Applied & PreventivePsychology 1:111-117 (1992). CambridgeUniversityPress. Printed in the USA.

Copyright © 1992 AAAPP 0962-1849/92 $5.00 + .00

Cognates of personal control: Locus of control,

C H R I S T O P H E R PETERSON
University of Michigan, Ann Arbor

ALBERT J. S T U N K A R D
University of Pennsylvania, Philadelphia

Abstract
Theories of personal control are extremely popular in contemporary psychology. What are the relationships
among such apparently similar notions as locus of control, self-efficacy, and explanatory style? Although these
cognates have similar correlates, they are not necessarily interchangeable. Our conceptual analysis suggests
that they exist at different levels of abstraction and generality. They interact with one another to influence
behavior. They are not simply alternative ways of talking about the same phenomenon. "Personal control"
should be regarded as multidimensional.
Key words: Explanatory style, Locus of control, Personal control, Self-efficacy

Personal control refers to a person's beliefs about how effectance motivation. The experience of effective interac-
well he or she can bring about good events and avoid bad tion with the world he called a feeling of efficacy.
events. In a previous paper, we generalized across its In the 1960s, an important change in psychology--the
cognates (Peterson & Stunkard, 1989). Here we do the cognitive revolution--had sweeping consequences for
opposite, drawing distinctions among three particularly theories of personal control. This change arose from rec-
well-known renditions: locus of control, self-efficacy, and ognition of the inadequacy of formulations ignoring the
explanatory style. These notions have attracted a great mental life of people and found expression in a host of
deal of research attention, but usually within different cognitive theories (Gardner, 1985). In these theories, the
traditions. As a step toward bridging these lines of work, terminology of information processing replaced that of
we examine the meanings of these cognates. stimuli and responses, of needs and drives. Theories of
personal control were not lost in the cognitive revolution;
The Emergence of Theories of Personal Control they were simply recast in the new language. Effectance
became not a motive, with biological connotations, but
Theorists throughout this century have been concerned an idea (belief, expectation, attribution, perception). The
with similar notions (Peterson, 1991; Weiner, 1990). Ac- early theorists emphasized drives to master the environ-
cording to the early formulations, a basic aspect of ment, whereas the new generation of theorists spoke of
human nature is a drive to master the environment. Ex- beliefs about whether or not this could be done.
pression of this motive is associated with effective adap- Some have decried one aspect of the impact of the
tation; its thwarting results in poor functioning. These cognitive revolution on the psychology of human behav-
early ideas were synthesized by Robert White (1959) in ior. A view of people as information processors is neces-
his classic paper on competence. According to White, sarily incomplete because it contains no goal or value to
people are motivated to interact in an effective way with explain the end to which information is processed.
the world. The motivation to be competent he called Recognizing this problem, cognitive theorists usually
speak of people's personal control with respect to good
This paper is based on a report submitted to the Henry J. Kaiser or bad events, thereby implying that personal control is
Foundation. We thank Lisa M. Bossio for her editorial advice. deployed in a hedonistic context. However, this assump-
Send correspondenceand reprint requests to ChristopherPeterson,
Department of Psychology,Universityof Michigan,580 Union Drive, tion is rarely explicit, and researchers may run afoul of it
Ann Arbor, MI 48109. when concerned with complex behavior.
111
112 Peterson and Stunkard

