Journal of Gerontology: PSYCHOLOGICAL SCIENCES 2006, Vol. 61B, No. 6, P333–P339

Copyright 2006 by The Gerontological Society of America

Perceptions of Forgetful and Slow Employees: Does Age Matter? Joan T. Erber and Beverly A. Long Department of Psychology, Florida International University, Miami.

Participants (perceivers) read a vignette describing a young or older employee (target) in a young-relevant or old-relevant work context who is either forgetful or slow. Regardless of work context, perceivers attributed older targets’ forgetful and slow behavior more to internal stable causes but the identical behavior of younger targets more to internal unstable causes. Perceivers also felt less anger and greater sympathy and were more likely to recommend a promotion and raise for older than for young targets. Perceivers’ anger for young and older targets was mediated by their internal unstable attributions, but their sympathy was not mediated by their internal stable attributions. Perceivers’ promotion and raise recommendations were associated with the degree of sympathy they felt more for young than for older targets.

I

N A perfect world, employees would remember to do all their job-related tasks and would accomplish each and every one with swiftness and efficiency. In the real world of work, however, employees do not always behave in ideal ways. Sometimes they forget to do things and oftentimes they do not perform their jobs quickly enough. How are such employees perceived by managers, who could affect their career paths and level of compensation? Do young and older employees who forget to do tasks or accomplish them too slowly receive a similar level of consideration, or is one age group given more leeway than the other? Although it is illegal to discriminate against older workers, numerous plaintiffs have won age-discrimination lawsuits against employers (McCann & Giles, 2002). However, research findings on age-based job discrimination are complex (Finkelstein, Burke, & Raju, 1995; Kite & Wagner, 2002; Pasupathi & Lockenhoff, 2002), suggesting the need for further investigation. If discriminatory practices toward older workers exist, these may stem from stereotypical expectations (Finkelstein et al., 1995; Finkelstein, Higgins, & Clancy, 2000; Heckhausen, Dixon, & Baltes, 1989; McCann & Giles), and an employee’s age may be particularly salient when his or her performance falls below certain standards. Perceivers tend to attribute everyday forgetting to internal stable causes when a target person is older but to internal unstable causes when a target is young (Erber, Szuchman, & Rothberg, 1990). We hypothesized that perceivers would apply this age-based double standard when making attributions not only for employees’ forgetting but also for their slowness. Speed of response has been widely studied with regard to age (Baltes & Lindenberger, 1997; Cerella, 1991, 1994; Fisk & Rogers, 1991; Salthouse, 1996), but there is little empirical data on attributions made when young versus older employees are slow to accomplish work-related tasks. In addition to attributions, how do managers react to employees who forget to do tasks or do not do their jobs quickly enough, and how do they evaluate such employees? We hypothesized that perceivers would use an agebased double standard when making these judgments as well. According to a model proposed by Weiner (1993), attributing another person’s behavior to internal stable causes

engenders sympathy and ultimately more lenient treatment, whereas attributing behavior to internal unstable causes engenders anger followed by harsher treatment. However, older adults who show signs of incompetence are often perceived as being warmer than young adults who behave the same way (see Cuddy & Fiske, 2002; Cuddy, Norton, & Fiske, 2005). Thus, we hypothesized that older employees would engender sympathy regardless of the attributions made for their forgetful or slow behavior, and also that they would be evaluated more leniently. However, there was little reason to expect that older adults would be exempt from engendering anger if perceivers thought they were just not paying attention or trying hard enough. Accordingly, we anticipated there would be a link (as predicted by Weiner’s model) between unstable attributions and anger regardless of employees’ age. One further issue that we address in this study is whether age discrimination in the workplace is more likely to occur in some job contexts than in others. In prior research, congruence between age and job type influenced judgments about employees (Cleveland & Landy, 1987). For young-type jobs (intermediate programmer, CD salesperson), older applicants were evaluated less favorably than young applicants were. However, older adults working in old-type jobs (appraising and selling stamps and coins) were viewed positively (Cleveland, Festa, & Montgomery, 1988; Perry, Kulik, & Bourhis, 1996). Therefore, we hypothesized that there would be a double standard wherein perceivers would react to older employees more positively and evaluate them more favorably when a job was age typed as old rather than young.

