Back to Journal Cover Page


CURRENT RESEARCH IN SOCIAL PSYCHOLOGY


Volume 6, Number 15
Submitted: January 18, 2001
Resubmitted: September 10, 2001
Accepted: September 10, 2001
Publication date: September 17, 2001

SELF-HANDICAPPING AS A METHOD OF SELF-PRESENTATION: AN ANALYSIS OF COSTS AND BENEFITS

Maurice J. Levesque
University of Connecticut

Charles A. Lowe
University of Connecticut

Catherine Mendenhall
Amherst College

ABSTRACT

To examine the self-presentational effectiveness of self-handicapping, participants read a survey supposedly completed by a student concerning activity before an exam. The handicapping manipulation was accomplished by varying the survey respondent's narrative description of activity the night before the exam. The respondent succeeded or failed the exam. Participants made attributions for the outcome and rated the respondent’s competence, responsibility, and sociability. Discounting of ability with failure emerged for the handicap present condition, but this benefit was offset by lack of effort attributions. Handicappers were seen as less generally competent, as less responsible, and as more sociable than non-handicappers.

[221]
---------------
[222]

INTRODUCTION

As originally defined, self-handicapping is "any action or choice of performance setting that enhances the opportunity to externalize (or excuse) failure and to internalize (reasonably accept credit for) success" (Berglas & Jones, 1978, p. 406). Rather than risk failure and the corresponding inferences of low ability, individuals who are uncertain about the likelihood of success on a given task may alter the conditions of the performance so as to provide a reasonable explanation for failure that does not reflect their competence. If the individual is successful on the task, the impediment represents an obstacle, which was overcome by superior ability. Thus, to protect against inferences of low ability following failure and to enhance inferences of high ability following success, self-handicapping capitalizes on the attribution principles of discounting and augmentation (Kelly, 1971). Considerable evidence exists suggesting the ubiquity of self-handicapping and the conditions likely to elicit self-handicapping have been identified (for a review see Higgins, Snyder, & Berglas, 1990).

Although self-handicapping behavior is well documented, less attention has been devoted to whether handicaps accomplish the self-protective and enhancing functions assumed to motivate the behavior. A few studies of handicappers' self-attributions for performance have found that handicappers do discount ability as a cause of failure (Isleib, Vuchinich, & Tucker, 1988; Rhodewalt, Morf, Hazlett, & Fairfield, 1991; Tice, 1991). However, most studies have failed to obtain augmentation (e.g., Mayerson & Rhodewalt, 1988 but see Rhodewalt et al., 1991 for an exception). Overall, it appears that handicaps function well to protect against self-judgments of low ability as a result of failure, but function less well for many individuals in terms of enhancing self-judgments of high ability following success.

Although the influence of self-handicapping on self-evaluations represents an important area of inquiry, handicapping seldom occurs in a social vacuum. Rather, the presence of a handicap is often reported to someone and others occasionally observe the acquisition of a handicap. Since the original conceptual and empirical work on self-handicapping (Berglas & Jones, 1978; Jones & Berglas, 1978), research has explored the degree to which self-presentational concerns motivate self-handicapping. Although Berglas and Jones (1978) did not obtain increased levels of self-handicapping under public conditions, Kolditz and Arkin (1982), using a more refined public-private manipulation, did observe an increased tendency to handicap under public as opposed to private conditions. Other research suggests that an individual is more likely to claim a handicap when observers are unaware that he or she has failed on a similar previous task (Baumgardner et al., 1985). These results suggest that self-presentational concerns may contribute to the tendency to engage in self-handicapping (Snyder, 1990). Given that possibility, discerning the effectiveness of self-handicapping as a self-presentational strategy is essential to a more complete understanding of self-handicapping (Snyder, 1990).

[222]
---------------
[223]

Effectiveness of Self-handicapping as Self-presentation
The few studies that have assessed observers’ attributions regarding a handicapped performance have focused primarily on attributions of ability. Arkin and Baumgardner (1985), for example, found that ability was discounted as a cause of failure in the presence of a handicap, but, for success, no evidence emerged to suggest that ability was augmented. This general pattern of results has been replicated by studies using different outcomes and handicaps (Smith & Strube, 1991; Springston & Chafe, 1987). Similar to findings regarding handicappers' self-evaluations, observers appear more willing to accept the handicap as a likely cause of failure than to ascribe greater ability to individuals who succeed despite a handicap.

