diener y dweck (1980). an analysis of learned ii the processing of success

Journal of Personality and Social Psychology1980, Vol. 39, No. 5, 940-952

An Analysis of Learned Helplessness:II. The Processing of Success

Carol I. Diener and Carol S. DweckUniversity of Illinois at Urbana-Champaign

Helpless children attribute their failures to lack of ability and view them asinsurmountable. Mastery-oriented children, in contrast, tend to emphasize mo-tivational factors and to view failure as surmountable. Although the performanceof the two groups is usually identical during success or prior to failure, past re-search suggests that these groups may well differ in the degree to which theyperceive that their successes are replicable and hence that their failures areavoidable. The present study was concerned with the nature of such differences.Children performed a task on which they encountered success and then failure.Half were asked a series of questions about their performance after success andhalf after failure. Striking differences emerged: Compared to mastery-orientedchildren, helpless children underestimated the number of successes (and over-estimated the number of failures), did not view successes as indicative of ability,and did not expect the successes to continue. Subsequent failure led them todevalue their performance but left the mastery-oriented children undaunted.Thus, for helpless children, successes are less salient, less predictive, and lessenduring—less successful.

When individuals view their actions as ir-relevant to subsequent outcomes, they may besaid to display "learned helplessness" (seeSeligman, Maier, & Geer, 1968). Such beliefsabout control over outcomes have been foundto be important determinants of behavior inachievement situations. For example, per-ceiving that one is unable to surmount failurecan have highly debilitating effects on per-formance, as seen in studies on learned help-lessness (Diener & Dweck, 1978; Dweck,1975; Dweck & Bush, 1976; Dweck & Rep-pucci, 1973). In contrast, perceiving that oneis able to avoid or to escape from failure canhave facilitating effects (Diener & Dweck,

This study is based on a dissertation submitted bythe first author to the Department of Psychology,University of Illinois. The research was supportedin part by U.S. Public Health Service Grant MH05089-31. The authors would like to thank RossParke, Bill Redd, Robert Wyer, and Ed Diener fortheir helpful comments.

Requests for reprints should be sent to Carol S.Dweck, who is now at Larsen Hall, Harvard Uni-versity, Cambridge, Massachusetts 02138.

1978; Dweck, 1975). Interestingly, thesedivergent responses to failure occur despiteequivalent performance prior to failure andequivalent performance on measures of abil-ity such as IQ and reading comprehension(Dweck & Licht, in press).

Much research on learned helplessness hasfocused on attributions as indicants of beliefabout control over outcomes. Specifically, ithas been shown that attributions of failures torelatively invariant factors, like a lack of abil-ity, are associated with performance decre-ments under failure; attributions of failuresto more readily modifiable factors, like a lackof effort, characterize the more "mastery-ori-ented" individuals and are associated withmaintenance or improvements in performancefollowing failure (e.g., Dweck, 1975; Dweck& Reppucci, 1973; Weiner, 1972, 1974).

More recently, Diener and Dweck (1978)sought to determine what other achievementcognitions might be important in mediatingresponses to failure. In prior studies, state-ments of particular achievement cognitionswere solicited at prespecified times. For ex-

Copyright 1980 by the American Psychological Association, Inc. 0022-3514/80/3905-0940$00.75

940

AN ANALYSIS OF LEARNED HELPLESSNESS 941

ample, in studying attributions, the experi-menter has typically been the one who defineswhen the failure has occurred and when itscauses should be explained. This approachrules out the possibility of obtaining informa-tion about cognitions other than attributions.Even considering only attributions, one learnsnothing about the time at which they wouldhave ordinarily occurred or if, indeed, theywould have occurred at all. To allow detec-tion of a wider constellation of cognitions,Diener and Dweck (1978) requested childrento report "what they were thinking about" asthey performed a task. They could therebymonitor potential differences in a variety ofachievement-related verbalizations underfailure.

The statements offered by helpless andmastery-oriented children in Diener andDweck's (1978) study were virtually indis-tinguishable prior to the introduction offailure. However, following failure there wasalmost no overlap in the types of statementsmade by the two groups. The helpless chil-dren made the expected attributions to un-controllable factors, but the mastery-orientedchildren did not offer explanations for theirfailures. Instead they appeared to be search-ing for a remedy to their current lapse inperformance by engaging in solution-directedverbalizations such as self-instructions andself-monitoring. The attitudes of the twogroups toward the task after the onset offailure differed as well. The mastery-orientedchildren tended to maintain positive affecttoward the task, and they continued toexpress a positive prognosis for their per-formance. Indeed, the mastery-oriented chil-dren did not appear to define themselves ashaving failed at all. In contrast, the helplesschildren began to express an appreciabledegree of negative affect along with state-ments that implied they had given up tryingto solve the problems. In sum, the verbaliza-tions of the helpless children under failurereflected a tendency to dwell on the present,to emphasize the negative, and to seek escapefrom the situation. The verbalizations of themastery-oriented reflected a tendency to looktoward the future, to emphasize the positive,and to redouble their task-directed efforts.

