Control and Efficacy as Interdisciplinary...

Review of General Psychology1999, Vol. 3, No. 4,317-337

Copyright 1999 by the Educational Publishing Foundation1089-268O/99/S3.OO

Control and Efficacy as Interdisciplinary Bridges

Jonathan HaidtUniversity of Virginia

Judith RodinUniversity of Pennsylvania

Control and efficacy are ideally suited as "bridges" or linking constructs for socialscientists working at different levels of analysis. Control and efficacy depend on the fitbetween individuals and the social systems in which they are embedded, and controland efficacy have measurable effects on neurotransmitter levels and endocrine re-sponses. This article presents an interdisciplinary perspective on control and efficacy.The authors survey the history of control-related constructs in psychology, from theirroots in animal learning to the present cognitive focus on beliefs about control. Theythen point out connections "up" to the sociological level and "down" to the physiologi-cal level. They propose a taxonomy of 6 useful constructs organized into 3 perspectives:motivational, cognitive, and systemic. Such a multilevel, multidisciplinary approachmay be particularly useful for approaching large real-world problems such as improv-ing schools or neighborhoods.

People generally like to control their environ-ments and their fates. They fight revolutions,

Jonathan Haidt, Department of Psychology, University ofVirginia; Judith Rodin, Department of Psychology, Univer-sity of Pennsylvania.

Funding for the report on which this article is based wasprovided by the Network on the Determinants andConsequences of Health Promoting and Health DamagingBehavior of the John D. and Catherine T. MacArthurFoundation Program on Mental Health and Human Develop-ment. Network administrator Grace Castellazzo providedinvaluable assistance from start to finish. The report is basedon 67 interviews conducted with members in each of theresearch networks of the MacArthur Health Program, as wellas with experts outside of the program. We thank thefollowing people, who gave generously of their time: NancyAdler, Joseph Allen, Barbara Anderson, Paul B. Baltes,Albert Bandura, Todd Bartko, Andrew Baum, Lisa Berk-man, Orville Gilbert Brim, Robert Cairns, Thomas Cook,Catherine Cooper, Richard Davidson, Michael Detke, RobinDiMatteo, Felton Earls, Jacquelynne Eccles, Glen Elder,Delbert Elliott, Susan Folkman, Ellen Frank, Frank Fursten-berg, Lynne Geitz, Henry Gleitman, Ira Harkavy, MardiHorowitz, James House, Richard Jessor, Boo Johannson,James Johnson, Robert Kahn, Ronald Kessler, Corey Keyes,David Kinney, David Kupfer, Margie Lachman, WilliamLovallo, Gerald McClearn, Hazel Markus, Karen Mathews,Bruce McEwen, John Monahan, Janet Moyer, John Nessel-roade, Mary Beth O'Hagen, Robert Rescorla, Robert Rose,Robert Rosen, Paul Rozin, Arnold Sameroff, Robert Sampson,Michael Scheier, Richard Schulz, Dianne Scott-Jones, MelvinSeeman, Teresa Seeman, Martin Seligman, Richard Settersten,Shobha Shagle, Richard Shweder, Burton Singer, J^aurenceSteinberg, Shelley Taylor, Christine Timko, Michael Useem,Karen Walker, Daniel Wegner, and William Julius Wilson.

Correspondence concerning this article should be ad-dressed to Jonathan Haidt, Department of Psychology,University of Virginia, Charlottesville, Virginia 22903.Electronic mail may be sent to [email protected].

consult astrologers, and buy air conditionerswith remote controls. They believe that theyhave control in many cases in which they havenone (Langer, 1983), and they live longer,healthier lives when their sense of control isaugmented (Langer & Rodin, 1976; Rodin &Langer, 1977; Taylor, 1989).

The construct of control has played a majorrole in modern psychology. The recognition ofan intrinsic motivation to affect the environmentwas an important challenge to behaviorism(White, 1959), and research on control hascontinued steadily since the 1950s. Lefcourt's(1966) review of the literature on internal versusexternal control of reinforcement is one of the10 most cited articles ever published in thePsychological Bulletin, and control has becomea central topic in social, developmental, clinical,health, and community psychology, as well as inseveral areas of sociology.

But research on control has in a way beenhampered by its own success. Control is relevantto so many areas of investigation that research-ers have developed dozens of control-relatedconstructs and measures, including locus ofcontrol, learned helplessness, self-efficacy, mas-tery, personal causation, personal competence,self-determination, autonomy, agency, empower-ment, and instrumentality. Skinner (1996) re-cently helped to clarify the situation by publish-ing a guide to more than 100 control-relatedconstructs. Skinner's guide organizes a tremen-dous amount of control-related research, but it

317

318 HAIDT AND RODIN

also reveals an implicit assumption guiding thatresearch: that it is people's beliefs about controlthat really matter. Psychological research oncontrol has focused primarily on people'sconsciously accessible and reportable beliefsand expectancies about the causal links amongagents, means, and ends.

The major claim of the present article is thatcontrol should be viewed in a broader and moreinterdisciplinary framework. The study of con-trol has been narrowed by the cognitiverevolution to the study of control beliefs,thereby losing touch both with the motivationalroots of the field in animal learning and with theimportance of control in sociological thinking.In this article, we describe three perspectives oncontrol that can help psychologically orientedresearchers move "up," to the sociologicallevel, and "down," to the biological level.Control and efficacy therefore provide interdisci-plinary bridges for the social sciences.

In the sections that follow, we begin with abrief history of control constructs in psychology,focusing on the shift from motivational tocognitive approaches. We then move up to thesociological level, showing how control servesas a "mind-society" bridge, and down to thebiological level, showing how control serves asa "mind-body" bridge. Finally, we present ataxonomy of control-related constructs whosethree main branches represent three overarchingperspectives on human behavior: motivational,cognitive, and systemic.

Control in the History of Psychology

The following literature review is highlyselective, taking a historical view of the fourmajor conceptualizations that have guidedpsychological research. More comprehensivereviews of current conceptualizations are avail-able in Bandura (1997) and Skinner (1995).

White's Effectance Motivation

The dominant schools of psychology atmidcentury—behaviorism and psychoanalysis—both denied the importance of human agency.They viewed behavior as determined by a fewpowerful drives whose expression was shapedby prior experience and present stimuli. Yet,within both schools, inexplicable findings were

arising that could not be made to fit intoorthodox drive models.

In 1959, Robert White surveyed both litera-tures and concluded that they had each run upagainst an important motivation that he labeledeffectance, which is a striving for "compe-tence." Animal learning researchers were find-ing that the three drives of hunger, thirst, and sexwere not sufficient to explain animal behavior.Rats, dogs, and monkeys would work, and evencross an electric grid, for the privilege ofexploring new territory or examining andmanipulating novel objects. Some researchersproposed that the list of drives ought to beexpanded to include drives for exploration(Butler, 1958), activity (Kagan & Berkun,1954), and manipulation (Harlow, 1953).

Simultaneously, psychoanalytic theorists werefinding that the two basic instincts of libido andaggression were not sufficient to explain chil-dren's behavior. Children showed a distinctpleasure in learning and perfecting skills,leading Hendrick (1942) to propose an addi-tional "instinct to master," which he character-ized as "an inborn drive to do and to learn howto do" (as cited in White, 1959, p. 307). Theinstinct to master yields pleasure when anefficient action "enables the individual tocontrol and alter his environment" (White,1959, p. 307). Similarly, Erikson (1953) pro-posed that children need "a sense of being ableto make things and make them well and evenperfectly: This is what I call the sense ofindustry" (as cited in White, 1959, pp. 311-312).

White's broad survey went on to describesimilar findings and proposals in other areas ofpsychology. Murray's (1938) well-known tax-onomy of needs included needs for "achieve-ment" and "construction." Maslow (1955)compared deficiency motivations with "growthmotivations," placing "self-actualization" at thetop of his hierarchy of needs. In developmentalpsychology, Piaget (1952) remarked on thechild's special interest in objects that areaffected by his or her own movements, harken-ing back to Groos's (1901) analysis of play, andthe child's "joy in being a cause."

It appeared that researchers in many areas ofpsychology were finding that they needed moreagent-centered and growth-oriented theories.Humans and other mammals seemed to beintrinsically motivated to explore their environ-ments, to interact with them, and to affect them.

CONTROL AND EFFICACY 319

In humans, this motivation seemed to take theform of a striving for skill acquisition, mastery,or competence. White (1959) argued that aninnate striving for competence and masterymade good evolutionary sense, particularly forhuman beings, for whom "so little is providedinnately and so much has to be learned throughexperience" (p. 329). White labeled this motiva-tion effectance and stated that its aim, or itsparticular pleasure, is the "feeling of efficacy."

But whereas White placed effectance on a parwith hunger as a basic human motivation, hestressed that effectance could not simply beadded to the list of deficit drives, because it doesnot aim for homeostasis. Effectance is, rather,the natural condition of human beings movingabout in the world, except during "episodes ofhomeostatic crisis," when we call on the skillswe have been so assiduously mastering. There isno minimum daily requirement for feelings ofefficacy, but neither is there any limit, point ofsatiety, or end of striving for competence:"Because there is no consummatory climax,satisfaction has to be seen as lying in aconsiderable series of transactions, in a trend ofbehavior rather than a goal that is achieved"(White, 1959, p. 322).

A number of theorists have extended White'smotivational view of control and efficacy.DeCharms (1968) asserted that "man's primarymotivational propensity is to be effective inproducing changes in his environment. Manstrives to be a causal agent. His nature commitshim to this path and his very life depends on it"(p. 269). Brehm (1966; Wortman & Brehm,1975) also thought there is an intrinsic need forcontrol and that whenever the ability to controlone's outcomes is taken away, the emotionalresponse is reactance, a state involving in-creased arousal and anxiety during which theperson attempts to recover control. Harter(1978) recast White's model in developmentalterms, showing how a child's effectance motivemight be strengthened or weakened by theactions of socializing agents. And the RochesterHuman Motivation Research Group developedWhite's motivational approach into a moremodern theory of intrinsic motivation, includinginnate needs for competence, autonomy, andrelatedness (Connell, 1990; Deci, 1975; Deci &Ryan, 1985; Skinner & Connell, 1986).

In summary, White (1959) criticized the drivetheories of the 1950s, yet he stayed within a

motivational framework. He developed anenhanced motivational perspective in whichpeople and animals have agentic motivations.We now move on to the next wave of theoristswho explored human agency, yet who did sousing the tools of the new cognitive revolution:beliefs, expectancies, and cognitive structures.