Some studies have included assessments that reflect audience judgments of the handicapper's dispositional competence. Arkin and Baumgardner (1985), for example, asked subjects to rate self-handicappers' dispositional competence and found that handicappers were rated as being less competent than non-handicappers regardless of performance outcome. Similarly, Springston and Chafe (1987) found that when asked to rate the self-handicapper’s future success, subjects expected that intentional handicappers would be less successful than either non-handicappers or accidental (unintentional) handicappers. By contrast, if predictions of future exam performance constitute a more general assessment of competence, the results of Luginbuhl and Palmer (1991) appear to contradict these findings. They found that observers predicted that self-handicappers would score better on a future exam than non-handicappers who had succeeded or failed on the first exam. However, these contradictory results might be explained by the nature of the handicaps. Although differences in the type of handicaps used as stimuli are not likely to affect attributions about the causes of a specific outcome (assuming relatively equal performance inhibitory effects), observers may rely on the specific nature of the handicap to infer its long-term impact. Thus, some handicaps (e.g., alcohol consumption [Arkin & Baumgardner, 1985; Springston & Chafe, 1987]) may be viewed as indicating a stable characteristic of the individual that is likely to affect future performance and therefore indicative of low dispositional competence. Other handicaps (e.g., going to the movies [Luginbuhl & Palmer, 1991]), however, may be seen as indicating more temporary states that are less likely to affect future performance.

Other research has examined observers' general evaluative judgments regarding handicappers and non-handicappers. Smith and Strube (1991) found that on a global measure of impression favorability, audiences thought less highly of self-handicappers, and Springston and Chafe (1987) found that intentional handicappers were liked less than unintentional handicappers or non-handicappers. Although the specific nature of the handicaps employed in these studies differed substantially, these results suggest that the benefits of handicapping in terms of attributions for a performance outcome may be offset by less favorable inferences of the handicapper's dispositional characteristics. Additionally they underscore the importance of measuring observers’ judgments of the dispositional characteristics of a handicapper as well as attributions regarding the causes of a specific outcome.

[223]
---------------
[224]

Present Study
The present study was designed to extend previous research on the self-presentational effectiveness of self-handicapping and to determine the relative benefits and costs associated with self-handicapping. Rather than employ a method that highlights the fictional nature of the handicapping story, as in the typical vignette methodology, individuals in the present study reviewed surveys supposedly completed by other students. The experimental manipulation of handicapping was accomplished in the reports that survey respondents gave concerning their activities the night before an exam. Alcohol consumption was chosen as the self-handicapping behavior on the basis of previous research documenting the willingness of individuals to handicap by consuming alcohol (Higgins & Harris, 1988; Tucker, Vuchinich, & Sobell, 1981) and Jones and Berglas' (1978) proposition that alcohol consumption is a common handicap. The present study also included a no handicap condition.

We propose that audiences who are aware of a handicapped performance must resolve two causal dilemmas. First, audiences must decide what caused the performance outcome and to what extent the handicap facilitated a failure or inhibited a success. Second, the audience must evaluate the handicapping behavior itself for which the principal issue becomes the extent to which the handicap indicates particular dispositional qualities.

In terms of attributions for the outcome of a handicapped performance (i.e., the first causal dilemma), the handicap is only one of many possible causes, and audiences judge the contribution of the handicap in comparison to other causes. For a successful performance, the audience evaluates whether a particular factor(s) was sufficient or necessary to overcome the inhibitory effects of the handicap. For a failed performance, the audience evaluates whether the handicap constitutes a sufficient explanation for the failure or whether other factors were equally or even more likely causes. Given that understanding a particular outcome often requires reference to multiple factors, this study sought to correct a shortcoming in previous research in which only ability attributions were assessed. It is possible, for example, that augmentation effects are difficult to obtain because audiences augment effort, not ability. In the same manner, ability may be discounted for failure in the presence of a handicap, but other internal causes may not be discounted (e.g., failure caused by low motivation). Thus, we assessed attributions to several possible causes.

