to be

theoretically or empirically related to either idealism or rule-univer-

salism, they were used to construct a separate scale which measured

willingness to generalize one's moral ideology to others.

Step Two: Final Revision of the EPQ

These two preliminary items were then completed by 56 students from

the Introductory Psychology subject pool, along with Edward's (1957)

SDS -- the Social Desirability Scale. First, additional item analysis

was conducted to delete any items which did not correlate significantly

(p < .01) with the totals for each scale. This analysis resulted in the

retention of 10 items for the idealism scale, 10 for the rule-universal-

ism scale, and 3 for the generality of judgments scale. These 23 items,

along with the directions for the EPQ, are presented in Table 3-2.

Next, the inter-scale correlations were computed to assess the orthogonality

of the measures. The correlation between rule-universalism and idealism

was + .05 (ns), and the correlations between these two scales and the

generality scale were + .37 (p < .05) and + .23 (ns), respectively.

Lastly, the correlation between each scale and the SDS was calculated.

These correlations, for rule-universalism, idealism, and generality,

were + .22, + .18, and + .26 (ns), respectively. These analyses suggest

that, as desired, the two scalesof universalism and idealism were not

Table 3-2

The Ethics PositionQuestionnaire (EPQ)

You will find a series of general statements listed below. Each represents
a commonly held opinion and there are no right or wrong answers. You will
probably disagree with some items and agree with others. We are interested
in the extent to which you agree or disagree with such matters of opinion.

Please read each statement carefully. Then, indicate the extent to which
you agree or disagree by placing on the line in front of the statement
the number corresponding to your feelings where:

1 = Completely Disagree 4 = Slightly Disagree 7 = Moderately Agree
2 = Largely Disagree 5 = Neither Agree Nor Disagree 8 = Largely Agree
3 = Moderately Disagree 6 = Slightly Agree 9 = Completely Agree

First impressions are usually best in such matters. Read each statement
carefully, decide if you agree or disagree and the strength of your opinion,
and then enter the appropriate number on the answer sheet. PLEASE give your
opinion on every statement.

1. A person should make certain that their actions never
intentionally harm another, even to a small degree.

2. Risks to another should never be tolerated, irrespective
of how small the risks might be.

3. The existence of potential harm to others is always wrong,
irrespective of the benefits to be gained.

4. It is sometimes necessary to sacrifice the welfare of others
in certain interpersonal situations.

5. One should never psychologically or physically harm another

6. One should not perform an action which might in any way
threaten the dignity and welfare of another individual.

7. If an action could harm an innocent other, then it should
not be done.

8. The compatibility of the action with ideas of what is the
"best" or most "perfect" kind of behavior must be taken
into account when judging actions.

9. Balancing the positive consequences against the negative
consequences of an action can result in a person performing
an action which will harm others to get positive consequences,
and hence is immoral.


Table 3-2 (continued)

10. The dignity and welfare of people should be the most
important concern in any society.

11. Other's actions should not be judged.

12. Although I use my moral principles to judge my own actions,
I don't use my principles in making judgments of others.

13. I would not condemn another who performs an action which
violates one of my moral rules if the doer of the act thinks
that is acting morally.

14. What is ethical varies from one situation and society to

15. Moral standards should be seen as being individualistic;
what one person considers to be moral may be judged to be
immoral by another person.

16. Different types of moralities cannot be compared as to
"rightness"; what one person considers to be moral may be
judged to be immoral by another person.

17. Questions of what is a moral behavior can never be resolved
since what is moral or immoral is up to the individual.

18. Moral standards are simply personal rules which indicate how
a person should behave, and are not to be applied in making
judgments of others.

19. Ethical considerations in interpersonal relations are so
complex that individuals should be allowed to formulate
their own individual codes.

20. Rigidly codifying an ethical position that prevents certain
types of actions could stand in the way of better human
relations and adjustment.

21. No rule concerning lying can be formulated; whether or not
a lie is permissible or not permissible totally depends upon
the situation.

22. There are some ethical principles that are so important that
they should be a part of any code of ethics.

23. Whether or not a lie is judged to be moral or immoral depends
upon the circumstances surrounding the action.

Note: The idealism score is computed by averaging responses to 1-10
(Item 4 is reverse scored).