Rotter's Internal-External Locus ofControl for Reinforcement

Social learning theory of the 1950s and 1960sattempted to integrate older reinforcementtheories with newer cognitive approaches. JulianRotter and his colleagues at Ohio State Univer-sity were trying to predict how reinforcementschanged expectancies in human subjects, butthey repeatedly found that these changes variedsystematically both as a function of the nature ofthe situation and as a characteristic of the personbeing reinforced (Rotter, 1966, 1975). The locusof control construct was initially proposed as away of capturing that systematic variation. As asituational variable, situations that produced thebelief that reinforcement was under outsidecontrol were called external control situations,and those that produced the belief that reinforce-ment was under the individual's own controlwere called internal control situations. As anindividual-differences variable, it referred to thefinding that people exposed to the same set ofcontingencies would vary in their tendency toexpect that reinforcements were contingententirely on their own behavior (internal locus ofcontrol), or that reinforcements were the resultof luck, chance, or fate or were under the controlof powerful others.

Scales were developed to measure individualdifferences in generalized control expectancies,and early factor analyses suggested that therewas one large factor and several small factors.The small factors were initially dropped, leadingto Rotter's (1966) well-known 23-item I-ELocus of Control Scale. Research with this andother scales demonstrated that an internal locusof control was generally associated with positiveoutcomes in health, sports, work, marriage,academic achievement, psychological adjust-ment, and other domains (Lefcourt, 1981, 1992).

Research with the locus of control constructgrew rapidly in the 1960s and 1970s, in partbecause the theory fit with the Zeitgeist in two

320 HAIDT AND RODIN

important ways (Lefcourt, 1992). Within psy-chology, the theory fit perfectly with thecognitive revolution's shift from motivationalinterpretations of deficit-driven behavior tocognitive explanations based on beliefs andattitudes. Social learning theory had alreadygained respectability among academic psycholo-gists, and Rotter's approach preserved thefamiliar language of learning and reinforcement.The introduction of generalized expectancies ofreinforcement therefore provided an easy bridgeinto the new world of cognition, where self-aware individuals processed information (aboutreinforcement) before acting.

The theory also fit in important ways with thepolitical Zeitgeist of the 1960s. Older motiva-tional interpretations of underclass behavior ledto the conclusion that there was somethingwrong with poor people, such as a lack of a workethic or of achievement motivation. The civilrights movement, however, had begun to pointout the devastating effects of oppressive socialconditions and lack of opportunity. The Cole-man report (Coleman et al., 1966) supported thisview, finding few differences between theachievement values or desires of Black andWhite children but apparent differences in theirlocus of control (see also Gurin, Gurin, Lao, &Beattie, 1969). The locus of control constructwas thus well suited for sociological and socialpolicy discussion. Unresponsive social, politi-cal, and economic structures taught BlackAmericans to have an external locus of control,whereas those same structures, being moreresponsive to White Americans, taught them tohave an internal locus of control. There was anobvious relationship between locus of controland sociological constructs such as alienationand empowerment.

Seligman 's Learned Helplessness

Dissatisfaction with traditional learning theo-ries grew throughout the 1960s. Two importantchallenges came out of Richard Solomon'slaboratory at the University of Pennsylvania,where it was demonstrated that animals extractinformation from their environment about con-tingency and controllability. In Pavlovian condi-tioning, Rescorla (1967) demonstrated thatanimals are not affected by the mere contiguityof a conditioned stimulus and an unconditionedstimulus, as had been previously thought. Rather, an

animal's responses are exquisitely sensitive to thedegree of contingency of the conditioned stimuluson the unconditioned stimulus.

In studies of operant conditioning, Overmeierand Leaf (1965) had discovered by accident thatdogs pretrained to associate a light withunavoidable shock while restrained in a harnesswere then unable to learn how to escape shock ina shuttlebox. Dogs that had not been givenunavoidable shock learned the avoidance taskeasily. There was no clear explanation for thiseffect within standard operant models, so twograduate students, Martin Seligman and StevenMaier, came up with their own. Seligman andMaier (1967) argued that animals do not justlearn to make responses to stimuli; rather, theyactually learn the degree to which they cancontrol an outcome. Dogs that have learned thatelectric shock is not controllable make no effortto control it, even in a new situation in which controlis possible. The dogs have learned that they arehelpless, and their helplessness generalizes to newsituations in which it is not appropriate.

In the 1970s, research on learned helplessnessmoved beyond the animal learning lab andlooked increasingly at the effects of uncontrolla-bility on people (Hiroto & Seligman, 1975;Seligman, 1975). The same basic effects werefound, yet it quickly became apparent that anadditional level of complexity was at work andthat the raw facts of controllability and uncontrol-lability were not sufficient to explain humanbehavior. People's interpretations and attribu-tions kept interfering. This realization occurredat the same time that attribution theory wasbecoming the dominant paradigm within socialpsychology, and thus was born the "attributionalreformulation" of the learned helplessnessmodel (Abramson, Seligman, & Teasdale, 1978).

Attribution theorists (Heider, 1958; Jones &Davis, 1965) see people as striving to makesense of their experiences and to assign causesto events. The attributional perspective wasparticularly compatible with the learned helpless-ness paradigm because attributions were seen tobe a form of "cognitive control" (Kelly, 1967).People feel particularly powerless when theycannot understand what is happening aroundthem. The attributional process relieves thisparalyzing confusion and is a prerequisite foreffective action. Abramson et al. (1978) lookedat the attributions people make about bad events,particularly their controllability. They selected


three parameters of causal explanation as thecritical cognitive mediators between events inthe world and human helplessness and depres-sion: (a) locus of causality as either internal tothe self or external, (b) whether the causes of anevent are perceived to be stable over time orunstable, and (c) whether causes are perceivedto be global or specific.

With these three parameters in hand, Abram-son, Seligman, and Teasdale brought learnedhelplessness fully into the age of cognition. Thenew focus of the theory was the mediatingvariable of explanatory style (initially calledattributional style). Explanatory style refers to aperson's habitual way of explaining events inher or his life on the three parameters justdescribed. A person who habitually invokesinternal, stable, and global factors to explainfailures and difficulties is said to have a"depressive explanatory style" and is most atrisk of becoming helpless and depressed in the faceof stressors and uncontrollable circumstances.

Bandura 's Self-Efficacy

Albert Bandura's notion of self-efficacy (1977',1982, 1995, 1997) bears some resemblance toRotter's locus of control and to Seligman'slearned helplessness. All three constructs grewout of a dissatisfaction with traditional learningtheories, and all three theorists pushed for amore cognitively oriented approach in whichbeliefs about control and agency guide behavior.But there are two principal differences that makeself-efficacy theory unique. The first concernsthe kind of belief being assessed. Seligman andRotter are concerned with beliefs about contin-gencies in the world (e.g., "Is that outcomecontingent on my behavior?"), which Banduracalls an outcome expectation. But Bandurapoints out that it is possible to believe that anoutcome is dependent on one's behavior yet notbelieve that one can produce the requiredbehavior. Bandura therefore focuses instead onbeliefs about the self and the self's abilities. Heshifts the psychological focus to what he calls anefficacy expectation, which is the belief that onecan successfully execute the behavior requiredto produce a particular outcome.

The second difference concerns the domainspecificity of the beliefs in question. Banduradoes not deny that people may have generalizedoutcome expectancies (i.e., that the world is or is

not controllable), but he insists that generalizedexpectations are of little use in predictingspecific behaviors. Efficacy expectations, whichare assessed within specific domains, are betterpredictors of behavior. Efficacy beliefs aremultidimensional; even within a broad domainsuch as school or child rearing, most peoplethink they are good at some tasks and not goodat others. Bandura's (1997) "sociocognitive"approach therefore "provides profiles of effi-cacy beliefs across diverse domains of function-ing rather than evading the distinctive patterningof human belief systems by using generalmeasures" (p. 52).

In a typical self-efficacy assessment, partici-pants are asked to rate how confident they arethat they can perform each of a series ofbehaviors, graded for difficulty (e.g., "I couldlook at a spider"; "I could let a spider walk upmy arm"). An important finding from Bandura'sresearch is that people's ratings of self-efficacyin challenging situations are better predictors oftheir subsequent behavior than are their pastrecords of success and failure (Bandura, 1977).It is rare in psychological research that anythingoutpredicts past performance.

But do self-efficacy beliefs cause effectiveperformance, or do they simply reflect people'sability to predict their own behavior? Banduraand his colleagues have provided evidence thatdirect manipulations of self-efficacy beliefsaffect performance. Weinberg, Gould, and Jack-son (1979), for example, gave participantsbogus feedback about performance in a competi-tion of muscular strength, thereby raising orlowering their self-perceptions of physicalefficacy. When tested on a different motor taskrequiring physical stamina, participants withartificially elevated efficacy beliefs displayedgreater physical endurance. Even more impor-tant is the finding that the two groups reacteddifferently to failure in a subsequent competi-tion. Participants whose self-efficacy beliefs hadbeen raised in the first stage of the experimentresponded to failure with even greater physicalefforts, whereas those whose efficacy beliefs hadbeen undermined turned in even weaker perfor-mances in the final stage. When faced withdifficulties, people who doubt their abilitiesquickly give up, whereas people with a strongbelief in their own efficacy will try even harderto rise to a new challenge (Bandura & Cervone,1983).

322 HAIDT AND RODIN

Questioning the Cognitive Focus

This brief and selective review was designedto show how control served as one of the bridgesthat brought psychology to its current stronglycognitive focus. The field began with challengesto behaviorism and psychoanalysis, which haddenied the importance of agency. White pro-posed that behaviorists and psychoanalystsshould accept their own evidence for theexistence of a motivation for competence.Rotter, Seligman, and Bandura continued thechallenge to traditional learning theories, aidedby the growing recognition of the importance ofcognition and information processing in the1960s and 1970s. By 1980, almost all research-ers had adopted a cognitive perspective, focus-ing on people's beliefs and attributions aboutcontrol, causality, and efficacy. Furthermore,several theorists had pointed out difficulties withthe motivational formulation of White, such asthat people do not always want control forcontrol's sake, particularly when they believethat they lack the information to make a gooddecision or when they believe that having controlwill not help them achieve their goals (Bandura,1989; Rodin, Rennert, & Solomon, 1980).

The recognition of the power of consciousbelief was undoubtedly an improvement overearlier dogmatic denials of the role of conscious-ness. But was anything lost in the transition?Should research on control and efficacy limititself to the study of people's individual beliefs?In her review of more than 100 controlconstructs, Skinner (1996) justified the focus on"subjective" control constructs by quotingLanger (1979, p. 306): "Virtually all researchersstudying the importance of control will agreethat the effects of objectively losing or gainingcontrol will only have psychological signifi-cance if the person recognizes (accurately orinaccurately) the gain or loss."

Yet, we have found that not all researchersagree on the primacy of conscious recognitionand belief. Animal researchers generally believethat conscious recognition is not necessary.Psychoanalysts do too, and they are joined bysome social and cognitive psychologists inter-ested in automatic or unconscious controlprocesses (Wegner & Bargh, 1998). And sociolo-gists since Durkheim appear to be perpetuallyfrustrated by psychologists' difficulty in seeingemergent social structures and forces, about

which individuals are often unaware. Con-sciously accessible beliefs about control andefficacy are clearly very important, but there ismore to the story. To see what more there is, we nowturn to the sociological perspective on control.