Thus far research has concentrated on cog-

nitions about failures. What about success?Even though the helpless and mastery-ori-ented groups typically solve equal numbers ofproblems in equal amounts of time prior tofailure, there is evidence to suggest that theymay differ in their perceptions of their pastsuccess. For example, Dweck and Reppucci(1973) showed different attribution patternsfor the two groups, with mastery-orientedchildren attributing success to factors thatimplied their successes were replicable (e.g.,effort) but with helpless children attributingtheir success to factors that did not suggestthat past performance was indicative offuture performance (e.g., external factors).In addition, in the Diener and Dweck study,several statements offered by the mastery-oriented children suggested that past suc-cess was perceived as highly relevant to fu-ture success, but those statements given byhelpless children suggested that they mightbe perceiving past success as irrelevant tofuture outcomes. For example, mastery-ori-ented children made statements like, "I wasdoing well before, so if I slow down andconcentrate I'll get it right again." Helplesschildren, on the other hand, appeared to ig-nore their earlier successes, and as their per-formance deteriorated in quality, they madestatements like, "I've never been good at thissort of thing" and "I never did have a goodrememory (sic)," even though only minutesearlier, before the failure began, they dem-onstrated quite admirable memory and abilityat the task. Although some helpless childrendid acknowledge their prior success, theytended to discount its bearing on present orfuture performance with statements like, "I'mgetting confused" and "I don't know what I'mdoing anymore." In view of these observa-tions, one might not expect success to pre-vent the negative effects of failure to the sameextent for helpless and mastery-oriented chil-dren; indeed, Dweck (1975) has shown thatfor helpless children, continued success in anarea was not sufficient in itself to buffer theeffects of failures.

It appears likely that just as there is avariety of achievement-related cognitionsabout failure that separate helpless and mas-tery-oriented children, there are also impor-tant differences in their perceptions of sue-

942 CAROL I. DIENER AND CAROL S. DWECK

cess. Critical questions relate to the precisenature of these differences and whether thedifferences occur before or after failure. Thepresent investigation was therefore concernedwith differences between helpless and mas-tery-oriented children in their perception oftheir successes. As in the Diener and Dweck(1978) study, all of the children were givena series of success problems, followed byfailure problems. However, either followingthe success problems or following the failureproblems, children were asked a set of ques-tions designed to tap their perceptions oftheir prior success and their predictions offuture success.

There are several possible ways in whichhelpless children may differ from mastery-oriented children in their evaluation of theirsuccess. Some of these differences may occurduring success, or they may be prompted byfailure. First, it is possible that helpless andmastery-oriented children differ in their per-ceptions of how successful they have been.Although both groups may recognize thatthey have correctly solved the problems, themastery-oriented children may think theyhave performed well, whereas the helplesschildren do not. For example, the helplesschildren may feel they should have solvedthe problem more quickly. Second, it is possi-ble that the two groups make different judg-ments about how other children would do atthe same task, with helpless children morelikely to view others as outperforming them.It is also possible that helpless children mayactually remember having had fewer suc-cesses; that is, compared to mastery-orientedchildren they may underestimate the numberof problems they solved. To evaluate thesepossibilities, children were asked after thesuccess problems to evaluate their perform-ance, to estimate how well most childrenwould do at the task, and to estimate howmany problems they had solved correctly.

A further possibility is that helpless chil-dren do not view their present success aspredicting future success. For example, assuggested by the Dweck and Reppucci (1973)findings, helpless children may attribute theirsuccesses to factors that do not imply con-tinued success. Mastery-oriented childrenmay, on the other hand, attribute success to

ability or to other factors that do predictfuture success. If this were true, then wewould expect helpless children, even whilethey are experiencing success, to be morepessimistic in their predictions of their futureperformance and to choose a level equal to orlower than their current performance. Mas-tery-oriented children would be expected topredict a future performance level at leastequal to their current level—or better—ifthey expect to improve with additional prac-tice. This possibility was assessed by askingthe children to apportion their attributionsfor success among several stable and variablecauses. The children were also asked to pre-dict their future performance and to estimatehow many problems they could get right ifgiven 15 more problems of the same type.

Another possibility is that helpless andmastery-oriented children perceive their suc-cesses similarly until failure occurs. It may bethat at that time, helpless children reevaluatetheir prior success in light of the currentfailure. For example, failure may overshadowthe actual successes of the helpless children—they may now underestimate the number ofpast successes and may overestimate theirfailures. Perhaps helpless children and mas-tery-oriented children have similar attribu-tions following success, but they revise theirattributions for success when failure occurs.If helpless children reinterpret their success,one would predict that following failure, theywould be less likely to attribute their pastsuccesses to ability and more likely to at-tribute them to external factors than theywere immediately following success. Mastery-oriented children would be expected to main-tain their attributions of success to internalfactors. To assess these possibilities, a secondgroup of helpless and mastery-oriented chil-dren were asked, following failure, to recallthe number of successfully solved problems,to estimate the number of unsolved problems,to identify the reasons for their earlier suc-cess, and to judge whether they would be ableto successfully complete the earlier (success)problems if they tried them again. It is alsopossible that helpless children are more likelythan mastery-oriented children both to de-value their successes as they occur and tofurther discount their successes after failure.


In summary, although past research hasnot addressed these issues directly, variousfindings suggest that mastery-oriented chil-dren tend to view their successes as predictiveof future success, even after failure. In con-trast, the available evidence implies that help-less children do not appear as likely to ac-knowledge their successes and that even whenthey do, they are less likely to see them asrelevant to future performance. The presentresearch was aimed at determining moreclearly how helpless and mastery-orientedchildren process their successes as they oc-cur and how they might reevaluate their suc-cesses in light of failure.

MethodOverview

Children, classified as helpless or mastery-oriented,received eight success and four failure problems ona discrimination task. The level of their hypothesis-testing strategy was monitored throughout. Half ofthe group answered a series of questions about theirperformance following the success problems andagain after the failure problems. The other half ofthe group answered the questions only at comple-tion of the task (following both success and failureproblems).