Recasting effectance motivation in terms of beliefs had expectation, in turn, is determined first by task-specific
two effects. First, the attention of researchers was di- characteristics and second by generalized expectations
rected toward specific aspects of the person. Beliefs are about the nature of reward, which Rotter termed locus of
always about something. Motivation can be general, but control.
beliefs are always specific. Prediction is greatly improved In its extreme cases, locus of control is represented by
by such specificity (e.g., Fishbein & Ajzen, 1974). an internal (I) orientation, in which the individual be-
Second, the attention of researchers was also directed lieves that rewards are brought about by his own actions,
onto the actual environment and the way that a person versus an external (E) orientation, in which the individ-
interacts with it. Whereas motives reside within a person, ual believes that rewards are due to the operation of
beliefs refer to both the person and aspects of his or her chance factors, fate, or powerful others. Locus of control
world. Again, current opinion holds that behavior is is therefore an individual difference, measured by re-
not well explained by looking just within the person. In- sponses to the Rotter (1966) I-E scale or similar self-
stead, understanding is greatly bolstered by taking into report instruments, which ascertain the degree to which
account both individual and situational characteris- someone endorses internal versus external statements.
tics (e.g., Bowers, 1973; Endler & Magnusson, 1976; Mis- Thousands of investigations have looked at locus of
chel, 1979). control. Phares (1978) summarized the thrust of these
The consequences of people's beliefs about control investigations by noting that "Our survey of the I-E liter-
were examined in a variety of paradigms in the 1960s in ature has revealed the typical internal to be one who
which control was operationalized as choice, predictabil- actively comes to grips with the world. Compared to the
ity, and contingency. Research results converged, show- external, the internal is resistant to social pressure and
ing that control was usually beneficial. dedicated to the pursuit of excellence" (p. 295). Like
Much of the research on personal control has taken others who comment on the locus of control area of
place within social psychology, reflecting a concern with research, Phares added a disclaimer that internality is not
the social environment in fostering or inhibiting per- necessarily good. Still, the bulk of the studies show that
ceived control and showing the influence of personality in a responsive environment--and this qualification is
psychology, in conceiving personal control as an individ- critical--individuals with an internal locus of control ac-
ual difference, and of clinical psychology, in designing crue to themselves all manner of benefits.
interventions to combat disorders of personal control. Rotter (1975) stressed that locus of control is but one
People do not need to exercise control to benefit from of the determinants of how a person responds to events.
it (Averill, 1973; Miller, 1979; Thompson, 1981). The Some researchers, however, use locus of control as a
mere perception of control is sufficient to reduce stress, personality disposition out of context, examining its
increase motivation, encourage performance, and so causes, consequences, and correlates and ignoring the
forth. For instance, Glass and Singer (1972) exposed factors deemed critical by Rotter in determining the
laboratory subjects to bursts of aversive noise. Half the exact nature of these relationships. Chief among these
subjects were told that they could terminate the noise by factors are the reinforcement value and the person's fa-
pushing a button; the other subjects were not given this miliarity with the task at hand.
instruction. All were then tested at a proofreading task. To understand the relationship between locus of con-
The subjects who believed themselves to have control trol and some behavior, the value of any reinforcement
over the original noise, even though they never pushed of that behavior must be known. Rotter (1975) criticized
the button, did better than the other subjects, despite the research from the 1960s in which a relationship was
fact that both groups experienced the identical (and in sought between locus of control and political activism.
actuality uncontrollable) noise. Does locus of control predict participation in a protest
Numerous theories of personal control compete in the march? One prediction is that internality leads to social
professional literature. We consider several of these in activism (presumably an attempt to change the world
turn and conclude that these cognates are not strictly through one's actions), but this explanation may be too
interchangeable; there are good reasons to keep in mind simplistic. Internal people may not protest, join a protest
the distinctions among them. group, or sign a petition simply because they do not
believe in the cause. On the other hand, very external
people may join a protest group because they like the
Locus of Control
other people who are members of the group, because it is
Julian B. Rotter (1954, 1966, 1975) proposed an influen- less boring than studying, because it will upset their par-
tial cognate of personal control within the context of his ents if they find out, because it is the conforming thing to
social learning theory. According to Rotter, reinforce- do, and so on (Rotter, 1975, p. 270).
ment strengthens responses to the degree that the individ- Rotter (1975) also pointed out that locus of control, as
ual expects the response to lead to further rewards. This a broad belief, is less likely to be related to behavior to
 Cognates of Personal Control 113