METHODS

Participants Participants (i.e., perceivers) were 167 adults (61 men and 106 women), ages 19 to 43 years (M ¼ 24.77, SD ¼ 4.99), most of whom attended evening classes at Florida International University. Perceivers had 12 to 16 years of education (M ¼ 14.51; SD ¼ .59) and scores from 7 to 24 (M ¼ 13.80; SD ¼ 3.19) on the last 25 items of the Shipley–Hartford Vocabulary

P333

ERBER AND LONG

P334

Table 1. Means and Standard Deviations for Attribution, Reaction, and Evaluation Ratings for Young and Older Targets

Rating Scale

Young Targets

Older Targets

M

SD

M

SD

3.39 2.81 4.80 4.90

1.55 1.53 1.59 1.41

4.34 3.90 3.23 3.65

1.46 1.69 1.83 1.88

3.99 3.52

1.59 1.42

3.16 4.67

1.51 1.49

2.92 3.18 3.69

1.29 1.46 .98

3.30 3.64 3.96

1.33 1.45 .93

Attributions Poor Memory Ability*** Mental Difficulty*** Lack of Effort*** Lack of Attention*** Reactions Anger** Sympathy*** Evaluations Promotion* Raise* Work rating

Note: SD ¼ standard deviation. *p , .05; **p , .001; ***p , .0001.

Test (Shipley, 1940), a measure of verbal ability. Of the 167 perceivers, 136 (81.4%) held jobs (71 had full-time jobs and 65 had part-time jobs), and they supervised between 0 and 150 employees (M ¼ 3.04, SD ¼ 13.46). Correlations between perceivers’ age in years and their ratings on the subsequently described scales were not significant.

Materials Vignettes in the main study specified the target’s chronological age but also included physical cues regarding the target’s age-group membership and a description of a youngtype or old-type work context. Physical cues. —To verify the age appropriateness of a target’s physical cues, we had 20 pilot perceivers (7 women and 13 men, 18–50 years of age, M ¼ 23.8) read descriptions and assign a chronological age to each. Average ages assigned to two young descriptions were 27.65 (SD ¼ 7.04) and 27.50 (SD ¼ 5.44), not significantly different in a paired-sample t test, t(19) ¼ 0.09, p ¼ .927. Average ages assigned to two older descriptions were 55.10 (SD ¼ 10.94) and 53.10 (SD ¼ 13.32), not significantly different in a paired-sample t test, t(19) ¼ 0.658, p ¼ .519. Work context. —To verify the age relevance of the work descriptions, we had a new sample of 20 pilot perceivers (9 men and 11 women, 17–28 years of age, M ¼ 21.85) estimate the proportion of employees at an e-commerce Internet firm and public library who would be 32 years of age or younger and also 55 years of age or older. In paired-sample t tests, perceivers estimated that a significantly greater proportion of e-commerce employees would be young rather than older, t(19) ¼ 14.127, p , .0001 (Ms ¼ 73.5% vs 12.25% for young and older employees, respectively), and a significantly greater proportion of library employees would be older rather than young, t(19) ¼ 4.033, p , .001 (Ms ¼ 57.5% vs 29.5% for older and young employees, respectively). A paired-sample t test also verified that pilot perceivers would be more surprised (1 ¼ not at all surprised; 7 ¼ very surprised) if told 75% of e-commerce

employees were older adults than if told 75% of e-commerce employees were young adults, t(19) ¼ 13.23, p , .0001 (Ms ¼ 5.95 vs 1.40, respectively). They would also be more surprised if told 75% of library employees were young than if told 75% of library employees were older, t(19) ¼ 4.76, p , .0001 (Ms ¼ 4.95 vs 2.25 for the young vs older employees, respectively). Accordingly, we considered the e-commerce Internet firm and the public library to be young-relevant and old-relevant work contexts, respectively.