Based on previous research (Arkin & Baumgardner, 1985; Luginbuhl & Palmer, 1991; Smith & Strube, 1991), evidence of discounting of ability for failure in the presence of a handicap was expected. Augmentation of ability for success in the presence of a handicap was not expected. Given the dearth of research concerning other causal factors, we made no a priori hypotheses concerning the role of factors such as effort. With respect to more dispositional judgments of ability, the present study included both predictions of future exam performance and ratings of the handicapper's general competence (e.g., intelligence). Because the handicap employed was the same one used by Arkin and Baumgardner (1985), it was expected that the present results would be consistent with their findings; such that, handicappers will be rated more negatively than non-handicappers on both predictions of future exam performance and dispositional competence.

[224]
---------------
[225]

Observers’ resolution of the second causal dilemma was expected to influence other general dispositional judgments. Trait scales were employed to tap two basic evaluative dimensions, namely responsibility and sociability. A pattern identical to that predicted for dispositional competence was anticipated for ratings of responsibility. Specifically, it was expected that handicappers would be rated less favorably than non-handicappers on responsibility related traits regardless of outcome. For sociability related traits, it was expected that the handicappers would be rated in a similar or even more positive manner than non-handicappers, a result likely given the nature of the handicap. Because these judgments are based on the handicapping behavior itself, performance outcome was not expected to influence these judgments.

METHOD

Participants
Participants were 83 introductory psychology students (37 males and 46 females) who participated as partial fulfillment of a course requirement. The design of the study was a 2 (outcome: success vs. failure) x 2 (handicap: self-handicapped vs. no self-handicap) between-subjects factorial.

Procedure
At each experimental session, individuals were asked to select any one of a number of experimental packets. Each packet included general instructions, the cover story, a fictitious student survey, and two sealed envelopes. The experimenter explained that the study was part of a previous experiment designed to assess how behavior the night before an exam might affect exam performance. Participants were further informed that introductory psychology students in a previous year had volunteered to complete a survey, immediately before taking an exam, regarding their preparation for the exam. The experimenter stated that a decision had been made to have peers evaluate the students and their performance and that the packet they had chosen contained a survey completed by one of the students, a grade report for that student, an evaluation questionnaire and a memory quiz. The experimenter stressed the importance of carefully reviewing the student survey.

Individuals proceeded by reading the general instructions, the cover story and the student survey. The general instructions stated that, after reviewing the survey, individuals could open envelope one which contained the student's grade and the evaluation questionnaire. At the end of the evaluation questionnaire, individuals were instructed to return all forms to envelope one and to complete the memory quiz and demographic form in envelope two. After all participants for the experimental session closed envelope two, they were debriefed and thanked for their participation.

[225]
---------------
[226]

Materials
An experimental assistant who varied his handwriting and writing implement completed the fictitious student surveys. To support the cover story, the name and social security number sections of the surveys were blacked out, supposedly to protect the anonymity of the respondent. Regardless of experimental condition, the survey indicated that the respondent was a second semester student with a 2.4 GPA intending to major in psychology which was important to him/her, had completed some of the required reading when it was assigned, and had studied approximately 4-5 hours for the exam. This information was designed to show that the respondent was an average student academically and that the exam was important. The final question on the survey concerned the respondent's activities the night before the exam. The responses to this question shared a common stem indicating that the student lived off-campus and had gone to the library to study. After talking with some friends, he or she had studied for four hours, taking a few short breaks, and had left the library at 11 p.m. Further, the respondent reported that it was snowing that evening and the roads were messy. The remainder of the response depended on the experimental condition. In the no handicap condition, the survey respondent reported driving slowly, arriving home and going to bed at midnight. In the handicap condition, the survey respondent reported driving home slowly and, after arriving home, going to a nearby party with his or her roommate. He or she reported drinking a few too many beers, being helped home, and going to bed at 3 a.m.

The exam grade report indicated that the respondent received either an A- (success) or a D (failure) and that the average grade on the exam had been a C+. To support the cover story, both the student survey and the grade report had the same respondent number, which subjects were instructed to verify to ensure that they received the correct grade report.