Control as a Mind-Society Bridge

Sociological theories have traditionally pos-ited the importance of supra-individual forcesand structures in explaining human behavior(e.g., Durkheim, 1897/1951). If one wants toknow why and how control is good for people,one must go beyond the study of individuals andtheir beliefs. One must also investigate a varietyof structural properties of social systems aboutwhich individuals may even be aware. However,even though one level of explanation cannot bereduced to the other, the two can be joined andarticulated into a broader and more powerfultheory. Many sociologists strive for such "micro-macro integration" (recently described as "con-silience" by Wilson, 1998), and a number ofbridges have been built between social struc-tures and individual outcomes, via control-related constructs. We explore four of thosebridges here (for major reviews of self-efficacywithin sociology, see Gecas, 1989; Gecas &Schwalbe, 1983; and Mirowsky & Ross, 1989).

Alienation and Anomie

Alienation is one of the oldest concepts insociology, rooted in the writings of Marx(1844/1963). In Marx's view of human nature,the self is created by participation in "praxis,"or habitual practices, particularly work activity.Efficacious and self-directed work activitycreates efficacious and self-directed individuals.A large part of Marx's critique of capitalism wasthat industrial production methods take controlaway from the individual and turn the individualinto a part of the larger machine. The psychologicaleffect of such work environments is alienation, astate marked by feelings of powerlessness andmisery, in which work is dull and unfulfilling. Marxthought there was a natural human need to createand produce and that this need goes unmet whenproducers are separated from products and whenlabor is performed only to obtain money.

Alienation can be a state of an individual or itcan be seen as an emergent property of socialsystems. Empirical studies of alienation have


concentrated on its manifestation in individuals,where alienation has been conceptualized in sixdifferent ways, according to Seeman (1959,1975): (a) powerlessness (the sense of lowcontrol vs. mastery over events), (b) meaningless-ness (the sense of incomprehensibility vs.understanding of personal and social affairs), (c)normlessness (high expectancies for sociallyunapproved means vs. conventional means forthe achievement of given goals [i.e., deviancepays]), (d) cultural estrangement (the individu-al's rejection of commonly held values in thesociety [or subsector] vs. commitment to thegoing group standards), (e) self-estrangement(the individual's engagement in activities thatare not intrinsically rewarding vs. involvementin a task or activity for its own sake), and (f)social isolation (the sense of exclusion orrejection vs. social acceptance).

The first form of alienation as powerlessnessis the most direct descendent of Marx's writings,and it was incorporated into Rotter's thinkingabout locus of control. The third form ofalienation as normlessness—and, to some ex-tent, the second form as meaninglessness—covers what Durkheim called anomie, a state ofderegulation characteristic of modern societiesin which traditional norms and standards havebeen undermined without being replaced by newones. Durkheim believed that social constraint ishealthy, because it gives people standards andstructure within which their ambitions can bemet. When constraints are relaxed and norms areunclear, as during times of rapid social change,human ambitions and demands expand like agas released into a vacuum. Nobody can besatisfied or happy, because nobody can achievehis or her ambitions. In Durkheim's sociologicalperspective, individual happiness requires social-structural supports: the existence of clear socialnorms and a widely accepted and respectedsocial order.

Reading Durkheim while thinking aboutcontrol and efficacy leads to the followingspeculation: Normlessness and chaos breedmisery and suicide in part because they arestructural impediments to the satisfaction ofeffectance motivation. Thoits (1983) has arguedalong similar lines that when people are not wellintegrated into cohesive groups, they are lesslikely to have a sense of certainty, purpose, andmeaning in life. Thorlindsson and Bjarnason(1998) recently found empirical support for this

Durkheimian position on the micro level,demonstrating that family integration counter-acted anomie and reduced suicidal ideation in alarge sample of Icelandic youth.

Work, Class, and Control

Marx and Durkheim are sometimes criticizedfor denying the importance of human agencyand for stressing the determinism of socialforces. But they both at least acknowledged inhuman nature a need for some sort of control,competence, or achievement. They are also inaccord that social structures such as social classinteract with these needs. Durkheim (1897/1951), for example, explained why an internallocus of control, or sense of mastery, shouldcorrelate with socioeconomic status (SES):

Wealth, . . . by the power it bestows, deceives us intobelieving that we depend on ourselves only. Reducingthe resistance we encounter from objects, it suggeststhe possibility of unlimited success against them. Theless limited one feels, the more intolerable all limitationappears, (p. 254)

Marx wrote even more extensively about thepsychological effects of social class. Alienationwas supposed to result from the disempoweringconditions of labor, yet labor conditions varygreatly by social class. Members of the domi-nant class retain power and personal control(e.g., through ownership of the means ofproduction) and so should not become alienatedin the course of their daily work. This is averifiable hypothesis, and a number of studieshave tested it.

The most comprehensive study of the psycho-logical effects of work conditions was done byMelvin Kohn and his colleagues (collected inKohn & Schooler, 1983). They conductedcross-sectional studies on a representative sam-ple of 3,100 employed American men in 1964,one quarter of whom were interviewed again 10years later to obtain longitudinal data. Theymeasured SES as well as a variety of occupa-tional conditions and psychological orienta-tions, including five of the kinds of alienationconceptualized by Seeman (1959). They foundthat three related occupational conditions werethe key to understanding a great variety ofpsychological outcomes, including alienation:(a) closeness of supervision in the workplace,(b) substantive complexity of work, and (c)routinization of work. They combined these

324 HAIDT AND RODIN

three conditions into a higher order constructlabeled occupational self-direction. Men whowere closely supervised in jobs with lowcomplexity and high routinization showed thehighest degree of three kinds of alienation: power-lessness, normlessness, and self-estrangement.

One part of the Marxist thesis was thereforesupported: Conditions of work, particularly alack of control and a lack of challenge, arerelated to alienation. However, another part ofthe thesis was not supported. Kohn and Schoolerdistinguished between two kinds of control:control over the product of one's labor (opera-tionalized as either owning one's own businessor having high rank in a supervisory hierarchy)and control over the process of one's labor (i.e.,occupational self-direction). In Marx's portrayalof early industrial capitalism, these two kinds ofcontrol were inseparable. However, in Kohn andSchooler's data they were quite separable, and,when examined separately, it was control overprocess that mattered. Ownership of the meansof production and high rank in a supervisoryhierarchy bore only a weak negative relationshipto alienation, a relationship that was mediatedby occupational self-direction. In other words,supervisors feel less alienated than lower levelemployees, but that is because, on average,supervisors enjoy greater occupational self-direction (Mirowsky & Ross, 1989, obtainedsimilar findings).

The implication of this finding is that wagelabor and low rank in a capitalist system neednot lead to alienation, if workers can be givensome degree of occupational self-direction. Infact, Kohn and Schooler argued that it isprimarily through its correlation with occupa-tional self-direction that social class exerts itseffects on personality. They concluded that

occupational self-direction has the most potent andmost widespread psychological effects of all theoccupational conditions we have examined. In terms ofthese effects, the central fact of occupational life todayis not ownership of the means of production, nor is itstatus, income, or interpersonal relationships on thejob. Instead, it is the opportunity to use initiative,thought, and independent judgment in one's work—todirect one's own occupational activities. (Kohn &Schooler, 1983, p. 84)

Thus, control-related variables again provide thebridge from social-structural properties (class,status, and working conditions) to individualoutcomes. Control appears to be part of the

explanation for the damaging effects of lowsocial class, whether the outcome variable isalienation, as just described, or physical health,as in recent findings that health and SES arecorrelated at all levels of SES (Adler et al.,1994; Marmot et al., 1991). The health-damaging effects of low status in a hierarchyhave even been extended into the animalkingdom by Sapolsky (1992), who found thatlow-ranking male baboons have higher basalcortisol levels but that the relationship betweenrank and cortisol level fluctuates with thestability of the social order.

Social Capital

A more recent construct for the sociologicalstudy of control and efficacy comes fromColeman's (1988, 1990) work on social capital.Coleman starts with the motivational theory ofrational action, borrowed from economics, inwhich individual agents strive to maximizeutility. He then gives these rational actors astrongly social nature, acknowledging thatpeople are governed by norms, rules, andobligations. Next, he asks what these social-rational agents need to maximize utility. He liststhree kinds of capital about which mosteconomists would agree: (a) financial capital(money), (b) physical capital (e.g., equipment,tools, and clothing), and (c) human capital(education, knowledge, and training). Colemanthen proposes a fourth kind of capital thateconomists have generally missed: social capi-tal, which inheres not in any person but in thestructure of relations among actors. There are atleast three kinds of social capital, but all of themhave the following two properties: "They allconsist of some aspect of social structures, andthey facilitate certain actions of actors—whether persons or corporate actors—within thestructure" (Coleman, 1988, p. S98).

The most important form of social capitalconsists of a dense network of obligations,expectations, and trustworthiness. Coleman givesthe example of the wholesale diamond market inNew York, where the high degree of trust andmutual obligation among the members allowsthem to make transactions without fear of theftor deceit by one another and therefore withoutexpensive monitoring and litigation procedures.If an alternate wholesale market were to befounded in which members lacked this interper-


sonal trust, its costs would be higher and itwould fail. Strong social ties and interpersonaltrust are a form of social capital, a kind ofinfrastructure that facilitates the goals of themerchants and the efficiency of the market.

The second form of social capital consists ofinformation channels. A system in which peoplecan rely on friends or colleagues to pass on tothem essential information will function moreeffectively than a system in which eachindividual must gather all of her or his owninformation from primary sources. The thirdform of social capital is norms and effectivesanctions. When widely held norms support thegoals of a social system, the system will workmore effectively.

The concept of social capital has provenextremely useful for the study of complex socialsystems. Social capital is particularly importantfor the study of control and efficacy because itprovides a way of thinking about efficacioussystems, as well as about the mechanismsthrough which efficacious systems create effica-cious individuals. A system with a high level ofsocial capital will, ceteris paribus, outperform asystem with lower social capital, and it will doso in part because it makes available to itsmembers a set of resources that enables them toreach their individual and collective goals. Morebroadly, it can be said that a fully capitalizedsystem, including sufficient levels of all fourforms of capital, provides individuals withgreater affordances, or opportunity structures,than a system that is undercapitalized.

Collective Efficacy

Bandura's theory of self-efficacy was devel-oped for individuals, but he has extended it tothe operation of groups as well. Bandura (1997,p. 477) denned collective efficacy as "a group'sshared belief in its conjoint capabilities toorganize and execute the courses of actionrequired to produce given levels of attain-ments." When the members of a group workindependently toward a common goal, such ason a gymnastics team, the collective efficacy ofthe group is essentially the sum of the efficaciesof the individual members. But when the groupis highly interdependent, as on a basketball teamor in many organizational settings, perceivedcollective self-efficacy becomes an emergentgroup-level attribute that is more than the sum

of its individual parts. As with individualself-efficacy, collective efficacy is multidimen-sional, and it grows primarily out of priormastery and success experiences. Also as withindividual self-efficacy, collective efficacy canbe experimentally enhanced with the samebeneficial effects on outcomes (Prussia &Kinicki, 1996).