Participants

The participants were 14 fourth graders, 72 fifthgraders, and 26 sixth graders from a working-classschool district. Approximately one third of thechildren were black and two thirds were white.Several children were randomly eliminated (beforethe collection of data on the individual task) sothat there would be equal numbers of males andfemales, helpless and mastery-oriented. The finalsample was composed of 56 males and 56 females.

Measure oj Helplessness

Since past research (Diener & Dweck, 1978;Dweck, 1975; Dweck & Reppucci, 1973; Floor &Rosen, 1975) has indicated that a major differencebetween helpless and mastery-oriented subjects liesin their respective tendencies to neglect or emphasizethe role of effort in determining their failures, thisrelative emphasis was used as the criterion for divid-ing children into helpless and mastery-orientedgroups. The Intellectual Responsibility (IAR) Scale(Crandall, Katkovsky, & Crandall, 1965) was usedfor this purpose. The IAR consists of 34 forced-choice items that describe either a positive or nega-tive achievement experience that frequently occursin the daily lives of children. The child must choose

between two alternative attributions on each item,one attributing the cause of the event to someoneelse in the child's environment (external responsibil-ity) and the other to his/her own behavior (in-ternal responsibility). A subset (10 items) of thequestions on the IAR specifically taps the child'sattributions of failure to lack of effort.

As in the Diener and Dweck (1978) study, boththe mean and median number of effort responseswere about 7 (6.83 and 7.00, respectively). Thosechildren scoring 6 or below were placed in thehelpless group, and those scoring 8 and above weredesignated as mastery-oriented. Subjects scoring 7were excluded from the present study. The IAR wasadministered in written form to all subjects at least1 month prior to the experimental session.

Task and Materials

The same task as the one used in the prior studyby Diener and Dweck (1978) was employed. Thetask consisted of a series of three-dimension, two-choice discrimination problems on which the childsearches for the one solution that is correct. Eachchild was presented with eight solvable problems(success problems) followed by four unsolvableproblems (failure problems). A problem consisted ofa set of four stimulus cards (see Figure 1), with eachcard containing two figures that varied on the threedimensions: color (e.g., red or blue), form (e.g.,square or triangle), and the symbol in the center ofthe form (e.g., dot or star).

The stimuli were varied in a systematic fashion sothat the child's hypothesis about the correct solutioncould be inferred unambiguously from his or herchoices. For example, a child who is testing thehypothesis "triangle" would choose cards in thesequence of left, left, right, left, as can be seen inFigure 1. In order to monitor hypothesis testing, a"blank trial" procedure was employed in which thechildren did not receive feedback about the correct-ness of their responses on the first three of everyfour trials (Levine, 1966). A hypothesis was definedas the consistent selection of the same stimulusproperty, such as the shape "triangle," over the fourtrials prior to feedback. The instructions explicitlypointed out that when the experimenter gave noresponses this meant neither right nor wrong. Thestimuli were ordered within a single block of fourtrials so that all useful hypotheses could be sepa-rated from position responses—for example, choosingthe right side of the card each time or alternatingright and left responses without considering thestimuli. The instructions also explicitly stated thatthe only possible solution was one of the colors,shapes, or interior symbols.

Procedure

Success problems. The problems were introducedto the child as being "a task for kids of your age."It was stressed that the child's job was to find the


Figure 1. Example of four consecutive stimulus cardsthat allow the tracking of a hypothesis over the no-feedback trials.

correct solution for each deck of cards. On the firstproblem, veridical feedback ("right" or "wrong")was given after every response. If the child wasunable to reach the solution on completion of a deckof 16 cards, the same problem was repeated, with ahint provided by the experimenter (see Diener &Dweck, 1978, for exact procedure). Each child wasgradually introduced to the no-feedback procedure,so that by the seventh and eighth problem they werereceiving feedback only after every fourth response.Each child was guaranteed to have solved all eightproblems correctly before continuing.

Following the eight success problems, one half ofthe children (equal numbers of boys and girls, help-less and mastery-oriented) were questioned abouttheir performance up to that point. The measuresincluded an evaluation of their performance, anevaluation of how well they thought other childrentheir age would do, an expectancy of future per-formance on the same task, an assessment of howmany they expected to get right if given 15 moresimilar problems, an accuracy of recall measure (for

number of successes), and attributions for theirsuccess.

The children were told, "There are a number ofthings I would like to ask you about these prob-lems. We are going to stop here for a few minutesso I can talk to you about them." Children werethen shown a chart that contained the numbersbetween 1 and 10, anchored by "not at all good" and"very, very good." The scale was explained, and thechildren were asked to choose the number thatcorresponded to how well they thought they haddone up to that point. The children were thenasked, "How well do you think most kids your agewould do on these problems?" They again chose anumber from 1 to 10, using the chart that was usedfor their own evaluation. Knowing the children'spredictions of others' performance allows a clearerinterpretation of their evaluation of their own per-formance. For example, a child may rate his or herown performance as 8 on a 10-point scale; but ifthat child thinks that most other children wouldrate a 9 or 10 on the scale, then he or she may notconsider 8 to be a successful score. On the otherhand, if the child believed most other childrenwould only rate a 4 or 5, then his or her perform-ance might be outstanding by comparison.

Using the same chart, the children were thenasked, "How well do you think you will do on therest of the problems?" To ascertain more preciselythe meaning of their performance estimates, thechildren were asked to estimate how many prob-lems they thought they could solve if they weregiven IS more problems of the same type.