Table 3-2 (continued)

The rule-universalism score is computed by averaging responses
to 14-23 (Item 22 is reverse scored).

The generality score is computed by averaging responses to

significantly correlated with one another. Also, while there was a

small positive correlation between scales and SDS, it did not reach


Step Three: Scale Characteristics

The EPQ, in its final form, was administered to 337 psychology

students in their classrooms in order to (1) obtain information about the

scale item (e.g., means, standard deviations, correlations with total

scores) and (2) identify a set of respondents who would participate in

the validational experiments. The 64 males and 64 females who were

selected on the basis of this first administration completed the EPQ

a second time -- approximately two weeks later in individual testing

booths -- allowing for the calculation of test-retest reliabilities

and comparison of internal consistency based on the first administration

(in the classroom, n = 337) and the second administration (individual

sessions, n = 128).

The characteristics of the Ethics Position Questionnaire are

summarized in Tables 3-3 to 3-7. Table 3-3 contains the inter-item

correlations, item-to-total correlations (corrected for the item itselff,

item means, item standard deviations, and item test-retest reliabilities

for the idealism scale. Tables 3-4 and 3-5 contain similar information

for the remaining scales of rule-universalism and generality. Table 3-6

contains the statistics of reliability for both the first and second

administrations of the EPQ. As previous analyses had indicated, the

idealism and rule-universalism scales have high internal consistency

as evidenced by Cronbach's alpha and Spearman-Brown criteria. Internal

consistency was, however, greater for second administration responses in

comparison to the first administration responses. Although several

factors could account for these results, this reduced consistency for

first testing could possibly be due to carelessness on the part of

subjects when they responded to the EPQ in the large group testing

sessions. Alternatively, the change in the reliability of the instrument

may be due to a practice effect that results in an increased coherence

in the responses of subjects who have taken the EPQ previously.

Test-retest reliabilities for the three subscales were computed by

correlating each subject's score for the first administration of the EPQ

with his or her score on the second administration. These correlation

coefficients are presented in Table 3-7, and while all are significant,

they are only moderate in magnitude. Although this less-than-desired

degree of test consistency across time may indicate that the dimensions

being assessed are unstable, several other factors could have limited

the size of the correlations. For example, variations in the adminis-

trative procedures and questionnaire formats used for the two assessments

may have been influential; second, if subjects did respond less carefully

on the first administration as the lowered internal consistency statistics

suggest, this would also reduce the size of the test-retest correlation.

Lastly, since these analyses are based on the responses of subjects with

extreme scores on the idealism and rule-universalism scale, statistical

regression towards the mean could have limited the size of the test-retest


The correlations between the second administration EPQ scales and

the original ethics position measure (the EPQl), the DIT, and the SEA

are also presented in 3-7. Both idealism and rule-universalism correlated






4-1 -
I- C

4-1 M-
- 0

O *



0 -a
O 4-




- E


U (,



E -





E c
0) -D




ro 0 eA
L- i-




0) 0


Q) O0

0) 00





I 4-1

E 0)


2r o


) f)l~

'.0 '-

o -" oo

oo 0 O0
0D O -
- CM M

M 0 C D
r4 C %.

N M -
l 0 m C\

Ln %o .

r C) 0o -
L m \D %

-~ N


0 '.

O .0 rN

-- -

n c rD C

S- -

0 e n f 0

0 -
0 %. -

ft >^ \

~r ~r o

'D0 0

SC) 0 '0 -
NC r N -

-3 -3 re\ -3 00 -
- T N N

C c -3 L \. r-. 0o mn -E

0 ) 0) 0) 0) a) (U 0) a)
4 44 44 44 44 44 41 4-4 44




m N

- N
.o LA






t0 --




0 0


E( 0
d- c
L ,

0 -0)

-I- -

E4 -








44J >-

E -

i -





0- --
4-J 44 1n
(0 0 N

-4 C



) -

4- E


0 E
0 (U
E -






4- -




LC -'.