Collective efficacy is particularly useful forunderstanding political and social change. Manygroups around the world are discriminatedagainst or excluded from power and prosperity,but it is primarily those that have a highcollective level of political efficacy that under-take vigorous and sustained action towardpolitical change (Bandura, 1997, chap. 11).Political efficacy, in turn, is partially a productof structural factors such as the material andsocial capital of the group and the openness orresponsiveness of the political system to thatgroup (Zimmerman & Rappaport, 1988). Butpolitical efficacy, and other forms of collectiveefficacy, can then reshape those structuralfactors in ways that enhance the group'ssubsequent political efficacy. Such a dramaticreciprocal change clearly happened in theUnited States with African American politicalefficacy in the 1960s.

In summary, sociological theories view indi-vidual beliefs within the broader context ofcomplex social systems. Societies and institu-tions with high social capital and low anomie, inwhich workers are given a high level ofoccupational self-direction, will produce en-gaged individuals with a healthy sense ofcontrol and a strong sense of collective efficacy.Efficacious individuals and groups will simulta-neously work to change the systems withinwhich they live. Control thus provides a"mind-society" bridge, revealing how social-structural factors affect minds, and vice versa.The sociological perspective is fully compatiblewith the individual-centered view prevalent inpsychology, but it suggests very differentavenues for intervention. Whereas psychologistsgenerally favor education and therapy aimed atchanging individual beliefs, sociologists point tothe alternate route of changing system-levelvariables (e.g., changing tax laws or zoningpolicies to strengthen families and neighbor-hoods, or improving the responsiveness ofauthorities and institutions to encourage moreactive participation by individuals and groups).

326 HAIDT AND RODIN

Control as a Mind-Body Bridge

In addition to bridging the psychological andsociological levels of analysis, control is alsowell suited as a bridge between the psychologi-cal and psysiological levels of analysis. Controland efficacy can help explain the mind-bodymechanisms by which social events can cause orprevent disease. Several of these bridges areoutlined in this section (for full reviews, seeAdler & Mathews, 1994; Cohen, 1990; O'Leary,1985; Peterson, Maier, & Seligman, 1993;Rodin, 1986a, 1986b; Taylor, Repetti, & See-man, 1997).

Control and Stress

Control beliefs modulate the physiologicaland psychological impact of various stress-provoking stimuli, even when the objectiveintensity of the stressor is unchanged. In aclassic study conducted by Glass and Singer(1972), two groups of participants were exposedto loud bursts of noise. Participants in one groupwere told they could terminate the noise bypressing a button, but they were asked not topress the button unless it was absolutelynecessary. None of these participants pressedthe button, yet the belief that they had someform of control made the noise less aversive andless likely to induce helplessness on subsequenttasks requiring persistence. An enormous num-ber of laboratory experiments and field studiesconfirm the tight relationship between controland stress. For example, Baum and his associ-ates have found that perceived loss of control isone of the most important mediators of reactionsto disasters such as flooding, exposure to toxicwaste, and the near meltdown of the Three MileIsland nuclear power plant (Baum, Cohen, &Hall, 1993; Baum, Fleming, Israel, & O'Keeffe,1992).

Control is related in complex ways to coping,but under most circumstances having the abilityand resources to engage in problem-focusedcoping reduces the physiological and psychologi-cal impact of stressors and daily hassles(Folkman, 1984). Social class is thereforecorrelated with a sense of control and inverselycorrelated with a sense of constraint (House etal., 1994; Lachman & Weaver, 1998), and thisrelationship appears to be an important contribu-tor to the SES-health gradient. People at any

level of SES are healthier than those below them(Marmot et al., 1991), and this correlationappears to be due in part to psychosocialvariables including control (Adler et al., 1994;House et al., 1994; Taylor et al., 1997). Evenamong individuals of low SES, those with astrong sense of perceived control show levels ofhealth characteristic of a higher SES (Lachman& Weaver, 1998).

Many studies have documented the physi-ological effects of control directly, particularlyon the cardiovascular and immune systems(Bandura, 1991,1997). Lack of control has beenassociated with elevated catecholamine levels(e.g., adrenaline) in animal and human studies(Frankenhaeuser, 1983; Weiss, Stone, & Harrell,1970). High levels of catecholamine are in turnassociated with increased blood pressure andheart rate, elevation of blood lipids, andventricular arrhythmia. There also appears to bea strong relation between control and levels ofcirculating corticosteroids (e.g., cortisol) in bothanimals and humans (Breier et al., 1987; Meier,Ryan, Barksdale, & Kalin, 1988). Corticoste-roids regulate the metabolism of cholesterol andother lipids involved in heart disease and play arole in regulating electrolyte balance and bloodpressure. These endocrine effects may accountfor much of the reported relation between heartdisease and control-related variables (Krantz,Glass, Contrada, & Miller, 1981).

Animal and human studies have also shown adirect effect of uncontrollable stress on theimmune system. Laudenslager, Ryan, Drugan,Hyson, and Maier (1983) gave rats eitherescapable shock or an identical amount ofinescapable shock and found that the rats thatlacked control showed suppressed lymphocyteproliferation and mitogen response. Seligmanand Visintainer (1985) injected live tumor cellsinto rats that received escapable shock, inescap-able shock, or no shock and found that ratsexposed to inescapable shock were more likelyto develop tumors and die than were rats in theother two conditions. In one of the few studies touse a direct manipulation of controllability tostudy human immune response, Sieber et al.(1992) found that exposure to controllable loudnoise produced no measurable effects on im-mune response, whereas yoked participants whowere exposed to an equal amount of uncontrol-lable noise showed a decrease in cytotoxicity ofnatural killer cells that was measurable up to


72 hr later. Sieber et al. also found that the effectof uncontrollability on natural killer cell toxicitywas greatest for participants who scored high ona desire for control scale (Burger & Cooper,1979), consistent with the idea that theseparticipants should be most strongly affected bythe absence of control.

However, the relationship between controland stress is complex (Averill, 1973; Burger,1989; Folkman, 1984; Thompson, Cheek, &Graham, 1988). Rodin et al. (1980) discussedcases in which increasing control can be stressinducing, such as giving a patient the responsibil-ity to make a medical decision that she or hedoes not feel qualified to make. Wright (1998)has demonstrated that a high sense of efficacyoften leads to greater cardiovascular reactivityin response to challenges, because people withlow self-efficacy, who believe a task is impos-sible, may simply not get aroused to meet thechallenge. Yet, despite these important rever-sals, it is generally the case that personal controlreduces the stressfulness of environmentalthreats and challenges.

Control and Positive Affect

Control clearly serves a stress-bufferingfunction, dampening negative affective andendocrine responses to threats and challenges.But Cohen (1990) and others think that controlalso has direct positive health consequences, inpart through its ability to produce positiveaffect. Davidson may have identified the braincircuits involved with his discovery of twodistinct motivational systems (Davidson, 1994;Davidson, Ekman, Saron, Senulis, & Friesen,1990; Sutton & Davidson, 1997). The approachsystem facilitates appetitive behavior and gener-ates certain types of positive affect that areapproach related, such as enthusiasm and pride.The approach system involves limbic andcortical structures that appear to have a conver-gence zone in the left prefrontal cortex. Aseparate withdrawal system appears to have aconvergence zone in the right prefrontal cortex,functioning to facilitate withdrawal from sourcesof aversive stimulation and to generate negativeaffect, including fear.

The approach system generates two distinctforms of positive affect. The first is labeledpre-goal-attainment positive affect, which Dav-idson (personal communication, April 1995) has

described as "the positive affect that arises asyou are progressively moving towards a desiredgoal." The second kind of positive affect iscalled post-goal-attainment positive affect, whichaccording to Davidson, arises once one hasachieved something one wants. This latter typeof affect "may be phenomenologically experi-enced as contentment and is expected to occurwhen the prefrontal cortex goes off-line after adesired goal has been achieved" (Davidson,1994, p. 743).

In other words, Davidson has identified areward system that might be the neural basis ofeffectance motivation. It is a reward systemtriggered not by success itself but by progress ingoal management and successive increases inmastery. The activation of this goal-pursuingsystem leads to increased engagement with theenvironment, including positive affect, pleasurefrom interaction, and increased persistence inthe face of obstacles (Davidson, 1994). David-son has found that activation of the approachsystem dampens activation of the amygdala,which is a principal trigger of the hypothalamic-pituitary-adrenal pathway, by which threatappraisals in the brain trigger rapid cortisolrelease by the adrenal glands. This findingsuggests a possible route by which the approachsystem may have health-protective effects.People who have as a trait a generally higheractivation level in their left frontal cortices(approach) rather than their right cortices(withdrawal) have been found to have greaterimmune competence, both at rest and inresponse to stressors (Kang et al., 1991). Suchfindings may partially explain why approach-oriented optimists stay healthier and live longerthan more fearful pessimists (Peterson et al.,1993).

Control and Health Behavior

Individuals with higher levels of perceivedcontrol and efficacy in health-related domainstake greater responsibility for meeting theirhealth needs, including seeking out informationand complying with medical regimens involvinglifestyle changes (DiMatteo, 1994; O'Leary,1985; Seeman & Seeman, 1983). When told toquit smoking, for example, people are less likelyto succeed if they believe either that they willnot be able to quit (i.e., low self-efficacy) or thatquitting will not affect their health outcome (i.e.,

328 HAIDT AND RODIN

external locus of control for health). People whohave a strong belief in their own self-regulatoryefficacy, in contrast, are more likely to undertakea program of personal habit change, more likelyto maintain the new habit over time, and moreresistant to relapse temptations (Carey & Carey,1993; DiClemente, Fairhurst, & Piotrowski,1995; see Bandura, 1997, chap. 7, for a review).

But there is a more subtle component to thecontrol-behavior connection: Control increasesenergy and vigor, perhaps via the mechanismsdescribed by Davidson (1994). Langer andRodin's (1976) nursing home intervention, forexample, produced its largest effect on the vigorof the residents. Conversely, Seligman (1975)described the passivity and lethargy that arehallmarks of learned helplessness. If feelings ofcontrol are often energizing, then they maymake any behavioral task easier, thereby encour-aging problem-focused coping strategies andproactive health-promoting behavior.

In summary, control and efficacy providebridges from the psychological world of mean-ings and beliefs to the physiological world ofstress hormones and immune suppression. Whencombined with bridges built "up" to thesociological level, the result is a set ofinterdisciplinary links by which one can begin totell a complete scientific story (Kahn, 1993;Wilson, 1998). One can explain how the loss oforder and predictability in a society (e.g.,modern Russia) might cause an increase inmorbidity and mortality rates. To facilitate theinterdisciplinary use of control and efficacy, wenow present an organizing taxonomy.