During piloting, children were asked to generateattributions for success, but they appeared to findthe whole idea foreign. That is, they failed to com-prehend what was being requested and why. Thisdid not seem to be a methodological problem be-cause when the nearly identical question was askedvis-a-vis failure in the earlier study (Diener &Dweck, 1978), the children had no difficulty inanswering the question. It appeared that "explain-ing" success was not something children find natural.However, this does not mean that different chil-dren may not understand their success differentlybut only that explaining success in an explicit fashionis not typical. There may well be a differentialtendency in what helpless and mastery-orientedchildren see as reflecting ability, and it was desira-ble to have a procedure to tap this. Therefore,attributions were measured using an "attributionwheel" that children could adjust to show the rela-tive importance of possible causes for their success(see Nicholls, 1975). The wheel was composed offour interconnected discs, each representing acausal factor. By moving the metal handles at-tached to each disc, the child could easily adjustthe discs to expose any proportion of a given disc.Each disc was clearly labeled with one of the fol-lowing choices: I was lucky; It was easy; I am goodat this; I am smart. Of interest was any differentialtendency on the part of helpless and mastery-ori-ented children to see success as reflecting ability(either general or task specific) versus impersonalfactors. An attribution to effort was not included


because the task clearly required an appreciableexpenditure of effort for success, and it was unde-sirable to have the choice of this factor obscure thedifference of interest. In short, this procedure wasdesigned not to tap "true" or "natural" attributionsbut to detect any differential emphasis on ability inidentifying the causes of success.

Children were shown how to use the wheel toindicate the relative importance of each of the fourcauses. Following the instructions, they were giventhe wheel with the four segments equally exposedand were asked to show the reasons for their success-ful solution of the problems. Each disc was dividedinto 35 equal segments, and the number of segmentsallotted to each causal factor was recorded. Afterthe children made their choices, they were told,"Let's go on now and do some more problems."

Failure problems. Following the eight successproblems, the children were given four problems of20 cards each with the consistent feedback of"wrong" (i.e., after every fourth response) on eachproblem. This allowed the children sufficient oppor-tunity to search for the solution but ensured thatgiven their strategy level, they would not havetested all possible solutions. This, in effect, made theproblems unsolvable for them, but not obviously so.To determine children's retrospective perceptions oftheir success, following the failure problems allchildren were asked the questions described at theend of the success problems, with appropriate word-ing changes. For example, the question dealing withattributions for success was reworded: "Why doyou think you did well on the earlier problems?"In addition, they were asked to estimate the totalnumber of problems for which they had not found asolution. Finally, to see if the children believedthey retained their ability to work the problemsthey had previously solved, the children were asked,"If we went back to these first problems (showingthem an earlier package), would you still be able todo them?"

Scoring procedures. The children's hypothesis-testing strategies can be classified as "useful" (i.e.,those that would eventually lead to problem solutionwhen followed perfectly) or "ineffective" (i.e., se-quences of hypotheses that would never lead toproblem solution). Within these two broad cate-gories, strategies can be further defined. The usefulstrategies, in descending order of efficiency, arefocusing, dimension checking, and hypothesis check-ing. In focusing, the child processes informationperfectly and eliminates all dimensions that havebeen logically disconfirmed on each feedback trial.Focusing was not included in this analysis becauseonly a very few children showed any evidence ofusing it. In dimension checking, the child proceedsthrough all three dimensions (color, form, interiorsymbol) in a systematic fashion. When the childtests one number of a dimension and receives feed-back, on the next trial, he or she chooses the stimu-lus that is consistent with that feedback, even ifshe or he is now testing a different dimension. Forexample, suppose the child tested the shape dimen-sion on the previous trial by choosing a yellowcircle and was told "wrong." If the child is now

Table 1Number of Helpless and Mastery-OrientedChildren Whose Hypothesis-Testing StrategyImproved, Remained the Same, orDeteriorated Following Failure

Group

HelplessMastery-oriented

Im-proved Same

4 1915 33

Deteri-orated

338

testing the color dimension, she or he will choosered. In hypothesis checking, the child eliminates onlyone possible solution per feedback. For example,suppose the child tested the shape dimension bychoosing a yellow circle and received "wrong" feed-back; if the child then decided to test the colordimension, she or he would still try both yellow andred.

Ineffectual task strategies are response sets thatcan never lead to the solution of the problem. Thethree ineffectual strategies are stimulus preference,position alternation, and position perseveration.Stimulus preference refers to the selection of asingle stimulus characteristic (e.g., the color red)independent of feedback. Position alternation occurswhen the child alternately chooses the left and thenthe right stimulus regardless of what they are, andposition perseveration occurs when the child choosesthe stimulus in the same position each time. Theseineffectual task strategies can be ordered on thebasis of the ages of children who most frequentlyuse them, with stimulus preference being used byolder children and position perseveration being usedby the youngest children.

The particular strategy being used by the childcould be determined by monitoring the child'shypotheses across blocks of trials. The scoringcriterion used by Diener and Dweck (1978) wasadopted for this study (cf. Gholson, Levine, & Phil-lips, 1972). If a child used the same strategy onthree out of five blocks of trials, he or she wasscored as employing that strategy.

In summary, each child received eight successproblems and four failure problems. The sophistica-tion of the child's strategy use was monitored. Halfof the children were questioned about their per-ceptions of success following Success and again afterfailure. The other half of the children were ques-tioned after failure only.