Procedure In the main study, experimenters instructed half the perceivers to imagine themselves as managers of an e-commerce Internet firm and the other half as managers of a large county library system. Experimenters told perceivers to rely on information provided by each employee’s immediate supervisor in order to make job-related evaluations of employees. The perceivers then read a vignette in which an immediate supervisor describes a young (or older) employee at an e-commerce Internet firm (or public library). Within each work context, the young or older employee was described as either forgetful or slow (see the Appendix for a description of the vignettes). Perceivers rated four attributions for why they thought the young (older) employee might be forgetful or slow (1 ¼ not at all a cause; 7 ¼ very much a cause). There were two internal stable attributions (poor memory ability, mental difficulty) and two internal unstable attributions (lack of effort, lack of attention). Perceivers then rated how much anger they, as managers, would feel toward this young (older) employee who is forgetful or slow (1 ¼ none at all; 7 ¼ very much) and also how much sympathy they would feel (1 ¼ none at all; 7 ¼ very much). Finally, perceivers evaluated the employee by using three scales: whether they would recommend this employee for a promotion (1 ¼ not at all likely; 7 ¼ very likely); whether they would recommend an additional raise over and above a cost-ofliving increase (1 ¼ not at all likely; 7 ¼ very likely); and how they would rate this employee overall (1 ¼ unsatisfactory; 7 ¼ outstanding). After perceivers completed their ratings, a manipulation check verified that they were aware of the target’s age group and gender.

RESULTS We performed analyses of variance (ANOVAs) to test whether perceivers used an age-based double standard in their (a) attributions, (b) reactions, and (c) evaluations of the young and older employees. For each set of analyses, a modified stepdown Bonferroni procedure (Holm, 1979; Jaccard & GuilamoRamos, 2002) ensured that p levels of the individual dependent measures could be deemed significant. In the following paragraphs, we report significant main and interaction effects. Table 1 shows the means and standard deviations for young and older targets on each rating scale.

Attributions In 2 (target age: young vs older ) 3 2 (target problem: forget vs slow) 3 2 (workplace: e-commerce Internet vs library) between-group ANOVAs, the target age main effect was significant for all four attributions: poor memory ability, F(1, 159) ¼ 17.36, p , .0001, g2 ¼ .098; mental difficulty,

AGE AND PERCEPTIONS OF FORGETFUL AND SLOW EMPLOYEES

P335

F(1, 159) ¼ 18.66, p , .0001, g2 ¼ .105; lack of effort, F(1, 159) ¼ 35.20, p , .0001, g2 ¼ .181; and lack of attention, F(1, 159) ¼ 25.77, p , .0001, g2 ¼ .139. Perceivers gave higher ratings to older targets on poor memory ability and mental difficulty, but they gave higher ratings to young targets on lack of effort and lack of attention. The target problem main effect was significant for two attributions: poor memory ability, F(1, 159) ¼ 7.99, p , .005, g2 ¼ .048, and lack of attention, F(1, 159) ¼ 10.61, p , .001, g2 ¼ .063. Perceivers gave higher ratings to poor memory ability when the target was forgetful (M ¼ 4.20) rather than when the target was slow (M ¼ 3.55), and they gave higher ratings to lack of attention when the target was forgetful (M ¼ 4.69) rather than when the target was slow (M ¼ 3.88).

Reactions In 2 (target age) 3 2 (target problem) 3 2 (workplace) between-group ANOVAs on scales concerning perceivers’ reactions to the forgetful or slow employee, the target age main effect was significant for both anger, F(1, 159) ¼ 11.42, p , .001, g2 ¼ .067, and sympathy, F(1, 159) ¼ 26.87, p , .0001, g2 ¼ .145. Perceivers would be angrier with young targets than with older targets and would have less sympathy for young targets than they would for older targets. The Target Age 3 Workplace interaction was not significant for anger ( p , .94), but it was significant for sympathy, F(1, 159) ¼ 5.74, p , .018, g2 ¼ .035. To gain further insight into this interaction, we conducted simple main effects tests within each level of target age to determine whether degree of sympathy differed as a function of the two work settings. Perceivers had greater sympathy for young employees in a library (M ¼ 3.88) than in an e-commerce firm (M ¼ 3.16), F(1, 159) ¼ 5.29 p , .023. For older employees, level of sympathy did not differ significantly across work settings, F(1, 159) ¼ 1.19, p , .276 (Ms ¼ 4.84 and 4.50 for an e-commerce firm and library, respectively).

