An Examination of the Relationships among Self-Perception Accuracy, Self-Awareness, Gender, and Leader Effectiveness

An Examination of the Relationships among Self-Perception Accuracy, Self-Awareness, Gender, and Leader Effectiveness

This article focuses on how membership in a self/rater agreement group (underraters, accurate raters, overraters) is related to self-ratings and others’ ratings of self-awareness and leadership effectiveness. It also examines gender differences in the likelihood of self/rater agreement and in perceived self-awareness. Finally, the article examines agree- ment group and gender differences in terms of two components of self-awareness: knowl- edge of self and willingness to improve. Contrary to common belief, our research shows that women are not more likely to underrate their own skills on measures of leadership competency, and that gender differences do exist, both in rated self-awareness and in one of its subcomponents, knowledge of self. In addition, this research found underraters were rated highest in self-awareness by direct reports and highest in terms of overall leadership effectiveness. Managers who tend to overrate themselves compared to others’ ratings were perceived as lowest of the three groups in both self-awareness and effectiveness. © 1993 by John Wiley & Sons, Inc.

INTRODUCTION

Behavioral feedback which compares self-ratings to co-worker ratings on various leadership competencies is often used to enhance manager’s awareness of their strengths and weaknesses. Such developmental as- sessment acfivities are usually based on two key assumpfions: (1) that awareness of any discrepancy between how we see ourselves and bow others see us enhances self-awareness, and (2) that enhanced self- awareness is a key to maximum performance as a leader (Tornow, 1992).

Although the empirical relationship between self/rater agreement and self-awareness is still somewhat unclear, self/rater agreenaent is often used as an operational definifion of self-awareness in research as well as in training. Self/rater agreement is used as a measure of self- awareness by both Atwater and Yammarino (1992) and Wohlers and

Human Resource Management, Summer/Fall 1993, Vol. 32, Numbers 2 & 3, Pp. 249-263 © 1993 by John Wiley & Sons, Inc. CCC 0090-4848/93/020249-15

London (1989). Atwater and Yammarino (1992) argue that the self-aware individual will have a more accurate self-assessment because “self- awareness stems from the individual’s ability to assess others’ evalua- tions of the self and to incorporate those assessments into one’s self- evaluation” (p. 143). Wohlers and London (1989) use several measures of self-awareness, including self/rater agreement, to identify managerial characteristics that managers are likely to agree on in their evaluations and to explore variables that may affect the level of agreement. These authors find that self-awareness is judged among the hardest competen- cies for others to rate, and that self-ratings of awareness are negatively related to self/rater agreement across thirty managerial characteristics.

Accurate self-reports have been shown to be related to traits associ- ated with effective leadership, including self-esteem (Farh & Dobbins, 1989), private self-consciousness (Froming and Carver, 1981; Gibbons, 1983; Nasby, 1989), intelligence, achievement status, and locus of control (Mabe & West, 1982). Other research has shown some relationships between feedback and subsequent change in self-view and/or perfor- mance (Wilson, O’Hare, & Shipper, 1990; Hazucha, Gentile, & Schnei- der, 1992; Van Velsor, Ruderman, & Young, 1991).

Research on self/rater agreement, however, has only begun to explore how congruent perceptions may be related to managerial effectiveness or performance. At least two studies (Bass & Yammarino, 1991; Atwater & Yammarino, 1992) have suggested that self/rater agreement may in- deed be related to effectiveness. In both studies, inaccurate self-raters tended to show poorer rated performance than people who rated them- selves as others rated them. In addition, the latter study found that the “overraters” not only rated themselves higher than “accurate raters” or “underraters,’ but that overraters were rated lowest by their co- workers. Of the three groups, co-worker ratings of performance were highest for people who underrate themselves. Atwater and Yammarino (1992) also found that self-awareness, as assessed by self-other agree- ment, is positively related to performance, and that the best predictors of leader behavior may differ depending on whether the leader tends to over-estimate, under-estimate, or accurately self-rate.