the degree that the person is in a situation that is struc- how confident they are, from not at all to completely,
tured, familiar, and unambiguous. In such circum- that they can perform a particular behavior in a given
stances, a person's specific expectation about the nature setting. An important outcome of Bandura's research is
of reward takes over. Rotter (1975) cited the example of the demonstration that an individual's sense of control
the relationship between locus of control and academic over a given response ("I am confident that I can let the
achievement. At early ages, internality predicts grades. spider walk up my arm") is a better predictor of this
At older ages, it does not. This makes perfect sense, response than the individual's past success or failure in
because with age the student learns more about the spe- performing it. One of psychology's few truisms is that
cific tasks required to achieve specific outcomes in partic- past behavior predicts future behavior; Bandura's re-
ular subjects. search on self-efficacy improves upon this idea.
Locus of control research may have peaked in popu- Bandura has used phobias as a paradigm with which
larity, but only because it has stimulated a host of com- to assess the effects of diminished and enhanced self-
petitors. One trend apparent in more recent I-E research efficacy. However, his theory extends far beyond this
is the development of domain-specific locus of control domain. Bandura (1977) argued that self-efficacy under-
measures (e.g., health locus of control). Relatedly, some lies all of behavior change, including actions involved in
researchers have called for a distinction between locus of health promotion (O'Leary, 1985).
control for good outcomes versus locus of control for Several additional notions of Bandura warrant men-
bad outcomes, resulting in a movement toward greater tion. First, he distinguishes between an individual's ef-
specificity of the notion. This specificity aids prediction ficacy expectancy and outcome expectancy. Outcome ex-
but detracts from Rotter's original conception of locus of pectancy is a belief about the consequences of a
control as a generalized expectation. particular behavior. A basketball coach might believe,
for instance, that the Chicago Bulls can be defeated
if they are held to fewer than 90 points. Whether he be-
Self-Efficacy
lieves that his team can hold the Bulls to fewer than 90
Albert Bandura's (1977, 1978, 1982b, 1986) widely points is of course a different matter. This is an efficacy
known notion of self-efficacy is an outgrowth of his ver- expectancy.
sion of social learning theory, which emphasizes the im- This distinction is important, even though Bandura
portance of vicarious processes in the acquisition and himself can be faulted for overlooking it at times. In
modification of behaviors. People learn through observa- some of his writing, he appears to ascribe people's diffi-
tion of others. Wholly novel responses may be performed culties mainly to efficacy expectations rather than to the
after relevant models are seen performing similar actions joint effect of efficacy and outcome expectations.
with desirable consequences (Bandura & Waiters, 1963). Second, Bandura restricts the study of personal con-
Much of Bandura's research efforts have been in the trol to very specific contexts, using a "microanalytic
context of behavior modification. How can phobic be- strategy." In keeping with his social learning bent, he
havior be changed? If changed, what is the critical ele- distrusts the possibility of broad dispositions like locus of
ment responsible? Phobias can be alleviated through control. Instead, he always specifies self-efficacy with re-
modeling. An individual afraid of spiders may stop spect to some particular situation and some particular
avoiding them and even be able to handle them without response.
fear after watching another individual perform such ac- Bandura's theory has been criticized as common-sensi-
tions and may also come to perform nonphobic behav- cal (Smedslund, 1978) and as too specific (Peterson &
iors vis-fi-vis spiders that were not explicitly modeled. Stunkard, 1989). Nonetheless, it has become a leading
What is going on here? According to Bandura, at the perspective in contemporary psychology because of the
heart of a phobia is the belief that one cannot cope with range of its applications and its ability to generate effec-
the phobic object. Phobic individuals believe that they tive strategies of behavior change (Peterson, 1992).
cannot perform the behaviors needed to master the situa-
tion presented by the spider, the wide-open space, or the
speech to be given. Modeling is effective to the degree Explanatory Style
that it strengthens the individual's personal control vis-fi- The notion of explanatory style emerged from the attri-
vis the situation in question. Bandura termed this self- butional reformulation of the learned helplessness
efficacy. A heightened sense of self-efficacy leads to model. This model is an account of why people act in an
changes in behavior not specifically modeled, to a de- inappropriately passive way and why they fail to cope
crease in emotionality, and even to a normalization of with the demands of a situation that seems fully within
catecholamine metabolism. their competence. According to Seligman (1975), people
In this type of work, self-efficacy is usually measured may act helpless because they have learned to be helpless.
with a simple rating scale. Subjects are asked to indicate The critical element in this learned helplessness is a di-
114 Peterson and Stunkard