Evaluations In 2 (target age) 3 2 (target problem) 3 2 (workplace) ANOVAs on the evaluation scales (one perceiver did not complete the raise scale), the target age main effect was significant for promotion, F(1, 159) ¼ 3.91, p , .05, g2 ¼ .024, and raise, F(1, 158) ¼ 4.41, p , .037, g2 ¼ .027. It only approached significance for work rating, F(1, 159) ¼ 3.47, p , .064, g2 ¼ .021. Perceivers were more likely to recommend a promotion and raise for older than for young employees. There was a trend for perceivers to give higher work ratings to the older employees.

Weiner’s 1993 Model To ascertain whether Weiner’s 1993 model applies to young and older employees, we conducted analyses to test (a) whether internal stable and internal unstable attributions mediate the association between target age and perceivers’ reactions and (b) whether target age moderates the relationship between perceivers’ reactions and their employee evaluations. To test for mediation, we employed a structural equation modeling approach (see Figure 1). Using this method, we assume mediation to exist if (a) the path from the distal variable to the mediator is statistically significant and (b) the path from

Figure 1. Structural equation model illustrating whether the internal stable and unstable attributions mediate the association between target age and perceivers’ affective reactions (sympathy and anger).

the mediator to the outcome variable is statistically significant (MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002). We tested the model by using a maximum likelihood fit function. For continuous variables in the model there was no notable nonnormality, with univariate skewness ranging from
0.36 to 0.18 and univariate kurtosis ranging from
1.11 to
0.78. The model is just-identified so the number of knowns and unknowns are equal and degrees of freedom are zero. We obtained parameter estimates by using the AMOS computer program. Paths that yielded statistically nonsignificant coefficients are indicated with dashed lines, whereas statistically significant path coefficients ( p , .05) are indicated by solid lines. For statistically significant paths, the unstandardized coefficients are shown in parentheses. All residuals are in standardized form. (Although we do not show correlated errors in Figure 1 so as to reduce clutter, we estimated them between the four mediators, because it is unreasonable to assume that target age is the sole source of correlations between them, as well as between the two outcome variables, because it also is unreasonable to assume that the four mediators are the sole source of the correlation between them. Correlated errors ranged from absolute values of 0.01 to 0.54, with a median of 0.14.) As seen in Figure 1, target age was a significant predictor of all four attributions. In addition, all four mediators were significantly correlated with sympathy. However, when we computed path coefficients reflecting unique explained variance, only the lack-of-effort coefficient achieved statistical significance. Both lack of effort and lack of attention were significantly correlated with anger. However, when we computed path coefficients reflecting unique explained variance, only lack of effort yielded a statistically significant coefficient. Also as seen in Figure 1, the mental difficulty and poor memory attributions did not mediate the effect of target age on sympathy. Indeed, there was a statistically significant path between target age and sympathy holding constant perceivers’ judgments about mental difficulty and poor memory ability.

ERBER AND LONG

P336

Table 2. Pearson r Correlations Between the Attribution, Reaction, and Evaluation Ratings for the Young and Older Targets Scale 1. 2. 3. 4. 5. 6. 7. 8. 9.

Mem Diff Effort Atten Anger Symp Promo Raise Rate

1 — .46*** .04
.01 .01 .18 .10 .20 .02

2 .37*** — .11 .21
.05 .08
.15
.07
.06

3
.04
.10 — .67*** .30**
.32**
.18
.08
.17

4

5

6

.03 .05 .34** — .32**
.15
.06
.02
.08

.16
.22* .25* .09 —
.22*
.07
.14
.20

.11 .29**
.23*
.10
.27* — .09 .13 .13

7
.13 .05
.31**
.24*
.41*** .47*** — .56*** .50***

8
.06
.02
.31**
.21
.31** .47*** .65*** — .52***

9
.03 .06
.35*** .01
.24* .27* .52*** .47*** —

Notes: Correlations for young targets (n ¼ 84) are above diagonal; correlations for older targets (n ¼ 83) are below diagonal. Mem ¼ poor memory ability; Diff ¼ mental difficulty; Effort ¼ lack of effort; Atten ¼ lack of attention; Anger ¼ anger; Symp ¼ sympathy; Promo ¼ promotion recommendation; Raise ¼ raise recommendation; Rate ¼ work rating. *p , .05; **p , .01; ***p , .001.