Research abounds suggesting there may be gender differences in the likelihood of overrating and underrating of managerial abilities. For ex- ample, research on causal attributions suggests that women may have a greater tendency to underrate their skills or performance because they tend not to take credit for success, attributing it more to external sources or to effort than to ability (Parsons, Meece, Adler, & Kaczala, 1982; Erkut, 1983; LaNoue & Curtis, 1985; Meehan & Overton, 1986). Beyer (1990, 1992) found gender differences in the accuracy of self-evaluations on masculine tasks, but not on feminine or gender-neutral tasks. On masculine gender-typed tasks, women consistently underestimated their performance and recalled more task failure than actually had oc- curred. To the extent that managerial roles and tasks are still seen as

250 / Human Resource Marmgement, Summer & Fall 1993

more masculine fhan feminine, women may be more likely fhan men to underestimate their competencies in these areas.

RESEARCH QUESTIONS

Our study had three purposes. First, we were inferesfed in learning more abouf self-rafer discrepancy, parficularly whaf accounfs for the differences between self-view and fhe views of ofhers, and how mem- bership in self/rafer agreemenf groups (i.e., underraters, accurate raters, overraters) is related to performance. In particular, we intended to repli- cate some aspecfs of fhe Afwafer and Yammarino (1992) research, using multiple managerial samples and a different multi-rater instrument.

Second, we intended to investigate the empirical relationship be- tween self/rafer agreemenf and rafed self-awareness. Alfhough Wohlers & London (1989) found fhaf self-rafings of self-awareness were nega- fively relafed fo self/rafer discrepancy (i.e., fhose who were rafed higher on self-awareness had less self-rafer discrepancy), fheir sfudy did nof disfinguish underrafers from overrafers. Our sfudy design allowed for fhis comparison by examining bofh self-rafings and others’ ratings of self-awareness for fhe fhree agreemenf groups.

Third, we infended fo learn more abouf gender differences in rafed self-awareness and in self/rafer agreemenf on performance in other im- portant managerial domains.

SAMPLES

Our data were obtained from fhree samples of managers who com- plefed Benchmarks (Cenfer for Creafive Leadership, 1990), a mulfi-rafer assessmenf insfrumenf confaining fwenfy-two scales. Sixteen of the twenty-two scales were used in this study. The normative data and reliability coefficients for these sixteen scales are listed in Table I.

The first sample was a group of 648 managers randomly selected from the Benchmarks database. The second sample comprised 168 upper- level managers from a Fortune 100 organization, while the third group was a sample of 79 hospital administrators who completed Benchmarks in 1990.

MEASURES

Categorization of Agreement Groups

In order to distinguish underraters from overraters in our analysis, managers in each of these samples were coded into one of three groups

Van Velsor, Taylor, and Leslie: Self-Perception and Leadership / 251

Table L Sixteen Benchmarks Scales: Means, Standard Deviations, Test/Retesl Reliabilities, Alpha coefficients.

1. Resourcefulness 2. Doing whatever it takes 3. Being a quick study 4. Decisiveness 5. Leading subordinates 6. Setting a developmental

climate 7. Confronting problem

subordinates 8. Work team orientation 9. Hiring talented staff

10. Building and mending relationships

11. Compassion and sensi- tivity

12. Straightforwardness and composure

13. Balance between person- al life and work

14. Self-awareness 15. Putting people at ease 16. Acting with flexibility

Mean

3.59 3.72 3.79 3.33 3.54

3.67

3.25 3.46 3.66

3.60

3.49

4.10

3.59 3.47 3.77 3.62

Standard Deviation

.55

.55

.68

.92

.57

.59

.73

.73

.66

.62

.65

.67

.81

.69

.95

.65

Test-Retest Reliability

.88

.91

.71

.86

.95

.90

.83

.78

.82

.88

.87

.91

.84

.80

.82

.86

Alpha Coefficient

.93

.92

.92

.75

.92

.85

.93

.81

.97

.93

.84

.88

.88

.82

.85

.85

Note. Excerpted from Benchmarks trainer’s manual (Lombardo & McCauIey, 1992).

(underraters, accurate raters, and overraters). Group membership was based on the average discrepancy between self-ratings and mean subor- dinate ratings across fifteen of the sixteen Benchmarks scales. One scale, Self-Awareness, was left out of the creation of this measure in order to later assess rated self-awareness among the three agreement groups.

An individual was coded as an underrater if the average self-rating across the fifteen scales was more than one-half standard deviation low- er than the averaged direct report ratings across the fifteen scales in the Benchmarks database sample. An individual was coded as an overrater if the average self-rating across the fifteen scales was more than one-half standard deviation higher than the averaged direct reports’ rating from the Benchmarks database sample across the fifteen scales. Finally, man- agers were placed in the accurate rating group if their self-ratings fell within one-half standard deviation plus or minus the overall averaged scores given by direct reports’ ratings on these scales.