) a) 0

41 41 44

-T L r-. I

CN -r r- '.0 r- I'
\D Mo ri c -
- N N CN N

'. N i C OCN 00

00 0 MN Ln

ON D 0 00 0 rr
CM L4 \ Ln %- r%0 -T

a- O
0 c" N -

r0 N N

r N -

\- r- L -
LA N N -

LA N N 0C)

r> -3 O N 0n o
- 0 0 0 -

"- 0O

o O

m o N
- N N NM

O) (U U
4 4-1 4- 4

C) -T-

_V C. --T

0N LA I" -_:

'30 N r-. 0
S N r\r C

0 N 0


D 0


L -


Co l0
-; *i;

- W)
L --
L. 4-J
O --



4-J --

1 0)

) -

- -0


o m

L -

4. 1)
I- -

L- -
0 Q)

E -o







E 0


0) -0
- 1LA




o N
0 CM4
ce% CM

tr\ -


C)4 -.a

- .-
I -C
c v
E *-
u -C
ca .
O -




0 c










I O-





'0 1 -r

L1 L'

Lo c o

C D m"



1) >-
> 4-L

C -

- C
a a)
a: 0

LA 00
o N- Lm

o 0 -e


a) >-
> 4.-

C -

a) a)
- C
Ce a)
3 : )









- (U


0- (0
0.. U



O o

ol 1








E c

c) O


C 0

a) U
4- O


en i/
-:: m

significantly with the corresponding subscales included in the original

EPQ, although the match was better for the idealism scale than for

rule-universalism; the correlation between idealism-EPQl and idealism-

EPQ was greater than the correlation between rule-universalism-EPI and

rule-universalism-EPQ (p < .05). In addition, the generality subscale,

which was not included in the EPQI, was found to correlate with the

idealism subscale of the original EPQ, but was not related to the

original rule-universalism subscale.

Rest's Defining Issues Test can be scored to yield an overall

measure of moral maturity as well as a stage classification based on

Kohlberg's theory of moral development (Rest, Note 4). The P-score,

which Rest contends is best for correlational analyses (Rest, Note 4),

summarizes the extent to which a subject's responses to the series of

moral dilemmas display reasoning typical of stages five or six. Thus,

the larger the P-score, the more morally principled or "morally mature"

a subject. As Table 3-7 indicates, neither the idealism nor the rule-

universal scales correlated significantly with P. However, generality

of moral judgments was correlated positively with P, suggesting that

moral maturity may be associated with the recognition that one's

conclusions about morality may be applicable only to oneself.

The DIT was also used to classify subjects into their stages of

moral development. Of the 128 subjects who participated, 39 (30.5%)

could not be classified clearly, 12 (9.4%) were classified as Stage 3,

30 (23.4%) as Stage 4, 46 (33.9%) as Stage 5, and only 1 (.8%) as

Stage 6. Analysis of the joint frequency distribution of the subjects

based on the EPQ ethical stance classification and the DIT stage

classification to determine, for example, if more deontologists were

also classified as Stage 4 or if skeptics were more likely to be

classified at a higher stage of reasoning than teleologists. The

analysis revealed no significant relation between the two measures;

X2 (12) = 14.48, p = .30. It is of interest to note, however, that the

one Stage 6 subject found in the sample was classified a deontologist.

According to Hogan (1973), individuals who receive high scores on

the Survey of Ethical Attitudes (The SEA) adopt an ethics of social

responsibility that emphasizes acceptance of societal regulatory

standards, while low scorers rely on an ethics of personal conscience

which argues that a society's standards can be violated if they conflict

with "higher law." As the correlations in Table 3-7 indicate, idealism

was unrelated to SEA, but both generality and rule-universality were

negatively correlated with the measure. Thus, individuals who express a

disbelief in the possibility of formulating universal standards of

morality and a reluctance to generalize moral judgments to others also

tend to adopt the ethics of personal conscience.

Hypotheses and Methods


Three experiments were conducted examining the classificatory

adequacy of the revised EPQ. All three experiments were similar in that

subjects, who included idealistic-absolutists, idealistic-relativists,

pragmatic-absolutists, and pragmatic-relativists, were asked to make moral

attributions about the actors in particular situations which were described

to them. The three experiments were designed and conducted at the same

time, and so they should not be interpreted as a sequence of studies.