The first page of the evaluation questionnaire (dependent measures) asked individuals to rate on 7-point scales (0 [not at all] to 6 [a great deal]) the degree to which the following factors had affected the respondent's performance: intellectual ability, effort during the exam, effort in studying before the exam, good or bad luck, amount of preparation, and the difficulty of the exam. Participants were then asked to form an impression of what the respondent might be like and to indicate their impression on the following 7-point bipolar trait scales: honest-dishonest, competent-incompetent, sociable-reserved, happy-sad, active-passive, responsible-irresponsible, dependable-undependable, intelligent-unintelligent, reliable-unreliable, and likely to succeed-not likely to succeed. A factor analysis performed on the trait ratings indicated two general factors. One factor consisted of ratings of dependable, responsible, reliable, intelligent, competent and likely to succeed. The other factor labeled Sociability consisted of ratings of social, active, and happy. To facilitate comparison with previous research and based on a priori assumptions, the first factor was bifurcated to form a Competence factor (i.e., intelligent, competent, and likely to succeed) and a Responsibility factor (i.e., responsible, dependable and reliable).

[226]
---------------
[227]

The final page of the questionnaire asked participants to predict the respondent's grade on the next exam in the same course. In addition, a manipulation check question asked individuals to rate how honest they thought the respondent's answers to the survey had been and to write down anything they thought was less than honest. The memory test asked seven questions regarding the respondent’s survey answers.

RESULTS

All analyses were conducted using 2 (success vs. failure) x 2 (handicap vs. no handicap) ANOVAs with outcome and handicap as between-subjects factors. All post hoc analyses were done using Tukey tests.

Manipulation Checks
To assess the degree to which individuals attended to the handicap and grade manipulations, memory test responses to two questions assessing these manipulations were coded for correctness (0=wrong, 1=partially correct and 2=correct). Responses were summed and analyzed but no significant differences emerged. The overall performance on the memory test (M = 13.11 of a possible 14) indicated nearly perfect recollection of the survey information.

Responses to the question asking subjects whether they believed the respondent revealed a significant interaction between outcome and handicap, F (1, 79) = 10.28, p < .01. Post-hoc analyses revealed that non-handicappers were seen as more honest when they succeeded (M = 6.52, SD = .51) as compared to when they failed (M = 5.48, SD = 1.57). A self-handicapper was perceived to be honest regardless of outcome (success, M = 6.20, SD = .83; failure, M = 6.52, SD = .60). For the failure condition, handicappers were rated as more honest than non-handicappers, but no difference emerged for the success condition. Review of the open-ended responses corresponding to this question showed that subjects questioned the reported amount of studying rather than the information pertaining to the handicapping manipulation. Overall, ratings of honesty indicate that subjects generally thought the respondent had been honest.

Outcome Attributions
Four factors (i.e., ability, effort, luck, and task difficulty) were used to assess beliefs about the "causes" of the student's exam performance.

Ability
A significant main effect for outcome, F (1, 75) = 35.56, p < .001, and a significant effect for handicap, F (1, 75) = 4.39, p < .05, emerged for ratings of the importance of ability as a cause for the outcome. Ability was rated as more important for success (M = 4.68, SD = 1.06) than for failure (M = 3.43, SD = 1.29) and as more important in determining the outcome of non-handicappers (M = 4.29, SD = 1.26) than of handicappers (M = 3.80, SD = 1.38).

[227]
---------------
[228]

In addition, a significant interaction between handicap and outcome emerged for ratings of ability, F (1, 75) = 6.52, p < .05. Specifically, ability was rated as more important for success than for failure in the handicap present condition, but no such discrimination occurred in the no handicap condition. Additionally, ability was seen as less important as a cause of a handicapper’s failure than a non-handicapper's failure. Thus, there was evidence for discounting ability for failure in the presence of a handicap; however, no support was found for the augmentation of ability when success occurs in the presence of a handicap.

Table 1. Mean Ability Ratings as a Function of Self-handicapping and Outcome

Outcome Non-Handicapped Self-Handicapped
  M SD M SD
Success 4.62 (a) 1.16 4.75 (a) .97
Failure 3.95 (b) 1.28 2.90 (c) 1.09

Note. Means with different parenthetical letters differ at p < .05.

Effort
Ratings of effort before the exam and degree of preparation were highly correlated, r (81) = .68, p < .001, and were averaged to create a single measure of effort. A significant main effect emerged for outcome, F (1, 79) = 4.75, p < .05, such that effort was perceived to be more important in determining failure (M = 4.48, SD = 1.41) than success (M = 3.87, SD = 1.10). No other significant effects emerged for ratings of effort.