Ratings from direct reports were used in this research for two rea- sons. In one of our samples, we had only direct report data and saw value in using measures that would facilitate comparability of results

252 / Human Resource Management, Summer & Fall 1993

across samples. Although some may argue that boss or superior ratings are the most important measures of performance, a growing body of research indicates that follower perceptions of leader behavior comprise the most meaningful data (Atwater & Yammarino, 1992; Bass, 1990; Ashford 1989). In this research we also wanted to avoid use of the boss ratings because with Benchmarks, as with many other feedback instru- ments of tbis kind, superiors are aware that tbeir responses are not con- fidential (i.e., they are broken out from other ratings in the feedback), and we suspect these ratings may be inflated as a result.

Self-ratings were compared to mean direct reports’ ratings instead of to each individual subordinate rating because we believed, as did Wohlers and London (1989), that “when combined, ratings tend to match the target managers’ views of selves more than target managers’ views match any one coworker’s ratings” (p. 251). In addition, several studies have indicated that the average of ratings is more reliable than a single rafing (French & Bell, 1978; Mount 1984; Latham & Wexley, 1981; Miner 1968). So, self/rater agreement, using this method, may actually be a conservative estimate of “real” agreement between target managers and individual raters or groups of raters.

Effectiveness Measures

In the hospital administrators sample, we were able to obtain four criterion measures of managerial effectiveness and subordinate satisfac- tion with the leader—all collected at the same time tbat Benchmarks was administered. The effectiveness and satisfaction items were: How likely is this person to derail (be plateaued, demoted, or fired) in the next five years? What is this person’s level of performance in his/her present job? How satisfied are you with this person as a leader? For purposes of this analysis, these four items were converted to a single performance scale (alpha = .92). For our other two samples, we did not have the benefit of an independent criterion measure of performance.

RESULTS

Distribution of Managers across Agreement Groups

Although research would suggest that women may be more likely to underrate themselves on competencies related to effective leadership, our results do not support this hypothesis. As can be seen in Table II, the distribution of men and women across the three agreement groups is not significantly different (x^ = 1.08, p = .58). While much research suffers from an underrepresentation of women in the sample, this cannot be

Van Velsor, Taylor, and Leslie: Self-Perception and Leadership / 253

Table II. Frequency Distribution (N, %) of Women and Men across Agreement Groups” (Sample 1)

Women

Men

Underrater Group

124 27.49

46 27.06

170 27.38

Agreement Group

176 39.02

60 35.29

236 38.00

Overrater Group

151 33.48

64 37.65

215 34.62

451 72.62

170 27.38

621 100.00

X2 = 1.079; ;- = .583. “Cell sample sizes and standard deviations available upon request from the authors.

the explanation here, since almost three-fourths of our sample were women.

What Accounts for Self/Rater Discrepancy?

Until Atwater and Yammarino (1992) published their research, the meaning of the discrepancy between self- and other-ratings was unclear. The appearance of elevation in the self-ratings of overraters, for exam- ple, could be a function of lowered ratings by others, rather than actu- ally reflecting elevated self-ratings. In other words, overraters could conceivably have identical self-ratings to underraters, but receive lower ratings from others.

Comparing the overall (average across Benchmarks scales) self-view and the overall average rater view for the three agreement groups (over- raters, accurate raters, underraters), we found, as did Atwater and Yam- marino (1992), that discrepancy results from differences in both self- ratings and other-ratings. This finding was consistent across all three samples, and no gender differences were found. As can be seen in Table III, underraters consistently rate themselves lower than do accurate rat- ers or overraters, AND underraters are consistently rated highest by their direct reports. Overraters, on the other hand, are consistently rated lowest by their subordinates, while consistently rating themselves high- er than the other two groups. In fact, in our large database sample, the direct reports’ rating for overraters (3.61) is almost one and one-half standard deviations lower then the self-ratings of overraters (4.05).