minished sense of personal control. Thus, learned help- most thorough such account is that of Dweck (1975;
lessness theory is most directly an explanation of why Dweck & Licht, 1980), who contrasts helplessness-ori-
and how personal control can go awry. Less directly, it is ented children with those she calls mastery oriented.
an account of how persot]al control can be fostered. These two groups of children differ in the persistence they
According to the learned helplessness model, helpless- show in the face of failure. Mastery-oriented children are
ness results from experience with uncontrollable events less likely to acknowledge failure. What appear as set-
(Maier & Seligman, 1976). Uncontrollability is defined in backs to the observers are regarded by these children as
terms of the contigency between responses and the out- steps toward the final goal.
comes of concern. When events occur regardless of what Explanatory style has been frequently investigated, yet
the subject does or does not do, then the events are important questions remain unexplored (Peterson, 1991).
uncontrollable. When the subject learns this (non)contin- The hypothesized roles of the individual dimensions of
gency and represents it as an expectation of future un- explanatory style have not been systematically investi-
controllability, learned helplessness occurs. gated. Also, the attributional reformulation of helpless-
A full explanation of when helplessness does or does ness theory proposes that explanatory style interacts with
not follow uncontrollable events should consider the per- bad events to produce difficulties, yet the vast majority of
son's causal interpretation of the events (Abramson, studies have not ascertained the occurrence of bad
Seligman, & Teasdale, 1978). According to a reformula- events. Finally, the helplessness model is a process ac-
tion of helplessness theory that considers these interpre- count of how difficulties develop. Because most studies
tations, when people encounter an uncontrollable event, use a cross-sectional design, the presumed sequence of
they ask "why?" Their answer dictates their response to events linking explanatory style and failures of adapta-
uncontrollability (Peterson & Seligman, 1984). tion has not been mapped out.
Three dimensions of causal explanation are important.
An internal explanation ("it's me") makes self-esteem
Taking Stock
loss more likely than an external explanation ("it's the
economy"). A stable explanation ("it's going to last for- Locus of control, self-efficacy, and explanatory style
ever") leads to more prolonged helplessness than an un- share a family resemblance (Peterson & Stunkard, 1989).
stable explanation ("it's one of those days"). A global Each construct is explicitly cognitive. Each is related to
explanation ("it's going to undermine everything that I good versus bad functioning and in particular the vigor
do") produces more pervasive deficits than a specific ex- or passivity with which someone meets the demands of
planation ("it's the heat in that place"). the world. Indeed, the following summary of research
An individual's causal explanation for uncontrollable findings (Janis, 1983) applies equally well to all three
events determines the extent of helplessness following constructs:
these events. Because individuals show a characteristic
style of offering causal attributions, what is called their "Perceived control" is one of the core concepts that
explanatory style, habitually favoring certain types of has evolved from research bearing on the sense of
explanations over others, some people are more at risk mastery. Loss of perceived ability to control aversive
for helplessness than others. One's explanatory style can events, which is strongly influenced by environmental
be measured with a questionnaire developed for this pur- circumstances, is now generally recognized as a major
pose (Peterson et al., 1982). An explanatory style in psychological determinant of reactions to stressful life
which uncontrollable events are attributed to internal, events. When a person notices that protective actions
stable, and global explanations is a risk factor for help- are having little observable effect in bringing an end to
lessness and passivity in the face of uncontrollability, an extremely disagreeable experience, his or her initial
whereas the converse style makes an individual robust reaction is usually an upsurge of anger or protest. If
and impervious to disruption by failure (Peterson & the person's efforts to regain a sense of control con-
Seligman, 1984). Recent evidence suggests that stable tinue to be thwarted, he or she is likely to become
and global explanations for bad events can be risk factors demoralized. After that happens, the person copes less
for physical illness (Peterson & Bossio, 1991). effectively and ultimately develops profound feelings
Helplessness theory emphasizes how personal control of helplessness and depression. These extreme reac-
can be thwarted in one situation and have undesirable tions, which are usually accompanied by apathy and
consequences in another. Other theories have attempted social withdrawal, are pertinent to both mental health
to depict the other side on the coin, the encouragement and physical health. There is a growing body of evi-
of personal control through mastery experiences. The dence that the malignant emotional sequence as-
constructs of these theories have different names: learned sociated with loss of perceived control, which often
industriousness, learned resourcefulness, self-control, occurs among people who are ill or incapacitated, not
transfer of persistence, and so on. At the present time, the only increases subjective suffering but also impedes
 Cognates of Personal Control 115