Luck
Ratings of the importance of luck revealed a significant main effect for outcome, F (1, 74) = 35.30, p < .001, such that luck was rated as more important to success (M = 3.28, SD = 1.77) than to failure (M = 1.36, SD = 1.34). The interaction between handicap and outcome was also significant, F (1, 74) = 6.88, p < .01. Luck was seen as more important for success than for the failure of both handicappers and non-handicappers. Within outcome, good luck was seen as more important for the success of a handicapper than for the success of a non-handicapper. By contrast, bad luck was not rated as an important cause of failure for either the handicapper or non-handicapper.

[228]
---------------
[229]

Table 2. Mean Ratings of Luck as a Function of Self-handicapping and Outcome

Outcome Non-Handicapped   Self-Handicapped  
  M SD M SD
Success 2.57 (b) 1.40 4.05 (a) 1.84
Failure 1.52 (c) 1.44 1.19 (c) 1.25

Note. Means with different parenthetical letters differ at p < .05.

Task Difficulty
Only a main effect for outcome was obtained for ratings of task difficulty, F (1, 78) = 4.68, p < .05. Task difficulty was seen as more important for failure (M = 3.93, SD = 1.22) than for success (M = 3.38, SD = 1.10).

General Competence
Predicted Future Grade
Participants’ predictions about the respondent's next exam grade were first subtracted from the scale value of the manipulated grade to obtain a difference measure sensitive to predictions of improvement (positive values indicating a predicted grade higher than the manipulated grade) or decline (negative values indicating a predicted grade lower than the manipulated grade). Only a significant main effect for outcome was obtained, F (1, 75) = 293.37, p < .001. Participants expected the performance of successful respondents to decline somewhat (M = -1.07, SD = -.93), while unsuccessful respondents were expected to show substantial improvement (M = 3.17, SD = 1.30). We had anticipated a self-handicapping effect for predicted future grade such that participants would expect handicappers to show less improvement on a future exam. However, contrary to our hypothesis, which was based on the results of Arkin and Baumgardner (1985), no such effect emerged; handicappers were not expected to show less improvement on a future exam.

Dispositional Competence
Analysis of competence traits ratings (e.g., intelligent) yielded significant effects for outcome, F (1, 79) = 75.19, p < .001, and handicap, F (1, 79) = 9.30, p < .01, but the interaction of these factors was not significant, F < .3, ns. Successful respondents were rated as more competent than were unsuccessful respondents (M = 5.82, SD = .79 and M = 4.11, SD = 1.06, respectively), and non-handicappers (M = 5.26, SD = 1.16) were seen as more competent than handicappers (M = 4.64, SD = 1.30). These results lend support to the hypothesis that handicapping exerts a negative influence on inferences of general competence.

[229]
---------------
[230]

General Evaluations
Responsibility
Consistent with the analysis of Competence, significant effects for outcome, F (1, 79) = 56.12, p < .001, and handicap, F (1, 79) = 41.82, p < .001, were obtained for ratings of Responsibility. Successful respondents were rated as more responsible than unsuccessful respondents (M = 5.02, SD = 1.24 and M = 3.49, SD = 1.01, respectively), and non-handicappers (M = 4.90, SD = 1.24) were seen as more responsible than handicappers (M = 3.58, SD = 1.15). These findings support the hypothesis that handicapping may be a costly method of self-presentation.

Sociability
Only the handicap effect emerged as significant for ratings of Sociability, F (1, 79) = 17.47, p < .001. The handicapper (M = 6.10, SD = .69) was rated as more sociable than the non-handicapper (M = 5.29, SD = 1.03). This benefit for inferences of sociability likely is derived from the stereotype of someone who "parties," especially when he/she chooses to "party" the night before an exam.

DISCUSSION

The present study was designed to determine the effectiveness of self-handicapping as a method of self-presentation; specifically, do individuals that are aware of a handicapped performance view the self-handicapper more positively with success and less negatively with failure than they do an individual achieving the same outcome who has not self-handicapped? As expected, the results suggest that an adequate answer to the effectiveness question depends on the criterion used to assess effectiveness. Each criterion used in this study, specifically, ability attributions, other attributions for the performance outcome, and dispositional judgments, is discussed as it relates to the question of effectiveness.