Although greater detail concerning the specific hypotheses under test

will be provided in subsequent chapters, a general description of each

of the studies and its focus follows.

Experiment I, "Consequences of Acts and the Attribution of

Morality," examined how individuals who adopt different ethical ideologies

differ in their emphasis of negative and positive consequences. Subjects

read four scenarios describing a series of events which culminated in

outcomes which were both good and bad. Respondents then made moral

judgments about the actor who produced the mixed consequences, and also

evaluated the quality of the different consequences. Based on Schlenker

and Forsyth's results, it was predicted that idealistic-absolutists

would emphasize the negative consequences more than the positive, that

pragmatic-absolutists would emphasize the positive more than the negative,

and that relativists would balance both positive and negative consequences.

Because of this differential impact of the consequences on moral

attribution, idealistic-absolutists were expected to be more severe

in their moral attributions, pragmatic-absolutists least, and relativists

would be intermediate.

Subjects in Experiment II,"Responsibility, Consequences and Moral

Evaluations," read a series of twenty short scenarios which were developed

by Shaw and his associates (e.g., Shaw, 1968; Shaw & Reitan, 1969;

Shaw & Sulzer, 1964). In half of the stories, the outcome of the

described action was extremely positive, while in the remaining half

the outcome was extremely negative. In addition, elements within each

scenario were varied so that each of Heider's five levels of responsibility

was represented. Thus, the outcome which was described was either good

or bad, and the central character in the story was either associated with

the outcome, caused the outcome, caused the foreseeable outcome unin-

tentionally, intentionally produced the outcome, or justifiably produced

the outcome.

After reading a story subjects were asked to make judgments about

the morality and responsibility of the actor. Based on Shaw's work, it

was predicted that greater amounts of responsibility would be attributed

to the actor up through the level of intentionality, but that attribution

of responsibility would decrease at the level of justification. In

addition, the actor would be held more responsible for a negative outcome

than a positive one. It was further hypothesized that the extent to

which this increased attribution for negative consequences occurs would

be dependent in part upon the ethical ideology of the attributor.

Because an idealistic individual assumes that negative consequences can

be avoided, these individuals should be more likely than pragmatists

to hold the actor who produces negative consequences responsible. How-

ever, while this attributional assymetry should hold for both idealistic-

absolutists and relativists in the attribution of responsibility, when

subjects are attributing morality the severity of consequences effect

should be pronounced only for idealistic-absolutists. Since idealistic-

relativists are less likely to blame a person for violating moral

expectations in any given situation, their attributions of morality

will not be as directly dependent upon their attributions of respon-


Experiment III, "The Consistency of Action with Norms of Morality

and Moral Judgments," was conducted to determine how individuals within

the different ethical ideologies are influenced by the consistency of an

action with moral standards. Subjects who participated in this study

read a series of sentences, each describing an action and outcome. The

action was either consistent or inconsistent with a moral standard (e.g.,

lying vs. telling the truth, breaking a promise vs. keeping a promise),

and the outcomes were extremely good, somewhat good, somewhat bad, or

extremely bad. Because idealistic-absolutists emphasize the importance

of acting consistently with rigid rules of morality, it was expected that

these attributors would be most influenced by the consistency of the

action with a moral standard. Therefore it was hypothesized that

idealistic-absolutists would be more negative in their judgments of an

actor who violated a moral standard than pragmatic-absolutists and

relativists, and that these differences would be most pronounced when

the consequences associated with the violation were negative.


Sixty-four males and 64 females recruited from the Introductory

Psychology subject pool participated in the experiments. The data of one

additional female participant were excluded from the analysis since she

failed to follow the instructions. These 128 subjects were selected from

a larger sample of 337 psychology students who were given the revised

EPQ in their classrooms. Using these responses to the EPQ, the 64

highest and lowest scores on the rule-universalism and idealism subscales

were identified, contacted, and asked to complete additional measures.

When both ethical ideology and sex of subject are considered, the sampling

selected subjects to fill the eight cells of the 2 absolutisticc vs.

relativistic) by 2 (idealistic vs. pragmatic) by 2 (males vs. females)

factorial design.