These data suggest that underraters are perceived as the highest per- forming or most effective managers, even more so than managers who tend to see themselves accurately and much more so than managers who overrate their skills. To test this result further, we used the hospital

254 / Human Resource Management, Summer & Fall 1993

c

01

6

-ac

Se ll

M ea

n

_; ^̂ 01

3 f2

01 UI

3

ns01

ve ne

ss

•.c

u 13 C ro

X

>

“^ aj 00

tn

2 o

c o

so

c ^ ‘x: rtj L- ro

0) ^

en 60

— c c “3 ‘Ĵ n! t- ra O O i

‘ ^ f^ ro

•>* •^ ro

O; 00 t^ m fO r*̂

t^ ô in IT) 00 ^

00

ino

Ci. 3

;.. tc 0; ro VH

01 T3 C D

CL.

3

, . C 01

re e

<

ro up

L a

at e;

er r

>

O

Van Velsor, Taylor, and Leslie: Self-Perception and Leadership / 255

administrators sample and compared the scores on the four-item effec- tiveness scale for the three agreement groups, using analysis of vari- ance. As can be seen in Table III, the differences between groups on the independent effectiveness measure are significant for two of the three pairwise comparisons. Underraters again receive the highest effective- ness rating from direct reports and overraters receive the lowest. The only pairwise comparison that is not statistically significant is the differ- ence between underraters and accurate raters. This finding lends addi- tional support to the idea that managers who underrate themselves are perceived by direct reports as being more effective than those whose self- and others-ratings are accurate.

Self/Rater Discrepancy and Self-Awareness

By definition, both overraters and underraters lack self-awareness, since neither group’s members rate themselves “accurately” (i.e., as others rate them). So we might expect that both underraters and overra- ters would be rated lower on self-awareness than would accurate raters. An analysis of variance comparing the three groups in terms of the direct ratings of self-awareness by self and by subordinates did show differences between the groups, but these differences were not entirely as expected, especially in terms of rater perceptions of managers’ self- awareness.

While we might have expected the accurate raters to have the highest rated self-awareness, it was the underrater group who was rated highest by direct reports in this area. True to their tendency to underrate them- selves, this group’s self-ratings of self-awareness were lowest of the three groups. So, although the self/rater discrepancy of underraters on this self-awareness scale was high, their rated self-awareness by direct reports was also high. It may be that for this group, self/rater discrepan- cy and self-awareness are not the same phenomenon.

As shown in Table IV, the ratings of accurate raters tended to fall between underraters and overraters, both in terms of their self-ratings and in direct reports’ ratings of self-awareness. Overraters had the high- est self-ratings of self-awareness of any of the three groups. Although this group’s members saw themselves as most self-aware, their raters gave them the lowest rating of any group on self-awareness, indicating that self/rater discrepancy may be an adequate indicator of self- awareness for managers who tend to overrate themselves compared to others.

We found some gender differences in rated self-awareness as well. Although no gender differences showed up in self-ratings of self- awareness. Table V shows direct reports rated women significantly high- er on self-awareness than they rated men. So, while women managers

256 / Human Resource Management, Summer & Fall 1993

Table IV. ANOVA of Self- and Direct Reports Ratings on Self-Awareness by Level of Agreement

Underrater group

Agreement group

Overrater group

f

Sample 1 (Database) JV = 633

Self- Rating

3.50

3.78

4.00

.15 56.23

p< .01

Direct Report Rating

3.83

3.66

3.40

22 86.93

;’ < .01

Sample 2 (Senior Execs.)

N = 168

Self- Rating

3.44

3.79

4.07

.11 11.08

p< .01

Direct Report Rating

3.72

3.58

3.50

.24 28.06

p< .01

Sample 3 (Hosp. Admin)

N = 74

Self- Rating

3.79

3.88

4.27

.17 7.22

p< .01

Direct Report Rating

4.11

3.70

3.46

.16 6.54

p< .01

do not perceive themselves as more self-aware than men do, they are perceived as such by the people who report to them.

Components of Self-Awareness

In thinking over how to interpret these findings, we were drawn back to the similarity between the Benchmarks self-awareness scale and At-

Table V. Gender Differences in Others’ Ratings of Self-Awareness (Sample 1).

Women

Men

Underrater Group

3.87

3.72

3.83 (n = 170)

Agreement Group

3.66

3.65

3.66 (n = 236) p< .01

Overrater Group

3.41

3.36

3.40 (n = 215)

3.63 (n = 451)

p< .02 3.56

(n = 170)

Notes: N = 648, i = 3.61. No significant interaction effects.