Inferences of Ability as a Cause of Success/Failure
To date, the most extensive evidence regarding the self-presentational effectiveness of self-handicapping has concerned attributions of ability (e.g., Arkin & Baumgardner, 1985; Luginbuhl & Palmer, 1991), reflecting the original conceptualization of self-handicapping, which emphasized that self-handicapping was motivated by a desire to avoid inferences of low ability with failure and to heighten inferences of ability with success. Consistent with previous research, the present results show that in the presence of a handicap, ability is discounted as a cause of failure (e.g., Smith & Strube, 1991). However, research on the augmentation of ability inferences with successful self-handicapped performance has yielded mixed results (Arkin & Baumgardner, 1985; Luginbuhl & Palmer, 1991) and the present data fail to support a clear augmentation effect. Clearly, augmentation, if it occurs, is a weaker counterpart to the discounting effect. Thus, with respect to specific performance ability inferences, handicaps appear to partially achieve the self-protective function "intended" by the self-handicapper.

[230]
---------------
[231]

Configuration of Attributions
The present results suggest that the benefit of self-handicapping in terms of ability inferences may be compromised, however, by the overall pattern of attributions for a self-handicapper’s performance. Specifically, the means for inferences of effort, task difficulty and luck reveal some general patterns that question the efficacy of self-handicapping. The success or failure of a non-handicapper was generally seen as the result of multiple causes and effort was cited as affecting success or failure equally. The pattern that emerges for the handicap condition appears to be somewhat clearer with respect to the importance of various causal factors. Ability and luck were cited as more important factors for success than for failure. Although ability was not credited for the handicapper's failure, effort and task difficulty were seen as likely causes.

Although no specific hypotheses were made concerning the pattern of attributions, these results do support the more general hypothesis that inferences regarding the role of causal factors other than ability would compromise the generally beneficial effect that self-handicapping has with respect to ability attributions. For perceptions of handicappers, the results suggest that any positive ability attributions for success are diluted by the contribution of luck as causing the observed success. Moreover, the benefit of discounting ability as an explanation for failure is offset by the less favorable ratings of the individual's effort in preparing for the exam. Thus, lack of ability may not cause the handicapper's failure but lack of motivation and effort, factors that are internal to the handicapper, were seen as strongly contributing to failure.

Thus, the benefits of self-handicapping with respect to protecting against inferences of low ability appear to be balanced by inferences of low motivation and effort. Consequently, if handicappers are generally concerned with observers' inferences regarding personal causal forces (i.e., ability and effort), handicapping may forestall one type of inference (ability) but not the other (effort). Interestingly, the presence of a handicap shifted observers' emphasis to effort demonstrating that, although all subjects were presented with the same effort relevant information, effort information is malleable and can be interpreted as either sufficient or insufficient depending on the outcome and other factors.

Dispositional Judgments
The results strongly suggest that the benefits of self-handicapping in terms of discounting ability for failure are limited and that self-handicapping may have more costly consequences. Specifically, handicappers may preclude inferences of inability to explain a specific failure, however, this benefit is lost when observers' judgments regarding dispositional competence are considered. In contrast to non-handicapped persons, the handicapper is seen as irresponsible and as less competent. These negative inferences are consistent with much of the existing literature on observers' judgments of self-handicappers (e.g., Springston & Chafe, 1987).

[231]
---------------
[232]

The negative dispositional judgments likely result from observers' analysis of the causes for the handicapping behavior itself. To the extent that the handicapping is perceived to be intentional and raises questions about particular aspects of the handicapper's character (perhaps because observers infer that the behavior reflects a pattern of similar behavior), the handicapper is perceived in a negative way. It remains for future research to consider the consequences of claiming a handicap that does not involve intentional behavior on the part of the handicapper. For example, it may be that when the handicap is perceived to be unintentional or when observers believe that the handicapping is an isolated, non-repetitive behavior, observers may suspend their tendency to infer negative dispositions.

The only general "benefit" of handicapping seems to be the perception of sociability. Handicappers are seen to be more sociable than non-handicappers likely because of the specific handicap employed in this study (alcohol consumption at a party) or because any "handicap" suggesting a preference for spending time with others as opposed to preparing for a task activates observers' stereotype of a "happy-go-lucky" incompetent person who does not try to avoid failure and who lacks the competence to succeed.