A Taxonomy of Control Constructs

This report has covered three perspectives oncontrol: the motivational perspective fromwhich the field began, the cognitive perspectiveto which it moved, and the systemic perspectivetaken by sociologists. Each of these threeperspectives looks at a different kind of causalexplanation. The motivational perspective looksat intrinsic motivations that drive behavior, suchas effectance. The cognitive perspective looks atbelief structures that mediate action, such asself-efficacy beliefs. The systemic perspectivelooks at structural properties of social systemsthat foster or inhibit individual or groupcompetence and success, such as social capitalor social disorganization.

By labeling these approaches as "perspec-

tives," we mean to stress that each one can beused to look at a social or developmental issue,and that each perspective will yield its own setof insights and prescriptions for intervention.No perspective can be reduced to the terms ofthe others, and no perspective contradicts thevalidity of the others.

Within each perspective, there are multipleconstructs that can be used in theory construc-tion. We describe six constructs that we think areuseful and versatile, but this list is intended to besuggestive, not exhaustive (for an exhaustive listof control belief constructs, see Skinner, 1996).The six constructs should be thought of as a"toolbox" of control constructs that work welltogether. For each construct, we discuss measure-ment issues and suggest tasks for which theconstruct is particularly well suited.

The Motivational Perspective

The importance of control was first discov-ered in research with dogs, monkeys, and youngchildren, including prelinguistic infants (re-viewed in White, 1959). A great strength of themotivational perspective is that it works forcreatures lacking language and the complexbelief structures that language facilitates. Themotivational perspective posits an innate orintrinsic need to achieve a particular kind ofrelationship with the environment. This relation-ship is, at a minimum, one of control orcausality, and in its more elaborate forms it islabeled as competence, mastery, or achievement.

Later motivation theorists focused on humanbeings, but their theories work as well forinfants as for adults. Deci and Ryan (1985)posited an innate need for self-determination.Brim (1992) posited a universal human drive forgrowth and mastery, which makes peopleactively seek out challenges. Heckhausen andSchulz (1995) posited a universal human desireto exert primary control over the environment,which is defined as the production of "behavior-event contingencies." Common to these theo-rists is a notion of intrinsic reward. Mastery,competence, and control are (usually1) pleasur-able, and people will take on challenges, master

1 The existence of cases in which mastery, competence, orcontrol is not pleasurable does not argue against theories ofintrinsic motivation any more than the existence of cases inwhich food is not pleasurable argues against the existence ofan intrinsic motivation to eat.


new skills, or make things happen because itfeels good to do so, because people are built toderive pleasure from approaching their goals(Davidson, 1994; White, 1959).

The motivational perspective is an optimisticperspective on human nature. It assumes that "ifsocial contexts can manage to set up opportuni-ties, people will actively strive to become morecompetent" (Skinner, 1995, p. 16). The motiva-tional perspective lends itself to metaphors forgrowth. James Connell (personal communica-tion, May 19, 1995) speaks of "thriving" versus"surviving" in school and sees the difference asdue in part to whether the school environmentprovides the right "nutrients" for psychologicalneeds, including competence.

A variety of motivational constructs havebeen posited. Here we select just two forelaboration.

Construct 1: Effectance Motivation

Description. As described by White (1959)and Harter (1978), effectance is an intrinsicmotivation or striving for competence or mas-tery. The satisfaction of effectance leads to apleasurable feeling of efficacy. Several research-ers have proposed that White's effectance can bedivided for analytical purposes into two distinctneeds: a need for competence or mastery and aneed for autonomy or self-determination (Deci,1995; Deci & Ryan, 1985; Skinner & Connell,1986). We focus in this report on effectance as aneed for competence, because this motivation ismore closely related to subsequent constructsand because it appears to be more culturallyuniversal than the need for autonomy, whichmay be a particularly Western need (Markus &Kitayama, 1991; Shweder, Balle-Jensen, &Goldstein, 1995).

Measurement. It is not clear at presentwhether there are global, stable differencesbetween individuals in their levels of effectancemotivation. On the one hand, effectance can beseen as a basic fact about people, as somethingthat "comes with the animal" (O. G. Brim,personal communication, March 20, 1995). Onthe other hand, Harter's (1978) analysis suggeststhat effectance might be strengthened or weak-ened in long-lasting ways by childhood experi-ences of mastery and failure and of support anddiscouragement. Davidson's (1994) suggestion

of plasticity in the neural development ofapproach and withdrawal systems also indicatesthe possibility of individual differences. Andthere have been several scales and behavioraltasks designed to measure individual differencesin effectance motivation in children (Harter &Zigler, 1974; Pearlman, 1984). However, thesescales have not been widely used, and there doesnot appear to be any well-validated method atpresent for measuring individual differences ingeneralized effectance motivation.

Uses. Until it is determined whether or noteffectance varies in a traitlike way acrossindividuals, this construct is best used as ageneral assumption about human needs anddesires (i.e., humans need food, water, compe-tence, and a few other things to thrive). It isespecially useful for thinking about the "goodlife," or the conditions of human satisfaction. AsBrim (1992) demonstrated, happiness comes notfrom material success but from living life at alevel of "just manageable difficulty" in whicheffectance is repeatedly engaged by properlygeared challenges. Effectance is also useful forthinking about life span transitions, whichalways involve new sets of challenges andperformance expectations. In thinking about anytransition, one can ask the following questions:Are opportunities for competence expanding orshrinking? Are opportunities closely matched toabilities and resources? Are challenges over-whelming or insufficient? The match betweeneffectance needs and environmental affordancesis often the key to understanding successfuldevelopment (Eccles, Midgley, Wigfield, &Buchanan, 1993; Elder, 1998).

Construct 2: Engagement

Description. Engagement is a quality ofintrinsically motivated behavior in a particularcontext. It is difficult to measure motivationssuch as effectance directly, yet a state ofengagement can be inferred from a person'sbehavior (e.g., working persistently or doingmore than the minimum), emotion (e.g., self-reports and others' ratings of interest vs.boredom), and task orientation (e.g., confidenceand willingness to take risks with more difficultchallenges; J. P. Connell, personal communica-tion, May 19, 1995). Engagement is similar toCzikszentmihalyi's (1991) notion of flow. En-

330 HAIDT AND RODIN

gagement is a visible sign of effectancemotivation at work (although engagement canresult from the operation of other intrinsicmotivations as well; see Connell, 1990).

Measurement. Engagement should not bethought of as a trait or as a property of personsdivorced from contexts. Engagement refers to away of being in a given context, and it can beassessed only in a given context. The RochesterHuman Motivation Research Group has devel-oped a series of instruments for measuringengagement in the classroom context known asthe Research Assessment Package for Schools(RAPS; Wellborn & Connell, 1987). The RAPSasks parents and teachers to rate various aspectsof a student's behavior and apparent emotions,and it asks students to report their own behavior,emotions, and thoughts.

Uses. Engagement has been demonstratedto be a critical mediating variable in a variety ofperformance contexts (Connell, Spencer, &Aber, 1994). People who work hard at a taskthey are committed to and interested in achievebetter results than people who are disaffected orunengaged. Engagement and disaffection cantherefore enrich process models whenever thecauses and correlates of high performance arebeing studied. (Note that engagement is notconfounded with outcome measures, such asgrades.) Engagement is particularly useful as amediating variable in intervention studies, be-cause one does not have to wait until grades areassigned at the end of the semester (for example)to determine whether the intervention is havingan effect. If the intervention increases studentengagement in academic activities, it is likely tobe effective.

The Cognitive Perspective

The motivational perspective makes it easy tosee the continuity between humans and animals.Indeed, White (1959) speculated that there maybe an ancient evolved motivation for effectiveinteraction with the environment. Yet, humanbeings have also evolved a variety of cognitiveabilities that intervene between motivation andbehavior, creating a more elaborate set ofmechanisms for the operation of control andefficacy. The cognitive perspective looks at theways in which beliefs about control, contingen-cies, and capacities affect human behavior,mood, and cognition.

Many distinctions among beliefs have beenproposed. The highest-level distinction, madeby many authors, is between beliefs about theself (e.g., Bandura's, 1977, self-efficacy beliefs,Gurin and Brim's, 1984, personal efficacyjudgments, and Skinner's, 1995, capacity be-liefs) and beliefs about contingencies in theexternal world (e.g., Rotter's locus of control,Seligman's learned helplessness, Bandura's out-come expectancies, Gurin and Brim's system-responsiveness judgments, and Skinner's strat-egy and control beliefs).

A second distinction is commonly madebetween domain-specific beliefs and general-ized beliefs. These two distinctions can becrossed to create a 2 X 2 table (see Table 1).Most of the research on control and efficacy fallsinto two of the four cells.

Construct 3: Self-Efficacy Beliefs

Description. Bandura (1997, p. 3) offered aclear definition: "Perceived self-efficacy refersto beliefs in one's capabilities to organize andexecute the courses of action required toproduce given attainments." Self-efficacy be-liefs are the product of learning (direct orvicarious) within particular domains of experi-ence. Self-efficacy beliefs strongly influence theeffort a person will expend in meeting achallenge, and they are among the best predic-tors of success in many contexts. Self-efficacy isclose to what Skinner (1996) described as the"prototypical" control construct, the one mostpsychologists think of when thinking aboutcontrol.

Measurement. Self-efficacy beliefs are as-sessed by a questionnaire or interview in whichparticipants are asked to rate their ability to

Table 1Four Types of Belief

Specificity

Domain specific

Generalized

Beliefs aboutself-capacities

Self-efficacybeliefs (Ban-dura)

Grandiosity? Self-esteem?

Beliefs aboutenvironmentalcontingencies

Domain-specificlocus of control

Generalized con-trol beliefs(Rotter andSeligman)


perform a variety of well-specified tasks rangingin difficulty. Well-crafted and well-validatedself-efficacy instruments are widely available inthe published work of Bandura (1997) and hiscolleagues, and they were recently used inlarge-scale studies of schools, families, andneighborhoods by Elliott et al. (in press);Furstenberg, Cook, Eccles, Elder, and Sameroff(1998); and King and Elder (1998).

Uses. Self-efficacy theory was designed toexplain, predict, and improve performance inspecific domains. It is exceptionally well suitedto performance domains in which a personperceives a challenge and is motivated tosucceed, yet is not certain about her or hisabilities (e.g., overcoming phobias and drugaddictions, sticking to diets or medical regi-mens, or doing well in school or sports).Self-efficacy measures are also useful in studiesthat measure changes over time in multiple lifedomains (e.g., Gurin & Brim, 1984; Lachman,1986; McAvay, Seeman, & Rodin, 1996).Because self-efficacy beliefs influence futurebehavior, such beliefs can be an effective entrypoint for interventions aimed at specific prob-lems or behaviors (Rodin, 1989).

