Chapter 18 Rotter and Mischel: Cognitive Social Learning Theory 543A third common misconception is that the scale divides people into two distinct
types—internals and externals. Rotter (1975, 1990) insisted that generalized expectan-
cies imply a gradient of generalization and that, in certain specific situations, a person
with generally high feelings of internal control may believe that the outcome of his
or her behavior is due mostly to fate, chance, or the behavior of powerful others.
Fourth, many people seem to believe that high internal scores signify socially
desirable traits and that high external scores indicate socially undesirable charac-
teristics. Actually, extreme scores in either direction would be undesirable. Very
high external scores might be related to apathy and despair, with people believing
that they have no control over their environments, whereas extremely high internal
scores would mean that people accept responsibility for everything that happens
to them—business failure, delinquent children, other people’s misery, and thunder
storms that interfere with planned outdoor activities. Scores somewhere in between
these extremes, but inclined in the direction of internal control, would probably be
most healthy or desirable.
Interpersonal Trust Scale
Another example of a generalized expectancy (GE) that has provoked considerable
interest and research is the concept of interpersonal trust. Rotter (1980) defined
interpersonal trust as “a generalized expectancy held by an individual that the word,
promise, oral or written statement of another individual or group can be relied on”
(p. 1). Interpersonal trust does not refer to the belief that people are naturally good
or that they live in the best of all possible worlds. Neither should it be equated
with gullibility. Rotter saw interpersonal trust as a belief in the communications
of others when there is no evidence for disbelieving, whereas gullibility is foolishly
or naively believing the words of other people.
Because many of our rewards and punishments come from other people, we
develop generalized expectancies that some type of reinforcement will follow from
verbal promises or threats made by others. Sometimes these promises and threats
are kept; other times they are broken. In this way, each person learns to trust or
distrust the words of others. Because we have differential experiences with the
words of others, it follows that individual differences will exist among people with
regard to interpersonal trust.
To measure differences in interpersonal trust, Rotter (1967) developed an Inter-
personal Trust Scale, which asked people to agree or disagree to 25 items that assessed
interpersonal trust and 15 filler items designed to conceal the nature of the instrument.
The scale is scored on a 5-point gradation from strongly agree to strongly disagree
so that strongly agree and agree responses would indicate trust on 12 items and
strongly disagree and disagree responses would indicate trust on the other 13 items.
Table 18.3 reveals several sample items from Rotter’s Interpersonal Trust Scale.
Scores for each of the 25 items are added so that high scores indicate the presence
of interpersonal trust and low scores mean a generalized expectancy of distrust.
Is it more desirable to score high or low on the scale, to be trustful or dis-
trustful? When trust is defined independently of gullibility, as Rotter (1980) con-
tended, then high trust is not only desirable but essential for the survival of
civilization. People trust that the food they buy is not poisoned; that the gasoline
in their cars will not explode on ignition; that airline pilots know how to fly the