physical recovery and sometimes leads to untimely Each cognate--by definition--is expected to predict
death. Fortunately, however, there is also evidence some aspect of behavior. Prediction would be bolstered
that the malignant sequence can be prevented or inter- and explanation aided, however, if researchers also took
rupted by psychological interventions that enable dis- into account the other determinants of behavior explic-
tressed people to see themselves as having sufficient itly recognized by Rotter, Bandura, and explanatory
control over what happens to them to cope success- style theorists. Locus of control should be measured
fully. (p. 10) in conjunction with a person's sense of the reinforce-
ment value of particular activities; self-efficacy should
In view of the conceptual and empirical overlap, should be looked at along with outcome expectancies; explana-
we regard these different cognates of personal control as tory style should be studied along with the actual causal
interchangeable? Are they simply different ways of talk- texture of settings. Common to all these theories is the
ing about the same psychological phenomena? need to locate people in the environments in which they
We think not. Locus of control, self-efficacy, and ex- behave.
planatory style do have similar correlates, but if we are These research traditions can be sensibly bridged only
content with this summary statement, we give up any by bringing constructs together with their full theoretical
attempt to explain just why these beliefs translate them- meanings. Simply including measures of all three con-
selves into effective coping on the one hand versus inef- structs in a battery may not be informative. No doubt
fective coping on the other. Each construct is defined correlations in the .30 range would be obtained (Mischel,
within an explicit theory, and the theories themselves 1968), but so what?
differ concerning the determinants of action and emo- Longitudinal studies would be more informative. If
tion. Collapsing these constructs reduces the benefits of these notions do interact, it is over time. As generalized
having a theory in the first place. predispositions, locus of control and explanatory style
The meanings of these three cognates of personal con- might influence one's efficacy in particular settings. Yet
trol are far from identical. Locus of control refers to one could argue as well that they should influence one's
one's generalized expectancies about the origin of re- outcome expectancies more than efficacy expectancies
wards and punishments in the world, self-efficacy refers because they are beliefs about the world, not simply
to one's belief about whether a given behavior can be about one's own abilities.
enacted, and explanatory style refers to one's habitual Interestingly, the more exact parallels among these
way of explaining the causes of events. These three con- theories are in Bandura's outcome expectancy. Each the-
cepts are clearly distinct. An individual with an internal ory assigns an important role to one's expectation about
locus of control may offer external causal explanations: the relationship between behaviors and outcomes; in-
"If I am charming, I will be offered the job, but whether deed, expectation arguably is the crucial determinant of
or not I am charming depends on the mood of the inter- behavior in each theory. Yet what has interested re-
viewer." Individuals may entertain efficacious beliefs searchers, including those who proposed each theory, is
about their ability to perform a given behavior indepen- not the part of the theory stressing expectation but rather
dently of their locus of control or explanatory style. the part stressing personal control.
Each cognate exists at a different level of abstraction These different notions may each do their best job
and generality. Locus of control and explanatory style within some given domain of behavior. To judge simply
are, by definition, predispositions to more specific from the most popular areas of work, locus of control
thoughts and beliefs that in turn are the proximal deter- accounts for perseverance, self-efficacy for behavior
minants of action and emotion. Self-efficacy, in contrast, change, and explanatory style for demoralization. If re-
is defined as one of these proximal determinants. Al- searchers played these domains off against one another,
though whether people can be characterized as generally evidence for the discriminant validity of these cognates
efficacious or not is an empirical question, certainly no might result.
researcher should start with this assumption. Some ques- The correlates of locus of control, self-efficacy, and
tionnaires purport to measure "general" self-efficacy, explanatory style might look more distinct than they
quite apart from specific behaviors, but they are not currently do if criterion measures were selected for ex-
doing the concept justice. plicitly theoretical reasons. Explanatory style, for exam-
Each cognate has its richest--perhaps its only--mean- ple, has been frequently looked at in relation to depres-
ing within its particular theory. The widespread tendency sion. Theoretically, however, the severity of depressive
to pull a personal control cognate out of its theoretical symptoms need bear no particular relationship to a "de-
context and include its measure in a questionnaire bat- pressive" explanatory style. Rather, the duration and
tery is apt to yield exactly what we find: an array of generality of depressive symptoms should be under the
modest correlates yet no crisp conclusions about why sway of explanatory style. Investigations of this theoreti-
variables cohere as they do. cally crisp hypothesis are virtually nonexistent.
116 Peterson and Stunkard