Subjects were scheduled to complete the experimental questionnaires

and other personality measures individually. Upon arriving at the

laboratory room, subjects were greeted by a male experimenter who seated

them in individual testing booths which minimize distractions while

completing the instruments. Subjects were then asked to read the

instructions to the series of questionnaires and begin. The entire

procedure took an average of 1 1/2 hours to complete, but

times ranged from 1 to 3 1/2 hours.

For all subjects, the questionnaire booklet contained the following

measures: the revised EPQ, the original EPQI from Schlenker and Forsyth,

the Defining Issues Test (the DIT, Rest's measure of moral maturity

described in Chapter Two), the Survey of Ethical Attitudes (the SEA,

developed by Hogan to measure the ethics of conscience vs. legalism and

also described in Chapter Two), and the scenarios and dependent

measures for Experiment I. In addition, subjects' booklets also

contained the materials for Experiment II or Experiment III. Thus,

while 128 subjects participated in Experiment I, only half that number

were given the materials for experiments II and III. The EPQ was the

first measure in all the booklets since as unbiased an estimate of

ethical stance as possible was desired, but the order of the remaining

questionnaires was counterbalanced across subjects. The eight orderings

of booklets employed were randomly determined, and all eight

orders were used in each cell of the design. These orderings are described

in Table 3-8.


4-- 4-1 4
C C C -

E E E a
U.. m- c-, m L. ,
L- L. L-
CU 1) Q)
a a a
0. LQ. L
wj UJU

41 41) 41 -
S- .-- --

a a a





a a -



c0' 0' 0'








c' E



UJ 0'


J3 0)



- CM rm -3-

uL o .0 ~ co



Schlenker and Forsyth (1977) found that individuals weigh the

consequences of an action differently depending upon their ethical

viewpoint. Idealistic-absolutists ("deontologists") tended to look to

the negative consequences, pragmatic-absolutists teleologiststs") to

the positive, and the moral judgments of relativists ("skeptics")

covaried with both types of consequences. Thus, if asked to judge an

action which produced both good and bad outcomes, pragmatic-absolutists'

moral attributions should be more closely related to their evaluations

of the quality of the positive consequences while idealistic-absolutists'

attributions should be more closely related to their evaluations of the

negative consequences. Relativists' attributions, on the other hand,

should be related to evaluations of both the positive and negative conse-

quences produced by the action. Because of this divergence in the

emphasis of consequences, the three positions should also differ in the

severity of their moral attributions. If idealistic-absolutists are

indeed emphasizing negative consequences and pragmatic-absolutists

positive, then the former should be the most negative in their moral

attributions, the latter should be the most positive, and the relativists

should fall intermediate.


Hypotheses based on these suggested differences between the

different ideologies were investigated in Experiment I. The specific

predictions advanced concerning moral attributions about actions which

result in mixed consequences included:

Hypothesis I

Ethical ideology should influence the severity of moral attribution.

Idealistic-absolutists should be least positive in their evaluation of

the actor, pragmatic-absolutists the most positive, and relativists

should fall intermediate.

Hypothesis II

IIA. The attributions of morality made by idealistic-absolutists

should emphasize the negative consequences of the action more than the

positive. Thus, the correlation between the evaluation of the quality

of the negative consequences and attributions of morality should be

greater in magnitude than the correlation between the evaluation of the

quality of the positive consequences and attribution of morality.

11B. The attributions of morality made by pragmatic-absolutists

should emphasize the positive consequences of the action more than the

negative. Thus, the correlation between the evaluation of the quality

of the positive consequences and attributions of morality should be

greater in magnitude than the correlation between the evaluation of the

quality of the negative consequences and attribution of morality.

IIC. The attributions recorded by relativists should emphasize the

positive and negative consequences equally. Thus, the magnitude of the

correlations between moral attribution and the evaluations of the quality

of positive and negative consequences will not be significantly different.

Hypothesis III

liIA. The magnitude of the correlation between the evaluation of

negative consequences and moral attribution should be greater for

idealistic-absolutists than pragmatic-absolutists.

IIIB. The magnitude of the correlation between the evaluation of

positive consequences and moral attribution should be greater for pragmatic-

absolutists than idealistic-absolutists.