ResultsOverview

Performance measures taken during thesuccess problems (prior to failure) are pre-sented to document the equivalence of per-formance on the parts of helpless and mas-tery-oriented children. This similarity ofperformance was in marked contrast to the


Table 2Percentage of Helpless and Mastery-Oriented Children Exhibiting Each Type ofStrategy on the Four Failure Problems

Problem number

Group

HelplessUseful strategies

Dimension checkingHypothesis checking

Ineffectual strategiesStimulus preferencePosition alternationPosition perseveration

Mastery orientedUseful strategies

Dimension checkingHypothesis checking

Ineffectual strategiesStimulus preferencePosition alternationPosition perseveration

17.577.0

5.5.0.0

44.555.5

.0

.0

.0

20.051.5

16.09.03.5

50.044.5

5.5.0.0

5.048.5

19.525.02.0

44.551.5

2.0.0.0

7.041.0

17.523.510.5

55.544.5

.0

.0

.0

children's perceptions of those successes (andto their performance on failure problems).

Performance on Success Problems

None of the measures—total number ofhints needed for all eight success problems,number of hints on problems seven andeight, or the number of times a child requiredmore than one set of hints to solve the prob-lem—showed a significant correlation withthe subscales of the IAR, r = -.15 to .06.Chi-square analyses on the number of timeschildren used dimension checking versus hy-pothesis checking during training also yieldedno significant differences. Helpless childrenused dimension checking 50% of the timeand hypothesis checking 50% of the time, ascompared to the mastery-oriented childrenwho utilized dimension checking 41% of thetime and hypothesis checking 59% of thetime.

Performance on Failure Problems

To provide a general picture of change inperformance following failure, a 2 X 2 (help-less vs. mastery-oriented, male vs. female)analysis of variance was performed on the

number of times that each child used a dis-confirmed or ineffectual hypothesis on thetest problems. The results show a significantlygreater use of disconfirmed and ineffectualhypotheses by the helpless (M = 8.5) thanby the mastery-oriented group (M — 2.6),F(l, 108) = 130.6, p < .001. This generalfinding is further supported by the significantnegative correlation between the number ofdisconfirmed and ineffectual hypotheses usedand the IAR effort items (r = -.66, p <.01).

Table 1 shows the number of children ineach group whose strategies deteriorated,remained the same, or improved. A chi-squareanalysis for the data shown in Table 1 (com-bining the Same and Improved categories)yielded a highly significant difference betweenhelpless and mastery-oriented groups, x20)= 24.04, p < .01.

The percentage of helpless and mastery-oriented children who exhibited each type ofeffective and ineffectual strategy across thefour problems is shown in Table 2. Helplesschildren showed a progressive decrease in theuse of effective strategies and an attendantincrease in ineffectual strategies. In contrast,the mastery-oriented children showed littledeterioration in their use of strategies over


test problems, and, in fact, some mastery-oriented children showed a clear tendency tobecome more sophisticated in their strategyuse as they received failure feedback. All ofthe above findings are consistent with thosefound in the Diener and Dweck (1978)studies.

The foregoing results clearly show thatthere was no difference between the helplessand mastery-oriented children in acquisitionor performance while they were experiencingsuccess. Their objective performance duringsuccess was the same, but were their percep-tions of that performance equivalent?

Perceptions oj Success

Overall analyses. To examine the effects ofrace, sex, and time of questioning, a 2 (group:helpless vs. mastery-oriented) X 2 (sex:male vs. female) X 2 (race: black vs. white)X 2 (time of questions: after success vs.after failure) multivariate analysis of vari-ance (MANOVA; Finn, 1972) was performedon eight dependent variables. (Since four ofthe nine dependent measures [the attribu-tions] are ipsative and the fourth value isdetermined by the other three, only threewere used in the MANOVA.) The main effectfor group showed that helpless children re-sponded to the questions in a different fashionfrom the mastery-oriented children, F(8, 89)= 22.75, p < .001. The main effects for raceand sex were not significant (p < .35 and p< .51, respectively), and, therefore, allfurther analyses will be collapsed across thesetwo variables. The main effect for time ofquestioning (after success or after failure)approached significance (p < .07). This wasprimarily due to the two items concerningfuture success (the expectancy of future suc-cess and the estimate of how many theywould get right if an additional 15 similarproblems were given) and reflected the ten-dency of children who have experienced fail-ure to have lower expectations for the future.None of the higher order interactions wassignificant.

Multiple comparisons. Planned compari-sons using Tukey's honestly significant differ-ence (HSD) test (Kirk, 1968) were per-formed to test the major hypotheses of the

study. These analyses allow us to examineseparately the differences between helplessand mastery-oriented children's perception ofsuccess after they have experienced only suc-cess and the differences between the groupsafter they have experienced failures as well.(Again, these analyses are based on the re-sponses of one set of children who experiencedonly success compared to a second group ofchildren who experienced both success andfailure. The postfailure responses of the chil-dren who were questioned twice—after suc-cess and after failure—are not utilized inthese analyses.) Additionally, differences inperceptions of success before and after fail-ure were examined by comparing the responsesof the group of helpless children respondingafter success to the other group of helplesschildren who responded after failure. Similarcomparisons were done for the mastery-ori-ented children. The results of these compari-sons are summarized in Table 3. To maintaina conservative experimentwise significancelevel, .01 was adopted as the alpha levelnecessary for significance.