Van Velsor, Taylor, and Leslie: Self Perception and Leadership / 257

Table VI. Comparison of Three Agreement Groups on Two Components of Benchmarks Self-Awareness Scale (Sample 1)

Knows Self Willing to Improve

Underrater group 3.87 3.79 Agreement group 3.70 3.61 Overrater group 3.42 3.36

Noh\ Mests for all pairwise comparisons significant (p < .01).

water & Yammarino’s (1992) definition of self-awareness (“the individu- al’s ability to assess others’ evaluations of the self and to incorporate those assessments into one’s self-evaluation,” p. 143). The Benchmarks scale is made up of four items: two of which refer to knowledge of self, and two of which refer to willingness to change. Because these two components are conceptually distinct, we pursued an analysis of the subcomponents of this self-awareness scale.

In preparation for this analysis, we assessed the reliability of these two-item scales, using our large random sample. The “knows self” scale had an alpha coefficient of .84 and an item correlation of .73. The “willing to improve” scale had an alpha of .73 and an item correlation of .57.

We hypothesized that underraters may be rated highest by subordi- nates on self-awareness because of a perceived greater willingness to improve (possibly in keeping with an orientation toward humility over- all), but that their subordinate ratings on knows self may be lower than those managers in the accurate raters group. In terms of gender differ- ences, we saw no reason to believe that the self-awareness advantage that women appear to have would be more likely to result from either of these two components.

Contrary to both our expectations, we found that underraters score highest on both components of self-awareness. Furthermore, the gender difference in self-awareness appears to be in the knows self component of self-awareness, rather than in the willingness to improve aspect. As can be seen in Table VI, subordinate ratings for knows self and for willing to improve were highest for underraters and lowest for overra- ters. All two-way comparisons were significantly different. Underraters were rated more highly than were accurate raters on both components, and accurate raters were rated more highly on both components than were overraters. Table VII shows the significant difference (p < .02) between direct reports’ ratings of men (3.60) and women (3.70) on the knows self subscale. No significant gender differences were found on the subscale, willingness to improve.

258 / Human Resource Management, Summer & Fall 1993

Table VIL Gender Differences in Knowledge of Self Component of Self-Awareness (Sample 1). ‘

Women

Men

Underrater Group

3.91

3.74

3.87 {n = 170)

Agreement Group

3.70

3.70

3.70 {n = 236) p< .01

Overrater Group

3.43

3.37

3.42 (n = 215)

3.70 (n = 451)

p< .02

3.60 (n = 170)

Notes: N = 648, :* = 3.65.

DISCUSSION

Gender Differences

Confrary fo popular belief and some research, we have found fhaf women are nof more likely than men to underrate themselves on com- petencies related to effective leadership. This resulf may reflecf a grow- ing comfort among women with the sex role appropriateness of man- agerial roles, in fhaf women do nof appear fo be parficularly likely fo undervalue fheir confribufions or skills in fhis once fradifionally masculine area. Alfernatively, it may reflecf fhe greafer self-confidence of women who fend fo enfer managerial jobs. These women may be less likely than women in general to attribute success or compe- tence to sources outside themselves. In any case, this research indicates we should no longer assume that women’s inclination is to under- rate.

We did find gender differences on fhe self-awareness scale and on one of ifs subcomponenfs. Alfhough women do not rate themselves differ- enfly fhan men on self-awareness, women are rafed higher by ofhers. Yef unlike fhe resulfs for underrafers, women’s greafer self-awareness seems fo be relafed only fo fhe knowledge of self componenf, on which fhey were perceived significanfly more posifively by subordinafes fhan were men. There were no gender differences on fhe willingness to im- prove component of self-awareness.

Van Velsor, Taylor, and Leslie: Self-Perception and Leadership / 259

Working with Underraters

Our research has supported earlier research showing that self/rater discrepancies are due to differences in both self-ratings and ratings by others. Underraters both rate themselves lowest of any group and are rated by others as highest. The idea that managers who underrate them- selves may be perceived as more effective leaders was confirmed by our comparison of independent effectiveness ratings across the three agree- ment groups.

In addition to being perceived as most effective, underraters are also perceived by direct reports as most self-aware, even though self/rater discrepancy on the self-awareness scale is high for underraters. In terms of the components of self-awareness, knowledge of self and willingness to improve, our results were contrary to our expectations in that under- raters were rated higher by direct reports in both areas. In other words, underraters’ high ratings on self-awareness were nof due only to a great- er willingness to improve, but to a greater perceived knowledge of self as well.