Thus, counter to Weiner’s 1993 model, the effect of target age on sympathy is not mediated by the internal stable attributions. Rather, the effect of target age on judgments of sympathy operates independently of the internal stable attributions. Interestingly, however, some support for mediation is observed in the effect of target age on sympathy through perceivers’ judgments about an internal unstable attribution (lack of effort). To the extent that forgetful behavior and slow behavior are attributed to lack of effort, there will be less sympathy regardless of the target’s age-group membership. With regard to the effect of target age on anger, there was some support for mediation caused by lack of effort (but not lack of attention) based on the significance patterns of individual path coefficients. The direct path between target age and anger did not achieve statistical significance, which is consistent with the effect of target age on anger being mediated by the internal unstable attributions (most notably lack of effort): Higher attributions to lack of effort are associated with higher levels of anger. To ascertain whether target age moderates the relationship between perceivers’ reactions to and evaluations of the targets, we first computed Pearson r correlations between ratings of perceivers who read vignettes about young targets and ratings of perceivers who read vignettes about older targets. As shown in Table 2, sympathy was positively correlated with the evaluation variables for young but not older targets. Furthermore, anger was negatively correlated with the evaluation variables for young but not older targets. To further examine whether target age moderates the relationship between sympathy and evaluation, we entered target age (coded as 0 for young targets and as 1 for older targets), sympathy, and the Target Age 3 Sympathy interaction term into a regression analysis to predict each evaluation variable. The test for moderation was the significance of the interaction term (Jaccard & Turrisi, 2003). Target Age 3 Sympathy was significant for promotion (b ¼
.352, p , .009) and raise (b ¼
.353, p , .017), but not for work rating (b ¼
.101, p , .318). In the significant interactions (illustrated in Figure 2), the slopes for the effect of sympathy on promotion and raise for young targets are .429 and .475, respectively. For older targets, the slopes for the effect of sympathy on promotion and raise are .077 and .122, respectively. To determine whether target age moderates the relationship between anger and evaluation, we entered target age, anger, and

the Target Age 3 Anger interaction term as predictors of each evaluation variable. Target Age 3 Anger was significant for promotion (b ¼ .272, p , .034), but not for raise (b ¼ .157, p , .278) or work rating (b ¼ .029, p , .757). The slope for the effect of anger on promotion is
.331 for young targets and
.059 for older targets.

DISCUSSION Our findings are clear in demonstrating that perceivers use an age-based double standard when making attributions not only for employees’ memory failures but also for their slow behavior. Perceivers attribute forgetful and slow behavior more to internal stable causes when employees are older but more to internal unstable causes when employees are young. Furthermore, perceivers feel greater sympathy and less anger for older employees, and they evaluate older employees more leniently. We often evaluate people against a background of expectations about the group to which they belong (Jackson, Sullivan, & Hodge, 1993; Krueger, Heckhausen, & Hundertmark, 1995). Being forgetful and slow are stereotypes about older adults, which could explain why these behaviors are attributed more to internal stable causes when employees are older. Moreover, stereotypes can result in an ‘‘attribution of contrast,’’ whereby an individual considered typical of his or her group is judged less extremely than one considered atypical. Being forgetful or slow may not violate expectancies for older adults but young adults who behave the same way are not expected to, so they are reacted to more negatively and evaluated more stringently. Apparently, perceivers use shifting standards (Biernat, 2003; Biernat & Manis, 1994)—young employees who forget or are not quick enough are downgraded more than older employees are. Weiner’s 1993 model postulates that internal stable and internal unstable attributions for behavior are associated with sympathy and anger, respectively. However, our findings do not support the predicted stable attribution–sympathy relationship. A significant direct path between target age and degree of sympathy indicates that perceivers’ greater sympathy for the older targets is independent of their internal stable attributions (mental difficulty and memory ability), which is not in keeping with Weiner’s model. As suggested earlier, older adults who show signs of incompetence may be viewed as warm and thus