IIIC. The correlation between the evaluation of positive conse-

quences for pragmatic-absolutists and the evaluation of negative conse-

quences for idealistic-absolutists should not differ from these same

correlations computed for relativists.


The 64 males and 64 females who completed the materials for

Experiment I were selected according to their scores on the first

administration of the EPQ. An equal number of individuals from each of

the four ethical ideologies tapped by the EPQ partially fulfilled a

course requirement by participating.

Via written instructions subjects were told that they would be

asked to read several stories, each of which would be followed by a

series of questions. The instructions stressed that after they had

carefully read a story, they were to answer each of the subsequent

questions so that their responses were an accurate indication of their

feelings. The four described situations, which are contained in Appendix

B, were similar in that in each one the central character's behavior

produced both positive and negative outcomes. These outcomes, which

included three good outcomes and three bad outcomes, were all the direct

result of the actor's behavior, and were selected on the basis of

extensive protesting. The four vignettes described: (1) a doctor

committing a mercy-killing, (2) a patient being visited by a friend who

negligently reveals confidential information to the patient, (3) a

psychologist conducting an experiment which involves a faked accident,

and (4) the director of a research firm misrepresenting the firm's ability

to successfully investigate a question posed by a client.

The dependent measures which followed each story asked subjects to

rate the quality of the consequences and to make moral attributions about

the actor and action. All these items employed 12-point scales with

verbal labels, and included: (a) ratings of the morality of the action

and actor on scales with labels of "extremely immoral," "immoral,"

"neither moral nor immoral," "moral," "extremely moral"; (b) ratings

of the "reason or motive behind the action" on a scale which used the

labels "extremely bad," "bad," "neither good nor bad," "good,"

"extremely good"; (c) the attribution of responsibility on a scale with

the labels "not at all responsible," "slightly responsible," "somewhat

responsible," "very responsible," and "completely responsible"; (d)

ratings of each individual consequence and all the consequences taken

together on scales with the labels "extremely bad," "bad," "neither

good nor bad," "good," "extremely good." The ratings of the three

positive consequences which accompanied each story were averaged together

to yield an overall evaluation of the good outcomes which resulted from

the action, while the ratings of the three negative consequences were

averaged together to yield an overall evaluation of the bad outcomes.


After reading a scenario, subjects were asked to rate the morality

of the actor and action, the quality of the actor's motive, and his/her

responsibility on five items, each with 12-point scales. These responses

were submitted to a 2 X 2 X 2 X 4 split-plot analysis of variance which

treated sex of subject, rule-universalism, and idealism as between-subjects

factors, and scenario as the within-subjects factor. Multiple comparisons,

when appropriate, were conducted using Duncan's multiple range test at the

5 percent level of significance.

Hypothesis I

The attribution of morality. Subjects rated the morality of the

actor, the action, and the degree to which they believed the action should

have been done on three separate items. A main effect of scenario

revealed in the analysis of the item "How moral do you feel [the actor]

was?" F (3, 360) = 81.23, p < .05, indicated that the actors in the

four different situations were rated somewhat differently. Examination

of the means for each vignette shows that the doctor was rated most

positively, followed by the psychologist, the friend visiting the patient,

and lastly, the research director; the means, in order, were 9.16, 8.06,

6.60, and 5.38. The only other significant main effect was for rule-

universalism with scenario; F (3, 360) = 2.61, p < .05. As Table 4-1

shows, on three of the four scenarios absolutists were more negative

in their attributions than were relativists (ps < .05).

The only other significant effect on the attribution of morality

item was an interaction of rule-universalism and idealism; F (1, 120) =

7.62, p < .05. As predicted, pragmatic-absolutists were more positive

than idealistic-absolutists in their moral attributions (p < .05); the

means were 6.4 and 7.6, respectively. Contrary to Hypothesis I, however,

relativists' judgments were more positive than idealistic-absolutists'

but did not differ from pragmatic-absolutists'; the means for the idealistic

and pragmatic relativists were 7.7 and 7.5, respectively.

A main effect of scenario, F (3, 360) = 90.50, p < .05, was also

in evidence on the item "How moral do you feel the [actor's] action was?"