These findings suggest that self/rater discrepancy may not be measur- ing self-awareness, as we usually conceive of it, among the underrater group. It may also be the case that leniency bias on the part of raters may be operating here, rather than true high performance by underraters. Self-awareness has proven to be a difficult competency to rate (Wohlers & London, 1989), and it may be that a demonstrated or outward lack of self-confidence on the part of underrating managers has an impact on the willingness or ability of subordinates to objectively rate self- awareness. Raters may be more reluctant to give low self-awareness ratings to people they perceive as lacking self-confidence or as having a great deal of humility. In fact, if raters are generally reluctant to further deflate (underrater) managers’ self-perceptions by giving them low rat- ings on other skills or behaviors, we might question the validity of subordinate ratings across competencies for underrating managers.

When giving managers feedback, our research suggests that human resource professionals should pay attention to both the magnitude and the direction of the discrepancy between self and other ratings. Until we know more about how to interpret the findings on underraters, one should check for or ask managers to check for other sources that validate high performance ratings.

Working with Overraters

Our research also indicates that, in working with managers who tend to overrate themselves, self/rater discrepancy may be a good indicator of self-awareness. People whose self-views are significantly inflated as

260 / Human Resource Management, Summer & Fall 1993

compared to others’ views are seen by others as lowest in self- awareness. In addition overraters rate themselves highest of any agree- ment group in terms of performance, but are rated by others as the poorest performers. While effecfive human resource and training pro- fessionals have always paid parficular attention to individuals wbo rate themselves higher than others rate them, our research underscores the importance of this care. Both self-perception accuracy and performance tend to be problem areas for the managers in the overrater group.

Stafistical, personality, and performance-related factors probably all play a part in the results reported here. It will be important that future research attempt to identify and separate these effects so that we can confinue to enhance the ways we work with managers to belp tbem learn from feedback.

Ellen Van Velsor is Director of Product Dei>elopment Research at the Genter for Greative Leadership. In this capacity, she is responsible for promoting the use of research in the development of programs and feedback technologies and managing database research projects. Ellen is co-author of Breaking the Glass Ceiling: Can Women Reach the Top of America’s Largest Corporafions? (1987, Addison-Wesley). She has authored numerous other articles and reports, includ- ing Gender Differences in the Development of Managers (1990, Genter for Greative Leadership) and Feedback to Managers, Volumes I & II (1992, Genter for Greative Leadership). Before joining the staff at the Genter, Ellen was a Postdoctoral Fellow in adult development at Duke University. She holds a Ph.D. in Sociology from the University of Florida.

Sylvester Taylor is a Research & Program Associate at the Genter for Greative Leadership. Since joining the Genter in 1989, Sylvester has been involved in researching and teaching in the area of organizational creativity, innovation, and learning. His current research interests include leadership effectiveness and man- agerial learning and development. Prior to coming to the Genter, Sylvester was at the University of North Garolina in Ghapel Hill where he worked as an instructor in the Department of Psychology and as an administrator in the Division of Student Affairs. He has published and edited articles in scholarly and professional journals on human cognition, human behavior and organizational effectiveness. Sylvester is currently completing a Ph.D. degree in Organizational Development at the University of North Garolina in Ghapel Hill.

Jean B. Leslie is a Research Analyst at the Genter for Greative Leadership. Her current research focuses on psychometric analyses of Genter instruments and cross-cultural applicability of Leadership models. Leslie is co-author o/Feedback to Managers, Volumes I and II (1992, Genter for Greative Leadership), and several articles published in sociological journals. Before coming to the Genter, she taught sociology courses at Eton Gollege, Guilj^ord Technical Gommunity Gollege, and Greensboro Gollege, all in North Garolina. She has an M.A. in Sociology from the University of North Garolina at Greensboro.

Van Velsor, Taylor, and Leslie: Self-Perception and Leadership / 261

REFERENCES

Ashford, S. (1989). Self-assessments in organizations: A literature review and integrative model. Research in Organizational Behavior, 11, 33-174.

Atwater, L. E., & Yammarino, F. J. (1992). Does self-other agreement on leader- ship perceptions moderate the validity of leadership and performance predic- tions? Personnel Psychology, 45, 141-164.