As can be seen in Table 3, when one looksat responses given after only success, there isno difference between groups in their overallevaluation of their performance. Both judgedthemselves to be doing well (helpless = 6.39;mastery-oriented = 7.00). However, signifi-cant differences emerged for their expectancyof future success, the number they expectedto get right out of 15, and how well theybelieved that most children of the same agewould do. The mastery-oriented childrenexpected both to do better in the future andto get more right than did the helpless chil-dren. The mastery-oriented expected to getabout 90% of the problems correct if theywere given more of the same type, whereasthe helpless children expected to solve only50% of the problems. When asked to esti-mate how well most children pf the same agewould do at these problems, the helpless chil-dren believed that most other children wouldbe better at the task than did the mastery-oriented children. (It should be noted thatthe helpless children's evaluation of otherchildren's performance was higher than theirevaluation of their own performance, whereasthe mastery-oriented children believed that


Table 3Means and Multiple Comparisons

Mean

After success

Measure

Evaluation ofperformance

(1-10)Evaluation of

"most" chil-dren (1-10)

Expectancy offuture suc-cess (1-10)

How many rightout of 15

How many right(answer = 8)

Attributions(total = 35)

It was easyI was luckyI am good

at thisI am smart

Help-less

6.39

7.25

6.04

7.46

5.46

7.8913.32

8.045.71

Mas-teryori-

ented

7.00

5.89

8.75

13.43

7.57

4.897.32

8.3914.46

After failure

Help-less

5.18

6.61

4.36

6.07

5.14

13.9614.50

3.003.50

Mas-teryori-

ented

6.79

5.39

8.14

11.79

7.57

5.468.11

8.9612.14

p levels of multiple comparisons

Helplesscompared to

masteryoriented

Aftersuc-cess

ns

.01

.01

.01

.01

nsns

ns.01

Afterfailure

.01

ns

.01

.01

.01

.01ns

.05

.01

Changes fromsuccess

to failure

Help-less

.01

ns

.05

ns

ns

nsns

nsns

Mas-teryori-

ented

ns

ns

ns

.01

ns

nsns

nsns

most children would perform more poorlythan they had.) If being successful involvesperceiving that one is performing at a levelequal to or better than one's peers, it wouldseem to be more difficult for the helplesschild to be successful than it is for the mas-tery-oriented.

Were the two groups of children equallyaccurate in estimating the number of prob-lems they had solved? When asked followingthe eight success problems how many prob-lems they had solved successfully, the mas-tery-oriented children were quite accurate(M — 7.57) but the helpless children gavethemselves credit for significantly fewer suc-cesses (M = 5.46).

The responses of the helpless and mastery-oriented children following failure were alsocompared—their performance evaluation,their expectancy for future success, how manyproblems they expected to solve if given an

additional 15 problems similar to the successproblems, and their estimate of the numberof problems they had successfully solved.Although there was no difference between thetwo groups in how they rated their perform-ance before the failure, differences did appearon this variable after the introduction offailure: The helpless children judged theirperformance to be significantly poorer thanthe mastery-oriented judged theirs to be (M— 5.18 and M — 6.79, respectively). Signifi-cant group differences again occurred forexpectancy of future success (helpless =4.36; mastery-oriented = 8.14), how manythey expected to get right out of 15 (helpless= 6.07; mastery-oriented = 11.79), and howmany problems they had successfully solved(helpless = 5.14; mastery-oriented = 7.57).

The effect of failure on each group's percep-tion of success can be examined by comparingwithin-group responses made after only sue-


cess to responses made after failure. The help-less children showed a significantly lowerrating of their performance after versus be-fore failure, but the mastery-oriented did not.In addition, the ratings of helpless childrenreflected even more pessimism about theirfuture performance than they had prior tofailure, whereas the mastery-oriented tendedto display equally high expectations. Al-though the mastery-oriented children did ex-pect to get fewer right out of IS than themastery-oriented group had expected previ-ously, predominantly they still expected suc-cess (M = 11.79).

The pattern of attributions made by thetwo groups also differed substantially, as seenin Table 3. When asked to choose the relativeimportance of various attributions for suc-cess, helpless children did not tend to givethemselves as much credit for success: Evenprior to failure, the mastery-oriented childrenattributed their successes to their intelligencemore than did the helpless children. Followingfailure, this pattern of success attributionswas evidenced by general ability attributionsand in a trend toward task-specific attribu-tions. Moreover, with the introduction offailure, the helpless children attributed theirsuccess significantly more to the task's havingbeen easy than did the mastery-oriented chil-dren. Overall, then, the helpless childrendiscounted their successes as indicative ofability or as implying continued or futuresuccess, whereas mastery-oriented childrenstressed their ability as an important determi-nant of their success.

Do the two groups have equally accuratememories for their failures? The children wereasked, following four failures, to estimate thenumber of problems they had not been ableto solve successfully. The mastery-orientedchildren were fairly accurate in their percep-tion (M = 3.71), whereas the helpless chil-dren tended to overestimate their failures (M= 6.14). The difference between the groupswas highly significant, / (S4) = 4.91, p <.001.

When asked if they could find the solutionto one of the original eight success problemsif it were readministered, lOO^r of the mas-tery-oriented children said yes, whereas only65% of the helpless children thought they

could. Chi-square analysis on these datayielded a significant difference, x2 = 24.34,p < .01. It would appear that helpless chil-dren see so little predictive power in pastsuccess and trust their own ability so littlethat they do not expect to solve the sameproblems they had previously mastered.