The results for dispositional judgments raise an important interpretational problem. Are the dispositional inferences regarding the handicapper based on excessive alcohol consumption only or on that behavior in context (i.e., occurring the night before an exam)? It is possible that the observed effects for dispositional inferences reflect the simple influence of excessive alcohol consumption. Specifically, individuals perceive someone who drinks to excess as less responsible, less competent, and more social than a person who does not consume alcohol. However, we suggest that knowledge of an important evaluative performance the next day would be perceived as a reason to avoid excessive alcohol consumption. The fact that a self-handicapper chooses to drink despite the inhibitory influence of the exam likely augments dispositional judgments of irresponsibility, incompetence and, in this case, sociability. This augmentation effect cannot be tested with the present data, but, when possible, future research should evaluate perceptions of the behavior used as a self-handicap in the absence of an evaluative performance to address this concern. This interpretational concern notwithstanding, it remains clear that individuals choosing to self-handicap with alcohol are likely to be viewed as relatively irresponsible and incompetent, and, under certain circumstances, they may also be perceived as sociable.

[232]
---------------
[233]

Future Research
These present findings raise several other important issues relevant to future research on self-handicapping as self-presentation. First, given that self-handicapping encompasses a wide variety of behavior (e.g., Higgins, Snyder, & Berglas, 1990), a principled scheme of categorizing types of self-handicaps is needed to allow for specific predictions of self-presentational effectiveness with respect to particular judgments. Because self-handicapping as a method of self-presentation rests on attributional principles, the scheme developed by Weiner and his colleagues (e.g., Weiner, 1992; Weiner, Amirkhan, Folkes, & Verett, 1987) based on Weiner's (1979) classification of causal factors seems promising. Thus, the crucial dimension for classifying handicaps with respect to self-presentational effectiveness might be whether the handicap refers to forces that are controllable or uncontrollable by the handicapper.

Second, and related to the first issue, the skilled self-handicapper, aware that the controllability of the handicapping behavior may impact observers' judgments, is likely to attempt to present the handicap in such a way as to make it appear uncontrollable; that is, they will engage in presentational tactics (e.g., verbal statements) that stress the uncontrollable reasons for the existence of the handicap or claim handicaps that are by their very nature relatively uncontrollable. These additional self-presentational efforts may require that observers also consider the honesty of such claims. Thus, there is a need for additional research on self-handicappers’ communications to observers and on the factors influencing whether observers are persuaded by such communications. Moreover, the apparent costs of using self-handicapping as a self-presentational strategy raise other questions regarding the motivations underlying self-handicapping. Are self-handicappers aware of the self-presentational costs and do individuals self-handicap because the self-protective benefits generally outweigh the public image costs? These issues certainly deserve additional attention.

GENERAL CONCLUSION

Obviously an answer to the question of whether self-handicapping is an effective means of self-presentation is complicated. Self-handicapping does appear to benefit the handicapper with respect to attributions of ability, but that benefit is obtained at considerable cost in terms of attributions regarding effort. In the long-term, handicapping appears to be a costly strategy. The handicapper is perceived to be irresponsible and incompetent, although he or she is also seen to be sociable. A more complete understanding of the self-presentational effectiveness of self-handicapping, however, requires that future research establish the linkages between observers’ judgments and their behavior towards the handicapper.

[233]
---------------
[234]

REFERENCES

Arkin, R. M., and A. H. Baumgardner (1985). "When self-handicapping fails to serve a purpose: Impressions of the strategic procrastinator." Unpublished manuscript, University of Missouri, Columbia. Reported in Baumgardner, A. H., and R. M. Arkin (1987). Coping with the prospect of disapproval: Strategies and sequelae. Pp. 323-346. In Coping with negative life events: Clinical and social psychological perspectives, edited by C. R. Snyder and C. E. Ford. New York: Plenum.

Baumgardner, A. H., E. A. Lake, and R. M. Arkin (1985). Claiming mood as a self-handicap: The influence of spoiled and unspoiled public identities. Personality and Social Psychology Bulletin 11:349-357.

Berglas, S., and E. E. Jones (1978). Drug choice as a self-handicapping strategy in response to noncontingent success. Journal of Personality and Social Psychology 36:405-417.