Construct 4: Generalized Control Beliefs

Description. Control beliefs vary by do-main (Bandura, 1997; Skinner, 1995). Nonethe-less, some people are more likely than others tobecome helpless, to perceive an external locusof control, or to make external causal attribu-tions across multiple domains (Lefcourt, 1981;Peterson et al., 1993). Theories of control maytherefore be analogous to theories of intelli-gence that include a general factor (g), to whichspecific factors (S) are added. Seligman andRotter have proven the existence and impor-tance of a general factor, which can be reliablymeasured.

Generalized control beliefs can therefore bethought of as a personality trait, or as a stable"attributional style." Some people tend towardinternality, and others tend toward externality.Alternatively, some people have a "mastery-oriented" response style, whereas others have a"helpless" response style (Diener & Dweck,1978; Dweck & Leggett, 1988). This trait makessome contribution to behavior and outcomesacross a broad spectrum of domains, and itsinfluence is likely to be strongest in domains in

which a person has the least prior experience(i.e., those in which domain-specific beliefshave not yet been formed). It is important tonote, however, that both Seligman and Rotterare learning theorists. They both view this"trait" as the product of experience rather thanas an inborn cognitive temperament. Peoplewho are exposed to frequent noncontingencyand uncontrollability learn to expect noncontin-gency and uncontrollability, and these expecta-tions generalize beyond the domains in whichthey were formed.

Measurement. This is the only one of the sixconstructs that can be thought of as a stablepersonality trait, and it is therefore relativelyeasy to measure. General scales can be used inmany cases without customization. Seligmanand Peterson's Attributional Style Questionnaire(Peterson et al., 1982) and Rotter's (1966) I-ELocus of Control Scale and its variants are themost widely used scales. Pearlin's MasteryScale (Pearlin & Schooler, 1978) is a short andsimple scale that is also widely used. Seligmanand Peterson's approach to measurement is therichest, in that it assesses three dimensions ofattribution and differentiates between attribu-tions about positive and negative events. How-ever, the Attributional Style Questionnaireincorporates scenarios and questions that aremost meaningful for middle-class young adults,and it may need to be modified for use in otherpopulations.

Uses. Generalized control beliefs are lessuseful than specific efficacy beliefs in predictingspecific behaviors (e.g., who will quit smoking).But they are useful for predicting successfuladaptation in broad domains of activity, where itis generally found that people with an optimisticexplanatory style or an internal locus of controldo better than others (Scheier & Carver, 1985;Seligman, 1991). There are, however, a numberof exceptions to this rule, and there appear to besituations in which externality and pessimismare adaptive, such as, perhaps, law school(Satterfield, Monahan, & Seligman, 1997).Generalized control beliefs are particularlyuseful in new situations and contexts, wheredomain-specific self-knowledge may not beavailable. Also, because of the tight linkbetween helplessness and depression (Seligman,1975), generalized control beliefs are importantin studies concerned with mental health.

332 HAIDT AND RODIN

The Systemic Perspective

Motivations and beliefs do not arise in asocial vacuum, and researchers using thecognitive and motivational perspectives oftentry to spell out the relationships between theindividual and broader social systems, such asschools, neighborhoods, workplaces, and fami-lies (see especially the work of Skinner, 1995,and Connell, 1990, and the European traditionof action theory [e.g., Frese & Sabini, 1985]).These systems can be thought about in twoways. First, one can focus on the individualswithin a system and ask how social-structuralproperties, as independent variables, support orinhibit individual success and failure as depen-dent variables. This view gives us our firstsystemic construct, systemic supportiveness.Alternatively, we can leave the individual levelentirely and take the social system as the unit ofanalysis. We can think about efficacy andcontrol as properties of healthy social systems,and we can ask what factors lead to systemicefficacy as a dependent variable.

Construct 5: Systemic Supportiveness

Description. A supportive system is a sys-tem that provides the affordances that anindividual needs to take on and master newchallenges (i.e., resources, structures, and oppor-tunities that match the needs and abilities of theindividual). A supportive school will have adifferent set of features than a supportive family,but in general a supportive system may becharacterized by the following features: highlypredictable contingencies, clear expectationsand norms for performance and success, clearfeedback about performance, encouragementfor mastery attempts, high social capital (in theform of dense social networks that can be tappedfor information or resources), and sufficientphysical capital (see Coleman, 1988; Fursten-berg et al., 1998; Gurin & Brim, 1984; Sampson,1993; Skinner, 1995). A supportive system canbe described in motivational terms as a systemthat leads to high engagement and allowsindividuals to satisfy their effectance motiva-tion. A supportive system can be described incognitive terms as a system that bolsters self-efficacy beliefs by providing mastery opportuni-ties and that creates internal control beliefs by

making people's outcomes consistently depen-dent on their own actions.

Measurement. Assessment methods can in-volve a variety of objective and self-reportmeasures, aimed either at assessing the person-environment fit or at describing relevant featuresof an environment. Skinner and Connell includemeasures of "perceived structure" and othersystem-level variables in their school assess-ment instruments. Rudolph Moos and hisassociates (Moos, 1987, 1991; Moos & Lemke,1992; Timko & Moos, 1989) have developed avariety of social climate questionnaires aboutwork, family, school, and inpatient environ-ments that pay close attention to the ways inwhich environments support or discouragecontrol and autonomy. And a variety of sociologi-cal scales have been designed to measurealienation and anomie (Seeman, 1991).

Uses. Any study of human performance,development, satisfaction, or health shouldinclude measures of systemic supportiveness.Many psychologists are interested in person-environment interactions, and systemic support-iveness is another name for that interaction,viewed from the perspective of control andefficacy. Systemic supportiveness is particularlyuseful for understanding aggregate differencesacross demographic groups (e.g., ethnic commu-nities and social classes) or institutions (e.g.,why students at one school are doing better thanstudents at another). It is also useful fordesigning interventions to make systems moresupportive, which in many cases will be morepractical and cost-effective than trying to changeindividual behavior directly through educationor therapy.

Construct 6: Systemic Efficacy

Description. In the long run, we would allbe better off if our basic institutions and systemsworked well. But what does it mean for a systemor institution to work well? Much of the answeris likely to revolve around issues of quality andefficiency; for example, an efficacious schoolmust turn out well-educated children at areasonable price per student. But we may beable to borrow some additional ideas from thestudy of control and efficacy at the individuallevel. For example, systemic efficacy may growout of a good fit between a system and itsbroader social, legal, or economic context. Or


perhaps efficacious and inefficacious systemscan be distinguished by their ability to respondto challenge and failure. Ira Harkavy (personalcommunication, March 14, 1995) pointed outthe importance of linking systems (e.g., schoolsystems and health systems) so that they worktogether to meet people's needs and their own.For now we simply want to stress the impor-tance of thinking at the systemic level. Society isnot just a collection of individuals. Schools,businesses, neighborhoods, and governmentagencies continue on as individuals cyclethrough them, and the efficacy of a system is notjust the sum of the efficacy beliefs of the peoplepassing through.

Measurement. Measurement might focus onthe quantity and quality of a system's output,such as the academic achievement of a school'sstudents or the sales growth and customersatisfaction ratings of a business. The costs ofproducing such outputs should be analyzedrelative to comparable systems. Measurementcould also look for signs of problems andinefficiencies within the system (e.g., employeedissatisfaction or the number of sick days takenor lawsuits filed by employees; R. Rosen,personal communication, February 16,1995).

Uses. Systemic efficacy may be useful forthinking about school reform, corporate restruc-turing, or any other sort of institutional change.What resources and structural features does thesystem need to be efficacious? Systemic efficacymay be particularly useful for thinking aboutsocial policy. How can changes in the legal,economic, and social environment in whichsystems operate help schools, neighborhoods,workplaces, and families do their jobs well?

Conclusion

We believe that the three perspectives de-scribed here are highly compatible and that akind of synergy results from using all three intandem. People are motivated to attain compe-tence and mastery, their motivated behavior isguided by their beliefs about self-efficacy andcontrollability, and the development of thesemotivations and beliefs depends critically on thefit between the individual and the multiplesocial systems he or she participates in, whichcan either support or inhibit efficacious, agenticbehavior.

As psychologists take on increasingly large

real-world problems and lobby to have the nextdecade declared "the decade of behavior," itbecomes increasingly important for us to workat multiple levels of analysis. In particular,public health problems often require working atthe sociological, psychological, and biologicallevels simultaneously. Control and efficacyserve both as examples of how multiple levelscan be bridged and as powerful tools for creatingchange in real-world systems. Combining themotivational, cognitive, and systemic perspec-tives on control can help us to meet thesechallenges and to make these changes.

References

Abramson, L. Y., Seligman, M. E. P., & Teasdale,J. D. (1978). Learned helplessness in humans:Critique and reformulation. Journal of AbnormalPsychology, 87, 49-74.

Adler, N., Boyce, T, Chesney, M., Cohen, S.,Folkman, S., Kahn, R., & Syme, S. L. (1994).Socioeconomic status and health: The challenge ofthe gradient. American Psychologist, 49, 15-24.

Adler, N., & Matthews, K. (1994). Health psychol-ogy: Why do some people get sick and some staywell? Annual Review of Psychology, 45, 229-259.

Averill, J. R. (1973). Personal control over aversivestimuli and its relationship to stress. PsychologicalBulletin, 80, 286-303.

Bandura, A. (1977). Self-efficacy: Toward a unifyingtheory of behavioral change. Psychological Re-view, 84, 191-215.

Bandura, A. (1982). Self-efficacy mechanism in humanagency. American Psychologist, 37, 122-147.

Bandura, A. (1989). Human agency in socialcognitive theory. American Psychologist, 44, 1175-1184.

Bandura, A. (1991). Self-efficacy mechanism inphysiological activation and health promotingbehavior. In J. Madden (Ed.), Neurobiology oflearning, emotion, and affect (pp. 229-269). NewYork: Raven Press.

Bandura, A. (1995). Self-efficacy in changing societ-ies. Cambridge, England: Cambridge UniversityPress.

Bandura, A. (1997). Self efficacy: The exercise ofcontrol. New York: Freeman.

Bandura, A., & Cervone, D. (1983). Self-evaluativeand self-efficacy mechanisms governing the moti-vational effects of goal systems. Journal ofPersonality and Social Psychology, 45, 1017-1028.

Baum, A., Cohen, L., & Hall, M. (1993). Control andintrusive memories as possible determinants ofchronic stress. Psychosomatic Medicine, 55, 274—286.

334 HAIDT AND RODIN

Baum, A., Fleming, I., Israel, A., & O'Keeffe, M. K.(1992). Symptoms of chronic stress following anatural disaster and discovery of a human madehazard. Environment and Behavior, 24, 347-365.

Brehm, J. W. (1966). A theory of psychologicalreactance. New York: Academic Press.