Effects of repeated questioning. Previousresearch (Dweck & Gilliard, 1975) has indi-cated that asking repeatedly for expectancystatements changes children's responses—boththe expectancies they give and their actualperformance on the task. Since data wereobtained on one group of children who hadresponded to the set of questions twice, thesechildren were compared to those who an-swered once following failure only. A 2(group: helpless vs. mastery-oriented) X 2(sex) X 2 (race: black vs. white) X 2(questions asked once or twice) multivariateanalysis of variance was conducted witheight dependent variables. The main effectfor group was highly significant, F(8, 89) —30.543, p < .001, indicating that the helplesschildren responded in a distinctly differentfashion from the mastery-oriented children.The main effects for sex and race were notsignificant (p < .40 and p < .32, respec-tively). A significant F ratio, F(8, 89) =2.90, p < .01, was found for the number oftimes the questions were asked. That is, thegroup that responded to questions about theirperformance only after failure differed fromthe group that answered both after successand after failure. The subsequent univariateanalyses suggested that the major effects weredue to expectancy of future success (thoseresponding once = 6.25; those respondingtwice = 5.39), /?(!, 96) = 4.08, p < .05,and evaluation of performance (those respond-ing once = 5.98; those responding twice =4.91), F(l, 96) = 9.18, p < .01. One mightplausibly have hypothesized that respondingafter success would yield higher expectanciesand evaluations on the second set of ques-tions by virtue of calling children's attentionto their successes. However, the fact that onboth questions the means were lower for thegroup who responded twice suggested thatrepeated questioning resulted in the child'srevising his or her first response to reflectthe failures to a greater degree. It may be


that the original questioning heightened thesalience of the subsequent failures.

Discussion

Past findings have suggested that successdoes not act as a buffer against the negativeeffects of failure for helpless children, as itseems to for the mastery-oriented. The pre-sent study examined the manner in which thehelpless child's perception of success differsfrom that of the mastery-oriented child. Over-all, the results show that if there is a way todevalue one's present performance or to bepessimistic about one's future performance,the helpless children are likely to make useof it. Indeed, they do not even have to ex-perience a negative outcome for this tendencyto display itself. In sharp contrast, the mas-tery-oriented are realistically optimistic whenthey are succeeding and are surprisinglyundaunted by failure.

First, the helpless children do not recog-nize or remember the actual extent of theirsuccess. Relative to the mastery-oriented,they underestimate the number of problemsthey have correctly solved. Thus, from thebeginning, there is "less" success to act as abuffer. Additionally, the success that thehelpless children do acknowledge is not "assuccessful" as that experienced by the mas-tery-oriented. Although both helpless andmastery-oriented children, after experiencingsuccess only, rated their performance aboutthe same—somewhat above average—the sim-ilarity ended there. Their estimates of howwell most children would do on the same taskindicated that the helpless children expectedmost children to do quite well, whereas mas-tery-oriented children expected that the per-formance of most children would be closer tomerely adequate. This suggests either thatthe helpless children view most other childrenas more able or that they are using a differ-ent comparison group for themselves. If thehelpless children are comparing themselvesnot to the "average" but to a more proficientgroup, it would take a better performance forthem to be "successful" than it would for themastery-oriented. For example, if the helplesschildren consistently compared themselves tothe very best children in the class, then it

would be more difficult to achieve success bysocial comparison standards than if they com-pared themselves to average classmates. Thispossibility might be a fruitful avenue forfuture investigation.

A second major difference is that helplesschildren do not view their present success aspredictive of future success. Their expectancyfor future performance was only a littleabove the midpoint despite their eight con-secutive successes. When asked to make aspecific prediction about how many problemsthey believed they could solve if given 15more of the same type, they expected to getonly about half right. What mediates theirperception that successes are not replicable?The extent to which children see their suc-cesses as reflecting ability is likely to be oneof the factors involved. When asked to makeattributions for their success, mastery-ori-ented children credited their ability ("I amsmart"), an enduring and general quality,significantly more than helpless children.Moreover, following failure, helpless childrendiscounted their past successes as having beendue to task ease to a greater extent than themastery-oriented. The mastery-oriented chil-dren continued to view their success as re-flecting ability and therefore predictive offuture performance.

After the advent of failure, helpless chil-dren appeared to reevaluate (i.e., devalueeven more) their prior success in light oftheir current failure in other ways as well.The major changes were in their loweredevaluation of their performance and in atrend toward lowered expectancy of futuresuccess. Some change in performance evalua-tion would be realistic, since children werenow evaluating a performance that includedboth success and failure. However, the help-less children's plunge in expectancy for futuresuccess coupled with their previously low ex-pectancy would suggest that although successis not perceived to be predictive of futureperformance, failure certainly is. It wouldappear that helpless children view failure asmore "diagnostic" of their level of ability,whereas mastery-oriented children view suc-cess as more diagnostic (see Trope & Brick-man, 1975). Additionally, the helpless chil-dren appeared to be so oriented toward fail-


ure that they inflated the actual amount offailure they received. The mastery-orientedchildren were accurate in their estimate ofthe number of problems they had missed.

One of the most striking examples of thetendency of helpless children to view failure aspredictive and to discount their successes istheir response to the question of whether theycould solve one of the success problems if itwere presented again. All of the mastery-ori-ented children believed they could redo theproblem successfully, whereas only two thirdsof the helpless children were willing to pre-dict that they would be successful a secondtime. Not only is success not predictive offuture success on similar problems, but anumber of the helpless children are not sureit is even predictive of success on the sameproblems.

The tendency of helpless children to dis-count success cannot help having adverseeffects on their persistence on a task in theface of obstacles (Feather, 1965; Zajonc &Brickman, 1969). Why do the helpless chil-dren discount their success to such an extent?Could this inattention to present successserve an adaptive function for the helplesschildren? Perhaps the anticipation of failureprevents the helpless children from beingeven more adversely affected by failure whenit does occur. That is, if, following successes,they allowed themselves to believe that theyhad high ability but were still prone to seefailure as indicating a lack of ability, thenthe occurrence of failure might have evengreater negative impact. Indeed, Dweck(1975) found that a few helpless childrenactually appeared more sensitive to failureafter they had experienced prolonged success.If this is true, it is possible that if suchchildren come to believe in their ability, theymight be less willing to risk a failure thatindicated a lack of ability. On the other hand,there is also the possibility that making suc-cess salient to helpless children by teachingthem to identify success and to view it asindicative of ability might prove helpful. Forexample, just as the retraining of failureattributions led to improved performance onthe part of helpless children (Dweck, 1975),it might be useful as well to retrain helplesschildren to attribute success to more stable

and predictive factors such as ability. Theprecise relationship between perceptions ofsuccess and reactions to failure needs furtherexploration to clarify these possibilities.