Higgins, R. L., and R. N. Harris (1988). Strategic "alcohol" use: Drinking to self-handicap. Journal of Social and Clinical Psychology 6:191-202.

Higgins, R. L., C. R. Snyder, and S. Berglas (1990). Self-handicapping: The paradox that isn't. New York, NY:Plenum Press.

Isleib, R. A., R. E. Vuchinich, and J. A. Tucker (1988). Performance attributions and changes in self-esteem following self-handicapping with alcohol consumption. Journal of Social and Clinical Psychology 6:88-103.

Jones, E. E., and S. Berglas (1978). Control of attributions about the self through self-handicapping strategies: The appeal of alcohol and the role of underachievement. Personality and Social Psychology Bulletin 4:200-206.

Kelley, H. H (1971). Attribution in social interaction. New York:General Learning Press.

Kolditz, T. A., and R. M. Arkin (1982). An impression management interpretation of the self-handicapping strategy. Journal of Personality and Social Psychology 43:492-502.

[234]
---------------
[235]

Luginbuhl, J., and R. Palmer (1991). Impression management aspects of self-handicapping: Positive and negative effects. Personality and Social Psychology Bulletin 17:655-662.

Mayerson, N. H., and F. Rhodewalt (1988). The role of self-protective attributions in the experience of pain. Journal of Social and Clinical Psychology 6:203-218.

Rhodewalt, F., C. Morf, S. Hazlett, and M. Fairfield (1991). Self-handicapping: The role of discounting and augmentation in the preservation of self-esteem. Journal of Personality and Social Psychology 61:122-131.

Smith, D. S., and M. J. Strube (1991). Self-protective tendencies as moderators of self-handicapping impressions. Basic and Applied Social Psychology 12:63-80.

Snyder, C. R (1990). "Self-handicapping processes and sequelae: On the taking of a psychological dive." Pp. 107-150. In Self-handicapping: The paradox that isn't, edited by R. Higgins, C. R. Snyder, and S. Berglas. New York: Plenum Press.

Springston, F. J., and P. M. Chafe (1987). Impressions of fictional protagonists exhibiting self-handicapping behaviors. Paper presented at the Canadian Psychological Association Convention, Vancouver, B.C.

Tice, D. M (1991). Esteem protection or enhancement? Self-handicapping motives differ by trait self-esteem. Journal of Personality and Social Psychology 60:711-725.

Tucker, J. A., R. E. Vuchinich, and R. E. Sobell (1981). Alcohol consumption as a self-handicapping strategy. Journal of Abnormal Psychology 90:220-230.

Weiner, B (1979). A theory of motivation for some classroom experiences. Journal of Educational Psychology 69:506-511.

Weiner, B (1992). "Excuses in everyday interaction." Pp. 131-146. In Explaining one's self to others: Reason-giving in a social context, edited by M. L. McLaughlin, M. J. Cody, and S. J. Read. Hillsdale, NJ: Erlbaum.

[235]
---------------
[236]

Weiner, B., J. Amirkhan, V. S. Folkes, and J. A. Verett (1987). An attributional analysis of excuse giving: Studies of a naive theory of emotion. Journal of Personality and Social Psychology 52:316-324.

AUTHOR NOTES

Portions of this study were presented at the 1993 meeting of the Eastern Psychological Association in Washington, D.C. We want to thank Robert Hudock who provided assistance in conducting the study.

AUTHOR BIOGRAPHIES

Maurice J. Levesque is Associate Professor of Psychology at Elon University. His research concerns accuracy and bias in interpersonal perceptions and the determinants of behavior during dyadic interactions. Correspondence regarding this article may be addressed to Campus Box 2850, Elon University, Elon, NC 27244. E-mail: Levesque@elon.edu.

Charles A. Lowe is Professor of Psychology at the University of Connecticut, Storrs. His research concerns attribution theory and the factors that affect the nature of interpersonal perception. He has published articles concerning children’s understanding of basic attribution principles such as discounting.

Catherine Mendenhall is a graduate of Amherst College. She conducted data analysis and drafted parts of the manuscript under the direction of Maurice Levesque as a summer research fellow at Amherst College.

[236]
---------------
[237]

Back to Journal Cover Page