Bass, B. M. (1990). Bass and Stogdill’s handbook of leadership. New York: Free Press. Bass, B. M., & Yammarino, F. J. (1991). Congruence of self and others’ leadership

ratings of naval officers for understanding successful performance. Applied Psychology: An International Rcvieiv, 40 (4), 437-454.

Benchmarks (1990). Assessment instrument. Center for Creative Leadership. Greensboro, NC

Beyer, S. (1990). Gender differences in the accuracy of self-evaluations of perfor- mance. Journal of Personality and Social Psychology, 54, 5-12.

Beyer, S. (1992). Self-consistency and gender differences in the accuracy of self- evaluations. Paper presented at the American Psychological Association, Washington, D.C

Erkut, S. (1983). Exploring sex differences in expectancy, attribution, and aca- demic achievement. Sex Roles, 9, 217-231.

Farh, J. L., & Dobbins, G. H. (1989). Effects of self-esteem on leniency bias in self-reports of performance: A structural equation model analysis. Personnel Psychology, 42, 835-850.

French, W. L.., & Bell, C H., Jr. (1978). Organization development: Behavioral science interventions for organization improvement. Englewood Cliffs, NJ: Prentice-Hall.

Froming, W., & Carver, C (1981). Divergent influences of private and public self- consciousness in a compliance paradigm. Journal of Research in Personality, 15, 159-171.

Gibbons, F. (1983). Self-attention and self-report: The “veridicality” hypothesis. fournal of Personality, 51, 517-542.

Hazucha, J. F., Gentile, S.A., & Schneider, R. J. (1992). The impact of self- managed development, boss support and organizational support on skill development. Paper presented at seventh annual conference of the Society for Industrial and Organizational Psychology, Montreal, Canada.

Latham, G. P., & Wexley, K. N. (1981). Increasing productivity through performance appraisal, Reading, MA: Addison-Wesley.

LaNoue, J. B., & Curtis, R. C (1985). Improving women’s performance in mixed- sex situations by effort attributions. Psychology of Women Quarterly, 9, 337-356.

Lombardo, M., & McCauley, C (1992). Benchmarks: Developmental reference points for managers and executives. Greensboro, NC: Center for Creative Leadership.

Lombardo, M., & McCauley, C (1992). A manual and trainers guide, Greensboro, NC: Center for Creative Leadership.

Mabe, P. A., & West, S. G. (1982). Validity of self-evaluation of ability: A review and meta-analysis. fournal of Applied Psychology, 67(3), 280-296.

Meehan, A. M., & Overton, W. F. (1986). Gender differences in expectancies for success and performance on Piagetian spatial tasks. Merrill-Palmer Quarterly, 32, 427-441.

Miner, J. B. (1968). Management appraisal: A capsule review and current refer- ences. Business Horizons, 11, 83-96.

Mount, M. K. (1984). Psychometric Properties of Subordinate ratings of manage- rial performance. Personnel Psychology, 37, 687-702.

Nasby, W. (1989). Private self-consciousness, self-awareness, and the reliability of self-reports, fournal of Personality and Social Psychology, 56, 950-957.

262 / Human Resource Management, Summer & Fall 1993

Parsons, J. E., Meece, J. L., Adler, T. F., & Kaczala, C. M. (1982). Sex differences in attributions and learned helplessness. Sex Roles, 8, 421-432.

Tornow, W. W. (1992). Perception or reality: Is measurement a means or an end? Paper presented at the Society for Industrial and Organizational Psychology, Montreal, Quebec.

Van Velsor, E., Ruderman, M., & Young, D. (1992). Enhancing self-objectivity and performance on the job: The role of upward feedback. Paper presented at the Society for Industrial & Organizational Psychology, St. Louis, Missouri.

Wilson, C. L., O’Hare, D., & Shipper, F. (1990). Task cycle theory: The processes of influence. In Clark, K. E., & Clark, M. B. (Eds.), Measures of leadership. West Orange, NJ: Leadership Library of America.

Wohlers, A. J., & London, M. (1989). Ratings of managerial characteristics: Eval- uation difficulty, co-worker agreement and self-awareness. Personnel Psycholo- gy, 42, 235-261.

Van Velsor, Taylor, and Leslie: Self-Perception and Leadership / 263


Comments are closed.