Yet another important question relates tothe public versus private nature of the child'sperformance evaluations and predictions. Inthis study, the children made their responsesverbally to the experimenter. This may havemade the situation more evaluative, made thehelpless children focus even more on failure,and made them more cautious in their predic-tions. It is unlikely, however, that the publicnature of the responses in itself accounts forthe results. For example, how would this ac-count for the helpless children's high evalua-tion of their peers? Also, since the number ofsuccesses and failures can be verified easilyby the experimenter, the demand should befor accurate estimates. Yet the helpless chil-dren underestimated their successes and over-estimated their failures. A related and evenmore interesting question involves the degreeto which the helpless child's behavior dependson the performance situation's being a publicone. That is, would the helpless child be aslikely to discount success or react to failureas adversely if no one were monitoring theresults? Is it helpless children's own doubtsabout their abilities that cause their deteri-oration or is it their knowledge that othersmay witness or have already witnessed theirfailures (i.e., their lack of ability)? Thisquestion is of general importance, since al-most all academic situations involve publicevaluation of some sort.

One might argue that the mastery-orientedchildren are unrealistic in their assessment oftheir performance, since they do not revisetheir evaluation after a series of failure trials.However, this does not seem to be the casefor several reasons. First, they are accurateabout the number of problems solved and notsolved. Second, it appeared in an earlierstudy (Diener & Dweck, 1978) that mastery-oriented children did not define the "wrong"feedback as failure but only as informationsuggesting that they needed to alter theirstrategy. Third, they do appear to appreciatethat they might not be able to solve quite asmany problems as they had previously esti-mated. They also appear to be able to tolerate


a lapse from perfect performance (which isto be expected in most performance situa-tions) and still view themselves as successful.

The goal of the present investigation wasto broaden our understanding of learned help-lessness by examining perceptions of success.The results indicate that just as the mastery-oriented children do not seem to define them-selves as failing on the failure trials, thehelpless children do not define themselves assucceeding on the success trials.

References

Crandall, V. C., Katkovsky, W., & Crandall, V. J.Children's belief in their own control of reinforce-ment in intellectual academic achievement situa-tions. Child Development, 1965, 36, 91-109.

Diener, C. I., & Dweck, C. S. An analysis of learnedhelplessness: Continuous changes in performance,strategy, and achievement cognitions followingfailure. Journal of Personality and Social Psychol-ogy, 1978,3(5,451-462.

Dweck, C. S. The role of expectations and attribu-tions in the alleviation of learned helplessness.Journal of Personality and Social Psychology,1975, 31, 674-685.

Dweck, C. S., & Bush, E. S. Sex differences inlearned helplessness: I. Differential debilitationwith peer and adult evaluators. DevelopmentalPsychology, 1976, 12, 147-156.

Dweck, C. S., & Gilliard, D. Expectancy statementsas determinants of reactions to failure: Sex dif-ferences in persistence and expectancy change.Journal of Personality and Social Psychology,1975, 32, 1077-1084.

Dweck, C. S., & Licht, B. G. Learned helplessnessand intellectual achievement. In J. Garber & M.Seligman (Eds.), Human helplessness: Theory andapplication. New York: Academic Press, in press.

Dweck, C. S., & Reppucci, N. D. Learned helpless-ness and reinforcement responsibility in children.

Journal of Personality and Social Psychology,1973, 25, 109-116.

Feather, N. T. Performance at a difficult task inrelation to initial expectation of success, test anxi-ety, and need achievement. Journal of Personality,1965, 33, 200-217.

Finn, G. D. Multivariance. Ann Arbor, Mich.: Na-tional Educational Resources, 1972.

Floor, L., & Rosen, M. Investigating the phenome-non of helplessness in mentally retarded adults.American Journal of Mental Deficiency, 1975, 79,565-572.

Gholson, B., Levine, M., & Phillips, S. Hypotheses,strategies, and stereotypes in discrimination learn-ing. Journal of Experimental Child Psychology,1972, 13, 423-446.

Kirk, R. E. Experimental design: Procedures for thebehavioral sciences. Belmont, Calif.: Wadsworth,1968.

Levine, M. Hypothesis behavior by humans duringdiscrimination learning. Journal of ExperimentalPsychology, 1966, 71, 331-338.

Nicholls, J. G. Causal attributions and otherachievement-related cognitions: Effects of taskoutcome, attainment value, and sex. Journal ofPersonality and Social Psychology, 1975, 31, 379-389.

Seligman, M. E. P., Maier, S. F., & Geer, J. H.Alleviation of learned helplessness in the dog.Journal of Abnormal Psychology, 1968, 73, 256-272.

Trope, Y., & Brickman, P. Difficulty and diagnosti-city as determinants of choice among tasks. Jour-nal of Personality and Social Psychology, 1975,31, 918-925.

Weiner, B. Theories of motivation. Chicago:Markham, 1972.

Weiner, B. Achievement motivation and attributiontheory. Morristown, N.J.: General LearningPress, 1974.

Zajonc, R. B., & Brickman, P. Expectancy and feed-back as independent factors in task performance.Journal of Personality and Social Psychology,1969, 11, 148-156.

Received August 27, 1979 •

diener y dweck (1980). an analysis of learned ii the processing of success

Documents