Breier, A., Albus, M., Pickar, D., Zahn, T. P.,Wolkowitz, O. M , & Paul, S. M. (1987). Con-trollable and uncontrollable stress in humans:Alterations in mood and neuroendocrine andpsychophysiological function. American Journalof Psychiatry, 144, 1419-1425.

Brim, G. (1992). Ambition. New York: Basic Books.Burger, J. M. (1989). Negative reactions to increases

in perceived personal control. Journal of Personal-ity and Social Psychology, 56, 246-256.

Burger, J. M., & Cooper, H. M. (1979). Thedesirability of control. Motivation and Emotion, 3,381-393.

Butler, R. A. (1958). Exploratory and relatedbehavior: A new trend in animal research. Journalof Individual Psychology, 14, 111-120.

Carey, K. B., & Carey, M. P. (1993). Changes inself-efficacy resulting from unaided attempts toquit smoking. Psychology of Addictive Behaviors,7, 219-224.

Cohen, S. (1990). Control and the epidemiology ofphysical health: Where do we go from here? In J.Rodin, C. Schooler, & K. W. Schaie (Eds.),Self-directedness: Cause and effects throughout thelife course (pp. 231-240). Hillsdale, NJ: Erlbaum.

Coleman, J. S. (1988). Social capital in the creation ofhuman capital. American Journal of Sociology, 94,S95-S120.

Coleman, J. S. (1990). Foundations of social theory.Cambridge, MA: Harvard University Press.

Coleman, J. S., Campbell, E. Q., Hobson, C. J.,McPartland, J., Mood, A. M., Weinfeld, F. D., &York, R. L. (1966). Equality of educationalopportunity. Washington, DC: U.S. GovernmentPrinting Office.

Connell, J. P. (1990). Context, self, and action: Amotivational analysis of self-system processesacross the lifespan. In D. Ciccetti & M. Beeghly(Eds.), The self in transition: From infancy tochildhood (pp. 61-97). Chicago: University ofChicago Press.

Connell, J. P., Spencer, M. B., & Aber, J. L. (1994).Educational risk and resilience in African-American youth: Context, self, action, and out-comes in school. Child Development, 65, 493-506.

Czikszentmihalyi, M. (1991). Flow: The psychologyof optimal experience. New York: Harper.

Davidson, R. J. (1994). Asymmetric brain function,affective style, and psychopathology: The role ofearly experience and plasticity. Development andPsychopathology, 6, 741-758.

Davidson, R., Ekman, P., Saron, C , Senulis, J., &

Friesen, W. V. (1990). Approach/withdrawal andcerebral asymmetry: Emotional expression andbrain physiology, I. Journal of Personality andSocial Psychology, 58, 330-341.

DeCharms, R. (1968). Personal causation. NewYork: Academic Press.

Deci, E. L. (1975). Intrinsic motivation. New York:Plenum.

Deci, E. L. (1995). Why we do what we do: Thedynamics of personal autonomy. New York:Putnam.

Deci, E. L., & Ryan, R. M. (1985). Intrinsicmotivation and self-determination in human behav-ior. New York: Plenum.

DiClemente, C. C , Fairhurst, S. K., & Piotrowski,N. A. (1995). Self-efficacy and addictive behaviors.In J. E. Maddux (Ed.), Self-efficacy, adaptation,and adjustment: Theory, research, and application(pp. 109-141). New York: Plenum.

Diener, C. I., & Dweck, C. S. (1978). An analysis oflearned helplessness: Continuous changes in perfor-mance, strategy and achievement cognitions follow-ing failure. Journal of Personality and SocialPsychology, 36, 451-462.

DiMatteo, M. R. (1994). Enhancing patient adherenceto medical recommendations. Journal of theAmerican Medical Association, 271, 79-83.

Durkheim, E. (1951). Suicide (J. Spaulding & G.Simpson, Trans.). New York: Free Press. (Originalwork published 1897)

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation and personality.Psychological Review, 95, 256-273.

Eccles, J., Midgley, C , Wigfield, A., & Buchanan, C.(1993). Development during adolescence: Theimpact of stage-environment fit on young adoles-cents' experiences in schools and in families.American Psychologist, 48, 90-101.

Elder, G. H. J. (1998). The life course and humandevelopment. In R. M. Lemer (Ed.), Handbook ofchild psychology: Vol. 1. Theoretical models of humandevelopment (pp. 939-991). New York: Wiley.

Elliott, D., Menard, S., Rankin, B., Elliott, A.,Huizinga, D., & Wilson, W. J. (in press). Beatingthe odds: Successful youth development in disadvan-taged neighborhoods. Chicago: University ofChicago Press.

Erikson, E. H. (1953). Growth and crises of thehealthy personality. In H. A. Murray & D.Schneider (Eds.), Personality in nature, society,and culture (2nd ed., pp. 185-225). New York:Alfred A. Knopf.

Folkman, S. (1984). Personal control and stress andcoping processes: A theoretical analysis. Journal ofPersonality and Social Psychology, 46, 839-852.

Frankenhaeuser, M. (1983). The sympathetic-adrenaland pituitary-adrenal response to challenge: Com-parison between the sexes. In T. M. Dembroski,


T. H. Schmidt, & G. Blumchen (Eds.), Biobehav-ioral bases of coronary heart disease (pp. 91-105).Basel, Switzerland: Karger.

Frese, M., & Sabini, J. (Eds.). (1985). Goal directedbehavior: The concept of action in psychology.Hillsdale, NJ: Erlbaum.

Furstenberg, Jr., F. F., Cook, T., Eccles, J., Elder, G.,& Sameroff, A. (1998). Managing to make it:Urban families and adolescent success. Chicago:University of Chicago Press.

Gecas, V. (1989). The social psychology of self-efficacy. Annual Review of Sociology, 15, 291-316.

Gecas, V., & Schwalbe, M. L. (1983). Beyond thelooking glass self: Efficacy-based self-esteem.Social Psychology Quarterly, 46, 77-88.

Glass, D. C , & Singer, J. E. (1972). Urban stress:Experiments on noise and social stressors. NewYork: Academic Press.

Groos, K. (1901). The play of man (E. L. Baldwin,Trans.). New York: Appleton.

Gurin, P., & Brim, O. G. (1984). Change in self inadulthood: The example of sense of control. InP. B. Bakes & O. G. J. Brim (Eds.), Life-spandevelopment and behavior (pp. 282-334). NewYork: Academic Press.

Gurin, P., Gurin, G., Lao, R. C , & Beattie, M. (1969).Internal-external control in the motivational dynam-ics of Negro youth. Journal of Social Issues, 25,29-53.

Harlow, H. F. (1953). Mice, monkeys, men, andmotives. Psychological Review, 60, 23-32.

Harter, S. (1978). Effectance motivation reconsid-ered. Human Development, 21, 34—64.

Harter, S., & Zigler, E. (1974). The assessment ofeffectance motivation in normal and retardedchildren. Developmental Psychology, 10, 169-180.

Heckhausen, J., & Schulz, R. (1995). A life-spantheory of control. Psychological Review, 102,284-304.

Heider, F. (1958). The psychology of interpersonalrelations. New York: Wiley.

Hendrick, I. (1942). Instinct and the ego duringinfancy. Psychoanalytic Quarterly, 11, 33-58.

Hiroto, D. S., & Seligman, M. E. P. (1975). Generalityof learned helplessness in man. Journal ofPersonality and Social Psychology, 31, 311-327.

House, J. S., Lepkowski, J. M., Kinney, A. M., Mero,R. P., Kessler, R. C , & Herzog, A. R. (1994). Thesocial stratification of aging and health. Journal ofHealth and Social Behavior, 35, 213-234.

Jones, E. E., & Davis, K. E. (1965). From acts todispositions: The attribution process in personperception. In L. Berkowitz (Ed.), Advances inexperimental social psychology (vol. 2, pp. 219-266). New York: Academic Press.

Kagan, J., & Berkun, M. (1954). The reward value ofrunning activity. Journal of Comparative andPhysiological Psychology, 47, 108.

Kahn, R. L. (1993). An experiment in scientificorganization. Chicago: MacArthur Foundation Pro-gram in Mental Health and Human Development.

Kang, D. H., Davidson, R. J., Coe, C. L., Wheeler,R. W., Tomarken, A. J., & Ershler, W. B. (1991).Frontal brain asymmetry and immune function.Behavioral Neuroscience, 105, 860-869.

Kelly, H. H. (1967). Attribution theory in socialpsychology. In D. Levine (Ed.), Nebraska Sympo-sium on Motivation (vol. 15, pp. 192-238). Lin-coln: University of Nebraska Press.

King, V., & Elder, G. H. (1998). Perceived self-efficacy and grandparenting. Journal of Gerontol-ogy, 53B, S249-S257.

Kohn, M. L., & Schooler, C. (1983). Work andpersonality: An inquiry into the impact of socialstratification. Norwood, NJ: Ablex.

Krantz, D. S., Glass, D. C , Contrada, R., & Miller,N. E. (1981). National Science Foundation's five-yearoutlook on science and technology. Washington, DC:U.S. Government Printing Office.

Lachman, M. E. (1986). Personal control in later life:Stability, change, and cognitive correlates. InM. M. Baltes & P. B. Bakes (Eds.), The psychologyof control and aging (pp. 207-236). Hillsdale, NJ:Erlbaum.

Lachman, M. E., & Weaver, S. L. (1998). The sense ofcontrol as a moderator of social class differences inhealth and well-being. Journal of Personality andSocial Psychology, 74, 763-773.

Langer, E. (1979). The illusion of incompetence. In L.Perlmutter & R. Monty (Eds.), Choice andperceived control (pp. 301-313). Hillsdale, NJ:Erlbaum.

Langer, E. (1983). The psychology of control. BeverlyHills, CA: Sage.

Langer, E. J., & Rodin, J. (1976). The effects ofchoice and enhanced personal responsibility for theaged: A field experiment in an institutional setting.Journal of Personality and Social Psychology, 34,191-198.

Laudenslager, M. L., Ryan, S. M., Drugan, R. C ,Hyson, R. L., & Maier, S. F. (1983). Coping andimmunosuppression: Inescapable but not escapableshock suppresses lymphocyte proliferation. Science,221, 568-570.

Lefcourt, H. M. (1966). Internal versus externalcontrol of reinforcement: A review. PsychologicalBulletin, 65, 206-220.

Lefcourt, H. M. (Ed.). (1981). Research with the locusof control construct: Vol. I. Assessment methods.New York: Academic Press.

Lefcourt, H. M. (1992). Durability and impact of thelocus of control construct. Psychological Bulletin,112,411^14.

Markus, H. R., & Kitayama, S. (1991). Culture andthe self: Implications for cognition, emotion, andmotivation. Psychological Review, 98, 224-253.