AGE AND PERCEPTIONS OF FORGETFUL AND SLOW EMPLOYEES

more sympathetic regardless of the specific attributions made for their behavior. With regard to the unstable attribution–anger relationship, Weiner’s model is a better fit, as evidenced by the finding that the effect of target age on anger is mediated by unstable attributions. Thus, attributing forgetful or slow behavior to lack of effort is associated with anger regardless of the target’s age group. When older adults are perceived as not trying hard enough, they are just as likely as young adults are to engender anger. Perceivers showed less leniency in evaluating young targets, as evidenced by their lower likelihood of recommending promotion and raise for young employees. However, the likelihood of promotion and raise has a stronger positive association with degree of sympathy that perceivers feel for young employees. There is also some, though weaker, evidence that evaluations have a stronger negative association with the degree of anger perceivers feel for young employees. In contrast, both sympathy and anger are less associated with evaluations of older employees, who seem to enjoy a somewhat protected status as far as the implications of perceivers’ affective reactions are concerned. Whether older employees are given more leeway than young employees with similar performance records merits further study. The evaluation of employees could be influenced by variables other than those measured here (e.g., managers’ beliefs about the need for a promotion or raise). On the basis of our findings, however, one could argue that older workers may be given an allowance when their performance is not up to par. This possibility could contribute to recent concerns about reverse discrimination (Hafemeister, 2003). In light of our perceivers’ relative leniency in evaluating older targets, should we assume that managers hold older employees less accountable for substandard job performance? Does this signify a ‘‘compassionate stereotype,’’ that is, concern and advocacy for older adults because of assumed helplessness or incompetence (Binstock, 1983; Revenson, 1989)? Alternatively, older workers may be credited with respect because of assumptions about positive qualities such as being responsible, sincere, and warm (Cuddy & Fiske, 2002; Cuddy et al., 2005; Erber & Szuchman, 2002). There was no support for the hypothesis that older employees in particular would be judged more leniently if they worked in an old (age-congruent) job setting. Surprisingly, work context held some importance, but only for young employees. Young e-commerce Internet employees engendered less sympathy than young library employees did. Perhaps expectations are especially high when young employees work in age-congruent jobs, so those who are forgetful or slow do not garner much sympathy. In pilot research, e-commerce and library were rated as young-relevant and old-relevant, respectively, but these jobs differ from the young- and old-relevant jobs in prior studies (e.g., Perry et al., 1996). Adding further to the complexity, work environments change over time, so what are considered young- and old-relevant jobs may not remain constant. Our results afford insight into how young versus older employees may fare when their performance is not ideal. Before drawing a firm conclusion that older employees will be given more latitude than will their younger counterparts, however, we must consider that our perceivers judged target employees on the basis of a supervisor’s thumbnail sketch, not

P337

Figure 2. Relationship between perceivers’ degree of sympathy and two evaluation variables (promotion and raise) as moderated by target age.

an uncommon practice in large organizations. Age may be highly salient when managers judge employees about whom they have minimal individuating information (Kite & Johnson, 1988). Immediate supervisors responsible for evaluating employees are likely to have more detailed knowledge about an employee’s job history, so age may hold less importance. Furthermore, our targets were men, who are more likely than women to work in full-time positions with fewer interruptions (Flippen & Tienda, 2000). However, more women are joining the labor force, and future research can test for any genderbased patterns of age discrimination (Kite & Wagner, 2002).