Again, the doctor's action was rated most positively, followed by the

psychologist's, the friend's, and the researcher's; the means were 8.57,

7.48, 6.69, and 4.27. Table 4-1 again contains the means for the two-way

interaction of rule-universalism and scenario, F (3, 360) = 4.39, p <

.05, which qualified a significant main effect of rule-universalism. As

with moral evaluations of the actor, absolutists evaluated three of the

four actions more negatively than did relativists. Responses to the

item "Weighing the human values involved, do you feel that the [actor]

should have done what he/she did?" duplicated these effects. A

scenario main effect, F (3, 360) = 90.50, p < .05, was again obtained

as well as the rule-universalism X scenario interaction; F (3, 360) =

2.97, p < .05. The doctor's action was rated most positively followed

by the psychologist's, which was in turn rated more positively than

either the friend's or the researcher's. These final two actions were




L. Lf C- -n U% T-



0 N Lo O L0 CO l
.0 C LL-

LL 0

< O

S0 4-1
E a

S0U .0 04 -_ 0U 0 -
SN 0 CO -
- fl 00 co ( C
-. .O .* *. -
-n r co a r co in
3 l_ O" L O fo
a) >- m a)
4) > W 4-

0 I C c

L Ln
0 .0 mu u C 0-

C 0 '.0 co r- N 0 --
u I'D 0O N O ) 0)-

O 0 c 0o om co m c
Cu u L

U 0u
S tn n A (n un EC

o tn 0 U! (n ( U 41 i
- 4- *- Q- C
J3 C *- 3 *- 1 -0

S0 u 0 0 a) 0-
U. ) ac < o < '

4- ru 0
a 40
>0 L- II
44 =) 0
) in 0I m n .- 0a
L to C *- 0
a)m 0 0 *-u 0 C 0 0
m 1- 00 0 .0) 0
( 1 < 0 L -0 3Z
QX <2 >Ia

not evaluated differently. The means, in order, were 8.66, 7.75, 4.54,

and 4.21. The means for the two-way interaction of rule-universalism

and scenario, which parallel the other moral attribution items, are

shown in Table 4-1.

Motive. Subjects evaluated the quality of the actor's motive on

the item "To what degree do you think the reason or motive behind the

[actor's] action was good vs. bad?" A scenario main effect, F (3, 360) =

110.68, p < .05, followed the previously reported pattern for attributions

of morality; the means were 9.86, 9.24, 8.01, and 5.35. The three-way

interaction of rule-universalism, idealism, and scenario depicted in

Table 4-2 indicates that receptions of the actor's motive differed

dependent upon the ethical ideology of the attributor. For the two most

positively evaluated situations (i.e., the doctor who performs a mercy-

killing and the psychologist conducting an experiment), relativists

were more positive in their evaluation of the actor's motive than were

absolutists. However, on the intermediately evaluated vignette -- the

friend visiting a patient in a hospital -- pragmatic-absolutists were

more negative than judges of all the other ethical ideologies. Also,

idealistic-absolutists, while not differing from idealistic-relativists,

were more negative than pragmatic-relativists. These differences tended

to be reversed for the most negatively evaluated scenario which described

the research director misleading a client. In this instance, idealistic-

absolutists inferred a more negative motive than both pragmatic-

absolutists and idealistic-relativists.

Attribution of responsibility. Main effects of scenario, F (3,

360) = 12.68, p < .05 and rule-universalism, F (1, 120) = 3.68, p <

Table 4-2

The Effects of Ethical Ideology and Scenario on Perceptions
of Actor's Motives

Ethical Ideology

























Note: Means without a single common subscript differ by Duncan's
Multiple Range Test at the p = .05 level of significance.

.05, were in evidence on the item "In general, how responsible was the

[actor] for the outcomes which followed his action?" Unlike attributions

of morality and motive, the doctor was rated as the most responsible,

followed by the research director, the psychologist, and lastly, the

friend; the means were 9.16, 8.71, 8.14, and 7.51, respectively. In

addition, absolutists attributed more responsibility to the actor than

did relativistic judges; F (1, 120) = 3.68, p < .05. The means were

8.64 and 8.12.