Locus of control, self-efficacy, and explanatory style trol. The theoretical distinctiveness of these constructs
have become popular mostly because of the availability may be blurred by the extent to which specific question-
of straightforward questionnaire measures. In each case, naire items reflect more than one construct. To answer an
however, measurement is far from finished business. The explanatory style questionnaire item by offering the attri-
questionnaires used to ascertain locus of control and bution that a bad event happened because "I couldn't help
explanatory style have been criticized, and attempts to myself" is to offer a self-efficacy appraisal. Researchers
bolster their reliability and validity are ongoing (e.g., investigating two or more of these personal control cog-
Peterson & Villanova, 1988). nates are advised to scrutinize closely the particular opera-
Bandura's procedure of assessing self-efficacy by ask- tionalizations they employ to eliminate or control for
ing subjects to make simple confidence ratings has not items that tap more than one construct (Nicholls, Licht, &
often been criticized, perhaps because it seems so face Pearl, 1982).
valid. Nonetheless, in view of the microanalytic strategy It is premature to regard personal control as a mono-
favored by Bandura, reactivity on the part of subjects lithic notion. Overlap exists among such well-known
may threaten some of the obtained results. Rating one's constructs as locus of control, self-efficacy, and explana-
self-efficacy immediately before engaging (or not) in tory style, but the differences are just as important as the
the behavior in question might create stronger links be- similarities. The differences may be the most interesting
tween self-efficacy and behavior than actually exist. aspects of these constructs, which means that we should
Bandura (1982a) has argued against this possibility, bridge these research traditions with care so as not to
but we still suspect that it may occasionally confound strip away theoretical subtlety. Psychology has long suf-
research. fered from a lack of consensus about its appropriate
Another measurement issue pertains to the possible units (Allport, 1937). Although there is agreement about
semantic overlap between questionnaire items used to the importance of personal control, this does not mean
operationalize these different cognates of personal con- that we yet know how best to conceptualize it.