P338

ERBER AND LONG

Ongoing investigation is clearly warranted on the prospects of young versus older employees to determine whether, and under what circumstances, age discrimination or its reverse come into play. ACKNOWLEDGMENTS We gratefully acknowledge James Jaccard and Christian Resick for lending their statistical expertise, and we thank anonymous reviewers for their helpful comments on an earlier draft. Address correspondence to Joan T. Erber, Department of Psychology, Florida International University, University Park Campus, Miami, FL 33199. E-mail: [email protected] REFERENCES Baltes, P. B., & Lindenberger, U. (1997). Emergence of a powerful connection between sensory and cognitive functions across the adult life span: A new window to the study of cognitive aging? Psychology and Aging, 12, 12–21. Biernat, M. (2003). Toward a broader view of social stereotyping. American Psychologist, 58, 1019–1027. Biernat, M., & Manis, M. (1994). Shifting standards and stereotypebased judgments. Journal of Personality and Social Psychology, 66, 5–20. Binstock, R. H. (1983). The aged as scapegoat. The Gerontologist, 23, 136–143. Cerella, J. (1991). Age effects may be global, not local: Comment on Fisk and Rogers (1991). Journal of Experimental Psychology: General, 120, 215–223. Cerella, J. (1994). Generalized slowing and Brinley plots. Journal of Gerontology: Psychological Sciences, 49, P65–P71. Cleveland, J. N., Festa, R. M., & Montgomery, L. (1988). Applicant pool composition and job perceptions: Impact on decisions regarding an older applicant. Journal of Vocational Behavior, 32, 112–125. Cleveland, J. N., & Landy, F. J. (1987). Age perceptions of jobs: Convergence of two questionnaires. Psychological Reports, 60, 1075– 1081. Cuddy, A. J. C., & Fiske, S. T. (2002). Doddering but dear: Process, content, and function in stereotyping of older persons. In T. Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 3–26). Cambridge, MA: The MIT Press. Cuddy, A. J. C., Norton, M. I., & Fiske, S. T. (2005). The older stereotype: The pervasiveness and persistence of the elderly stereotype. Journal of Social Issues, 61, 267–285. Erber, J. T., & Szuchman, L. T. (2002). Age and capability: The role of forgetting and personal traits. International Journal of Aging and Human Development, 54, 173–189. Erber, J. T., Szuchman, L. T., & Rothberg, S. T. (1990). Everyday memory failure: Age differences in appraisal and attribution. Psychology and Aging, 5, 236–241. Finkelstein, L. M., Burke, M. J., & Raju, N. S. (1995). Age discrimination in simulated employment contexts: An integrative analysis. Journal of Applied Psychology, 80, 652–663. Finkelstein, L. M., Higgins, K. D., & Clancy, M. (2000). Justifications for ratings of old and young job applicants: An exploratory content analysis. Experimental Aging Research, 26, 263–283. Fisk, A., & Rogers, W. (1991). Toward an understanding of age-related memory and visual search effects. Journal of Experimental Psychology: General, 120, 131–149. Flippen, C., & Tienda, M. (2000). Pathways to retirement: Patterns of labor force participation and labor market exit among pre-retirement population by race, Hispanic origin, and sex. Journal of Gerontology: Social Sciences, 55B, S14–S27. Hafemeister, T. L. (September, 2003). Is age-based ‘reverse discrimination’ acceptable in the workplace? Judicial Notebook, APA Monitor on Psychology, 34, 66. Heckhausen, J., Dixon, R. A., & Baltes, P. B. (1989). Gains and losses in development throughout adulthood as perceived by different adult age groups. Developmental Psychology, 25, 109–121. Holm, S. (1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6, 65–70.