Sex differences. Males and females differed in their attributions

on several items. The means for a three-way interaction of sex, idealism,

and scenario obtained on evaluations of the morality of the actor and

action, Fs (3, 360) = 2.96 and 3.15, ps < .05, shown in Table 4-3

reveal differences only for the doctor vignette. For both items,

idealistic females were more negative than idealistic males and all

pragmatists. A three-way interaction of sex, rule-universalism, and

scenario was also significant on the evaluation of the action item; F

(3, 360) = 3.14, p < .05. For males, absolutists were more negative

than relativists on the doctor and psychologist scenarios, but more

positive when the research director's action was being appraised. For

females, absolutists were more negative than relativists on the friend

visiting a patient vignette only. These means are presented in Table

4-4. Lastly, sex interacted significantly with scenario on judgments

of whether the action should have been done and attributions of

responsibility; Fs (3, 360) = 2.76 and 4.08, respectively (ps < .05).

The means for these interactions are presented in Table 4-5.

Hypothesis I: Summary. A consistent pattern of evaluations held

throughout subjects' attributions of morality, ratings of motives, and

Table 4-3

The Effects of Scenario and Idealism on Males and
Females Attributions of Morality

Doctor Psychologist Friend Researcher

Morality of the Actor
Pragmatic Males 9.25a 8.28b 6.87c 5.44d

Idealistic Males 9.53a 7.66b 6.56c 5.56d

Pragmatic Females 9.94a 8.28b 6.47c 5.53d

Idealistic Females 7.91b 8.03b 6.50c 5.00d

Morality of the Action
Pragmatic Males 8.69b 7.63c 5.94d 4.31e

Idealistic Males 8.94ab 7.38c 5.69d 4.25e

Pragmatic Females 9.44a 7.28c 5.50d 4.25e

Idealistic Females 7.22c 7.63 5.63d 4.25e

Note: For each dependent measure, means without a single common
subscript differ by Duncan's Multiple Range Test at the
S= .05 level.

Table 4-4

The Effects of Scenario and Universalism on Males and Females
Attribution of Morality


Doctor Psychologist Friend Researcher

Morality of the Action

Absolutistic Males 7.75c 6.72de 5.53fg 4.88gh

Relativistic Males 9.88a 8.28bc 6.09ef 3.69i

Absolutistic Females 8.00bc 7.50cd 5.03 4.25hi

Relativistic Females 8.66b 7.41cd 6.09ef 4.25hi

Note: Means without a single common subscript differ by Duncan's Multiple
Range Test at the p = .05 level.






c 0*
44 41

EL -
(U -
3 0


0 C)



C0 -





o c





u "-


o -4
o o

IfA -3-

u u
u-L -T

.0 0
oo o

co r-

-o ro
un co
C C 0

oo o-








0 )




r o




>C La





attributions of responsibility. Overall, the motives of the physician

who allowed his terminally ill patient to die were rated most positively

and he was delegated the most responsibility, but he was also rated to be

the most moral of all the depicted characters. On the other hand, the

research director who purposefully lied to his client was also perceived

to be highly responsible for the negative consequences which followed his

action, but he was evaluated as the most immoral and to have the worst


Attributions also varied according to ethical ideology. Overall,

relativists more than absolutists felt the action which was described should

have been done and attributed less responsibility to the actor. Second,

and as predicted by Hypothesis I, idealistic-absolutists attributed less

morality to the actor than did pragmatic-absolutists. Relativists,

however, did not fall intermediate but instead attributed the same

amount of morality to the actor as did pragmatic-absolutists.

Consequences and the Attribution of Morality

Hypothesis II. Only marginal support was obtained for the

prediction that idealistic-absolutists would emphasize negative conse-

quences, pragmatic-absolutists positive, and relativists both. Table

4-6 contains the correlations between the averaged evaluation of good

and bad consequences and attributions of morality. Inspection of these

coefficients indicates that for idealistic-absolutists the correlations

between good and bad consequences and attribution of morality are small

and about equal, failing to support Hypothesis IIA. Some support for

Hypothesis 11B was obtained, however, since for pragmatic-absolutists

the evaluation of the negative consequences was not significantly

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