REFERENCES
Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Endler, N. S., & Magnusson, D. (1976). Toward an interac-
Learned helplessness in humans: Critique and reformula- tional theory of personality. Psychological Bulletin, 83, 956-
tion. Journal of Abnormal Psychology, 87, 49-74. 974.
AUport, G. W. (1937). Personality: A psychological interpreta- Fishbein, M., & Ajzen, I. (1974). Attitudes toward objects as
tion. New York: Holt. predictors of single and multiple behavioral criteria. Psycho-
Averill, J. R. (1973). Personal control over aversive stimuli and logical Review, 81, 59-74.
its relationship to stress. Psychological Bulletin, 80, 286-303. Gardner, H. (1985). The mind's new science: A history of the
Bandura, A. (1977). Self-efficacy theory: Toward a unifying cognitive revolution. New York: Basic Books.
theory of behavioral change. Psychological Review, 84, 191- Glass, D. C., & Singer, J. E. (1972). Urban stress: Experiments
215. on noise and social stressors. New York: Academic Press.
Bandura, A. (1978). The self system in reciprocal determinism. Janis, I. L. (1983). Foreword. In E. J. Langer (Ed.), The
American Psychologist, 33, 344-358. psychology of control (pp. 9-11). Beverly Hills, CA:
Bandura, A. (1982a). The assessment and predictive generality Sage.
of self-precepts of efficacy. Journal of Behavior Therapy and Maier, S. F., & Seligman, M. E. P. (1976). Learned helpless-
Experimental Psychiatry, 13, 195-199. ness: Theory and evidence. Journal of Experimental Psychol-
Bandura, A. (1982b). Self-efficacy mechanisms in human ogy, 105, 3--46.
agency. American Psychologist, 37, 122-147. Miller, S. M. (1979). Controllability and human stress:
Bandura, A. (1986). Social foundations of thought and action: A Method, evidence, and theory. Behaviour Research and
social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Therapy, 17, 287-304.
Bandura, A., & Waiters, R. H. (1963). Social learning and Mischel, W. (1968). Personality and assessment. New York:
personality development. New York: Holt, Rinehart, & Wiley.
Winston. Mischel, W. (1979). On the interface of cognition and personal-
Bowers, K. S. (1973). Situationism in psychology: An analysis ity: Beyond the person-situation debate. American Psychol-
and critique. Psychological Review, 80, 307-336. ogist, 34, 740--754.
Dweck, C. S. (1975). The role of expectations and attributions Nicholls, J. G., Licht, B. G., & Pearl, R. A. (1982). Some
in the alleviation of learned helplessness. Journal of Person- dangers of using personality questionnaires to study person-
ality and Social Psychology, 31, 674--685. ality. Psychological Bulletin, 92, 572-580.
Dweck, C. S., & Licht, B. G. (1980). Learned helplessness and O'Leary, A. (1985). Self-efficacy and health. Behaviour Re-
intellectual achievement. In J. Garber & M. E. P. Seligman search and Therapy, 23, 437-451.
(Eds.), Human helplessness: Theory and applications (pp. Peterson, C. (1991). The meaning and measurement of explana-
197-221). New York: Academic Press. tory style. Psychological Inquiry, 2, 1-10.
 Cognates of Personal Control 117

Peterson, C. (1992). Personality (2nd ed.). San Diego: Harcourt Rotter, J. B. (1966). Generalized expectancies for internal ver-
Brace Jovanovich. sus external control of reinforcement. Psychological Mono-
Peterson, C., & Bossio, L. M. (1991). Health and optimism. New graphs, 80(1, Whole No. 609).
York: Free Press. Rotter, J. B. (1975). Some problems and misconceptions relat-
Peterson, C., & Seligman, M. E. P. (1984). Causal explanations ed to the construct of internal versus external reinforce-
as a risk factor for depression: Theory and evidence. Psycho- ment. Journal of Consulting and Clinical Psychology, 43,
logical Review, 91, 347-374. 56~7.
Peterson, C., Semmel, A., yon Baeyer, C., Abramson, L. Y., Seligman, M. E. P. (1975). Helplessness: On depression, develop-
Metalsky, G. I., & Seligman, M. E. P. (1982). The Attribu- ment, and death. San Francisco: Freeman.
tional Style Questionnaire. Cognitive Therapy and Research, Smedslund, J. (1978). Bandura's theory of self-efficacy: A set of
6, 287-299. common sense theorems. Scandanavian Journal of Psychol-
Peterson, C., & Stunkard, A. J. (1989). Personal control and ogy, 19, 1-14.
health promotion. Social Science and Medicine, 28, 819-828. Thompson, S. (1981). Will it hurt less if I can control it? A
Peterson, C., & Villanova, P. (1988). An expanded Attribu- complex answer to a simple question. Psychological Bulletin,
tional Style Questionnaire. Journal of Abnormal Psychology, 90, 89-101.
97, 87-89. Weiner, B. (1990). Searching for the roots of applied attribution
Phares, E. J. (1978). Locus of control. In H. London & J. E. theory. In S. Graham & V. S. Folkes (Eds.), Attribution
Exner (Eds.), Dimensions of personality (pp. 263-304). New theory: Applications to achievement, mental health, and inter-
York: Wiley. personal conflict (pp. 1-13). Hillsdale, NJ: Erlbaum.
Rotter, J. B. (1954). Social learning theory and clinicalpsychol- White, R. W. (1959). Motivation reconsidered: The concept of
ogy. Englewood Cliffs, NJ: Prentice-Hall. competence. Psychological Review, 66, 297-333.