336 HAIDT AND RODIN

Marmot, M. G., Smith, G. D., Stansfeld, S., Patel, C ,North, E, Head, J., White, I., Brunner, E., &Feeney, A. (1991). Health inequalities amongBritish civil servants: The Whitehall II study.Lancet, 337, 1387-1393.

Marx, K. (1963). Early writings (T. B. Bottomore,Trans.). New York: McGraw-Hill. (Original workpublished 1844)

Maslow, A. (1955). Deficiency motivation andgrowth motivation. In M. R. Jones (Ed.), NebraskaSymposium on Motivation, 1955 (pp. 1-30).Lincoln: University of Nebraska Press.

McAvay, G., Seeman, T. E., & Rodin, J. (1996). Alongitudinal study of change in domain-specificself-efficacy among older adults. Journal ofGerontology: Biological Sciences, 51B, B243-B253.

Meier, S. R, Ryan, S. M., Barksdale, C. M., & Kalin,N. H. (1988). Stressor uncontrollability and thepituitary-adrenal system. Behavioral Neurosci-ence, 100, 669-678.

Mirowsky, J., & Ross, C. E. (1989). Social causes ofpsychological distress. New York: Aldine deGruyter.

Moos, R. H. (1987). The social climate scales: Auser's guide. Palo Alto, CA: Consulting Psycholo-gists Press.

Moos, R. (1991). Connections between school, work,and family settings. In B. J. Fraser & H. J. Walberg(Eds.), Educational environments: Evaluation,antecedents, and consequences (pp. 29-53). Ox-ford, England: Pergamon Press.

Moos, R. H., & Lemke, S. (1992). The multiphasicenvironmental assessment procedure: A user'sguide. Palo Alto, CA: Consulting PsychologistsPress.

Murray, H. A. (1938). Explorations in personality.London: Oxford University Press.

O'Leary, A. (1985). Self efficacy and health. Behav-ioral Research Therapy, 23, 437^t51.

Overmeier, J. B., & Leaf, R. C. (1965). Effects ofdiscriminative Pavlovian fear conditioning uponpreviously or subsequently acquired avoidanceresponding. Journal of Comparative and Physi-ological Psychology, 60, 213-218.

Pearlin, L., & Schooler, C. (1978). The structure ofcoping. Journal of Health and Social Behavior, 19,2-21.

Pearlman, C. (1984). The effects of level of effectancemotivation, IQ, and a penalty/reward contingencyon the choice of problem difficulty. Child Develop-ment, 55, 537-542.

Peterson, C , Maier, S. R, & Seligman, M. E. P.(1993). Learned helplessness: A theory for the ageof personal control. New York: Oxford UniversityPress.

Peterson, C , Semrnel, A., von Baeyer, C , Abramson,L. Y, Metalsky, G. I., & Seligman, M. E. P. (1982).

The Attributional Style Questionnaire. CognitiveTherapy and Research, 6, 287-299.

Piaget, J. (1952). The origins of intelligence inchildren (M. Cook, Trans.). New York: Interna-tional Universities Press.

Prussia, G. E., & Kinicki, A. J. (1996). A motivationalinvestigation of group effectiveness using socialcognitive theory. Journal of Applied Social Psychol-ogy, 81, 187-198.

Rescorla, R. A. (1967). Pavlovian conditioning andits proper control procedures. Psychological Re-view, 74, 71-79.

Rodin, J. (1986a). Aging and health: Effects of thesense of control. Science, 233, 1271-1276.

Rodin, J. (1986b). Health, control, and aging. In M.Baltes & P. Bakes (Eds.), The psychology ofcontrol and aging (pp. 139-165). Hillsdale, NJ:Erlbaum.

Rodin, J. (1989). Sense of control: Potentials forintervention. The Annals of the American Academyof Political and Social Science, 503, 29^2 .

Rodin, J., & Langer, E. (1977). Long-term effects of acontrol-relevant intervention with the institutional-ized aged. Journal of Personality and SocialPsychology, 35, 897-902.

Rodin, J., Rennert, K., & Solomon, S. K. (1980).Intrinsic motivation for control: Fact or fiction? InA. Baum & J. E. Singer (Eds.), Advances inenvironmental psychology (pp. 131-147). Hillsdale,NJ: Earlbaum.

Rotter, J. B. (1966). Generalized expectancies forinternal versus external control of reinforcement.Psychological Monographs, 80(1), 1-28.

Rotter, J. B. (1975). Some problems and misconcep-tions related to the construct of internal versusexternal control of reinforcement. Journal ofConsulting and Clinical Psychology, 43, 56-67.

Sampson, R. J. (1993). Family management and childdevelopment: Insights from social disorganizationtheory. In J. McCord (Ed.), Advances in criminologi-cal theory (pp. 63-93). New Brunswick, NJ:Transaction Press.

Sapolsky, R. N. (1992). Cortisol concentrations andthe social significance of rank instability amongwild baboons. Psychoneuroendocrinology, 17,1Q\-709.

Satterfield, J. M., Monahan, J., & Seligman, M. E. P.(1997). Law school performance predicted byexplanatory style. Behavioral Science and the Law,15, 95-105.

Scheier, M. F, & Carver, C. S. (1985). Optimism,coping, and health: Assessment and implications ofgeneralized outcome expectancies. Health Psychol- •ogy, 4, 219-247.

Seeman, M. (1959). On the meaning of alienation.American Sociological Review, 24, 783-791.

Seeman, M. (1975). Alienation studies. AnnualReview of Sociology, 1, 91-123.


Seeman, M. (1991). Alienation and anomie. In J.Robinson, P. Shaver, & L. Wrightsman (Eds.),Measures of personality and social psychologicalattitudes (pp. 291-371). San Diego, CA: AcademicPress.

Seeman, M., & Seeman, T. (1983). Health behaviorand personal autonomy: A longitudinal study of thesense of control in illness. Journal of Health andSocial Behavior, 24, 144-160.

Seligman, M. E. P. (1975). Helplessness: Ondepression, development, and death. San Fran-cisco: Freeman.

Seligman, M. E. P. (1991). Learned optimism. NewYork: Alfred A. Knopf.

Seligman, M. E. P., & Maier, S. F. (1967). Failure toescape traumatic shock. Journal of ExperimentalPsychology, 74, 1-9.

Seligman, M. E. P., & Visintainer, M. A. (1985).Tumor rejection and early experience of uncontrol-lable shock in the rat. In F. R. Brush & J. B.Overmeier (Eds.), Affect conditioning and cogni-tion: Essays on the determinants of behavior (pp.203-210). Hillsdale, NJ: Erlbaum.

Shweder, R. A., Balle-Jensen, L., & Goldstein, W.(1995). Who sleeps by whom revisited: A methodfor extracting the moral goods implicit in praxis. InJ. Goodnow, P. Miller, & F. Kessell (Eds.), Culturalpractices as contexts for development (pp. 21-39).San Francisco: Jossey-Bass.

Sieber, W. J., Rodin, J., Larson, L., Ortega, S.,Cummings, N., Levy, S., Whiteside, T., & Herber-man, R. (1992). Modulation of human natural killercell activity by exposure to uncontrollable stress.Brain, Behavior and Immunity, 6, 141—156.

Skinner, E. (1995). Perceived control, motivation,and coping. Thousand Oaks, CA: Sage.

Skinner, E. A. (1996). A guide to constructs ofcontrol. Journal of Personality and Social Psychol-ogy, 71, 549-570.

Skinner, E. A., & Connell, J. P. (1986). Controlunderstanding: Suggestions for a developmentalframework. In M. M. Bakes & P. B. Bakes (Eds.),The psychology of control and aging (pp. 35-69).Hillsdale, NJ: Erlbaum.

Sutton, S. K., & Davidson, R. J. (1997). Prefrontalbrain asymmetry: A biological substrate of thebehavioral approach and inhibition systems. Psy-chological Science, 8, 204-210.

Taylor, S. E. (1989). Positive illusions: Creativeself-deception and the healthy mind. New York:Basic Books.

Taylor, S. E., Repetti, R. L., & Seeman, T. (1997).Health psychology: What is an unhealthy environ-ment and how does it get under the skin? AnnualReview of Psychology, 48, 411^447.

Thoits, P. A. (1983). Multiple identities and psycho-logical well-being. American Sociological Review,48, 174-187.

Thompson, S. C , Cheek, P. R., & Graham, M. A.(1988). The other side of perceived control:Disadvantages and negative effects. In S. Spacopan& S. Oshamp (Eds.), The social psychology ofhealth (pp. 69-93). Beverly Hills, CA: Sage.

Thorlindsson, T, & Bjarnason, T. (1998). ModelingDurkheim on the micro level: A study of youthsuicidality. American Sociological Review, 63,94-110.

Timko, C , & Moos, R. H. (1989). Choice, control,and adaptation among elderly residents of shelteredcare settings. Journal of Applied Social Psychol-ogy, 19, 636-655.

Wallston, K. A., & Wallston, B. S. (1982). Who isresponsible for your health? The contract of healthlocus of control. In G. S. Sanders & J. Suls (Eds.),Social psychology of health and illness. Hillsdale,NJ: Erlbaum.

Wegner, D., & Bargh, J. (1998). Control andautomaticity in social life. In D. T. Gilbert, S. T.Fiske, & G. Lindzey (Eds.), Handbook of socialpsychology (4th ed., pp. 446-496). New York:McGraw-Hill.

Weinberg, R. S., Gould, D., & Jackson, A. (1979).Expectations and performance: An empirical test ofBandura's self-efficacy theory. Journal of SportPsychology, 1, 320-331.

Weiss, J. M., Stone, E. A., & Harrell, N. (1970).Coping behavior and brain norepinephrine level inrats. Journal of Comparative and PhysiologicalPsychology, 72, 153-160.

Wellborn, J. G., & Connell, J. P. (1987). The ResearchAssessment Package for Schools. Rochester, NY:University of Rochester.

White, R. B. (1959). Motivation reconsidered: Theconcept of competence. Psychological Review, 66,297-333.

Wilson, E. O. (1998). Consilience: The unity ofknowledge. New York: Alfred A. Knopf.

Wortman, C , & Brehm, J. (1975). Responses touncontrollable outcomes: An integration of reac-tance theory and the learned helplessness model. InL. Berkowitz (Ed.), Advances in experimental andsocial psychology (pp. 277-336). New York:Academic Press.

Wright, R. (1998). Ability perception and cardiovas-cular response to behavioral challenge. In M. Kofta& G. Weary (Eds.), Personal control in action:Cognitive and motivational mechanisms (pp. 197-232). New York: Plenum.

Zimmerman, M. A., & Rappaport, J. (1988). Citizenparticipation, perceived control, and psychologicalempowerment. American Journal of CommunityPsychology, 16, 725-750.

Received November 10, 1998Revision received April 29, 1999

Accepted May 19, 1999 •

Control and Efficacy as Interdisciplinary...

Documents

Transcript of Control and Efficacy as Interdisciplinary...