Jaccard, J., & Guilamo-Ramos, V. (2002). Analysis of variance frameworks in clinical child and adolescent psychology: Issues and recommendations. Journal of Clinical Child and Adolescent Psychology, 31, 130–146. Jaccard, J., & Turrisi, R. (2003). Interaction effects in multiple regression. Newbury Park, CA: Sage. Jackson, L. A., Sullivan, L. A., & Hodge, C. N. (1993). Stereotype effects on attributions, predictions, and evaluations: No two social judgments are alike. Journal of Personality and Social Psychology, 65, 69–84. Kite, M. E., & Johnson, B. T. (1988). Attitudes toward older and younger adults: A meta-analysis. Psychology and Aging, 3, 233–244. Kite, M. E., & Wagner, L. S. (2002). Attitudes toward older adults. In T. Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 129–161). Cambridge, MA: The MIT Press. Krueger, J., Heckhausen, J., & Hundertmark, J. (1995). Perceiving middleaged adults: Effects of stereotype-congruent and incongruent information. Journal of Gerontology: Psychological Sciences, 50B, P82–P93. MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. S. (2002). Comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7, 83–104. McCann, R., & Giles, H. (2002). Ageism in the workplace: A communication perspective. In T. Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 163–199). Cambridge, MA: The MIT Press. Pasupathi, M., & Lockenhoff, C. E. (2002). Ageist behavior. In T. Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 201–246). Cambridge, MA: The MIT Press. Perry, E. L., Kulik, C. T., & Bourhis, A. C. (1996). Moderating effects of personal and contextual factors in age discrimination. Journal of Applied Psychology, 81, 628–647. Revenson, T. A. (1989). Compassionate stereotyping of elderly patients by physicians: Revising the social contact hypothesis. Psychology and Aging, 4, 230–234. Salthouse, T. A. (1996). The processing-speed theory of adult age difference in cognition. Psychological Review, 103, 403–428. Shipley, W. C. (1940). A self-administering scale for measuring intellectual impairment and deterioration. The Journal of Psychology, 9, 371–377. Weiner, B. (1993). On sin versus sickness: A theory of perceived responsibility and social motivation. American Psychologist, 48, 957–965. Received June 15, 2005 Accepted April 17, 2006 Decision Editor: Thomas M. Hess

Appendix: Vignettes

Forgetful Target Sean (Mark) Smith is 28 (31) years old [or Robert (Harold) Smith is 61 (55) years old]. He has been working in the e-commerce graphics department [or the library circulation department] for about a year. I think you met him when you picked me up at our division to go to the seminar [or when you came to speak to our library staff] last spring. Sean (Mark) is about 59100, has wavy blond hair, and usually wears Docker pants and a striped shirt (is about 5980, has thick dark hair, and usually wears khaki pants and a light blue shirt) [or Robert (Harold) is about 59100, has thin gray hair, and usually wears a tan suit and a bow tie (is about 5980, is bald on top with white hair, and usually wears brown pants and a tan shirt)]. He always looks well groomed and he rarely misses a day of work. He usually contacts the retailers whenever we receive a new product [or He usually collects fines for overdue books when people try to check out new ones]. However, he does seem to be forgetful. Last week I reminded him we were opening early one day for an important inventory [or for a special event]. When I arrived, he

AGE AND PERCEPTIONS OF FORGETFUL AND SLOW EMPLOYEES

was still not there. When he finally arrived at the usual opening time, he seemed to have forgotten all about the important inventory [or special event]. On another occasion, I reminded him about an important staff meeting the next day. He did not attend. Later, someone from another department told me they had seen him in the employee lounge having coffee and doughnuts. I assume Sean [Mark, Robert, or Harold] forgot about the meeting.

Slow Target Sean (Mark) Smith is 28 (31) years old [or Robert (Harold) Smith is 61 (55) years old]. He has been working in the ecommerce graphics department [or the library circulation department] for about a year. I think you met him when you picked me up at our division to go to the seminar [or when you came to speak to our library staff] last spring. Sean (Mark) is about 59100, has wavy blond hair, and usually wears Docker pants and a striped shirt (is about 5980, has thick dark hair, and usually

P339

wears khaki pants and a light blue shirt) [or Robert (Harold) is about 59100, has thin gray hair, and usually wears a tan suit and a bow tie (is about 5980, is bald on top with white hair, and usually wears brown pants and a tan shirt)]. He always looks well groomed and he rarely misses a day of work. He usually contacts the retailers whenever we receive a new product [or He usually collects fines for overdue books when people try to check out new ones]. However, he does seem to be slow. Last week we got a new software shipment [or a new box of books arrived] and I asked him to process it [or them]. When I checked back later, he had still not finished. It took him a good part of the morning to finish a job that should have taken half the time. On another occasion, he went into the back room to get some sample software for one of our salesmen [or to get a book that was on reserve for one of our patrons]. He was gone much longer than should have been necessary. When he finally returned with the software [or book], he found that the salesman [or patron] had told one of the other employees [or librarians] that he could not wait. I assume the salesman [or patron] left because Sean (Mark, Robert, or Harold) was so slow.