A "Schema-Matching" Model of Occupational Gender Bias:
Trait and Behavior Components

Daniel Arkkelin
Valparaiso University
Presented at the Annual Meeting of the American Psychological Association, San Francisco, 1998

    Participants and Procedure: In Study 1, "personality profiles" were constructed which described a TP as scoring high/low on measures of masculine and feminine traits. The descriptions were attributed to a TP applying for a job that a pilot study indicated was perceived as low in both task and relations importance (bus driver), a low-task/high-relations job (sales clerk), a high-task/low-relations job (computer specialist), or a high-task/high-relations job (college professor). Responding to one TP in one job, 202 female and 103 male participants (Ps) made a hiring recommendation, rated their impressions of the TP's masculinity and femininity, and predicted the TP's job relations skills, task skills, and the probability that the TP was male. They also rated how masculine and feminine they perceived the job to be and estimated the percentage of male job incumbents. In Study 2, thirty two "candidate profiles" were constructed which described a male/female TP as scoring low/high on measures of job task and relations behaviors as well as scoring low/high on measures of masculine and feminine traits. The profiles were attributed to TPs applying for each of the above jobs. Responding to one of the jobs, 27 female and 19 male Ps made hiring decisions for each TP assuming that they would be either the TP's "supervisor" or "client" (i.e., passenger, customer, computer user or student).

    Results: In Study 1, separate 2 (High/Low TP Masculinity) x 2 (High/Low TP Femininity) x 2 (High/Low Job Task Importance) x 2 (High/Low Job Relations Importance) ANOVAs yielded main effects for both TP Masculinity and TP Femininity (see Table 1), indicating that high-masculine TPs were rated higher on task skills, lower on relations skills, as more masculine, less feminine, and more likely to be male than were low-masculine TPs. High-feminine TPs were rated lower on task skills, higher on relations skills, as less masculine, more feminine, and less likely to be male than were low-feminine TPs. Main effects for TP Femininity and Masculinity on hiring decisions indicated that a high level of either factor was desirable, but the Masculinity x Femininity interaction [F(1,289) = 12.80, p < .001] indicated Ps were more likely to hire androgynous TPs than masculine or feminine gender-typed TPs. Main effects for Job Task and Relations Importance indicated that high-task jobs were seen as more masculine, less feminine, and having a greater percentage of men than were low-task jobs. High-relations jobs were seen as more feminine and having a lower percentage of men than were low-relations jobs. Ps saw TPs applying for high-task and low-relations jobs as more likely to be male than TPs for low-task and high-relations jobs, respectively. Correlations (see Table 2) revealed that perceptions of TP masculinity/femininity were significantly related to inferred task/relations skills, and regression analyses indicated that inferred task/relations skills were the best predictors of hiring decisions.

    In Study 2, separate 2 (Low/High TP Task Skills) x 2 (Low/High TP Relations Skills) x 2 (Low/High TP Masculinity) x 2 (Low/High TP Femininity) x 2 (Male/Female TP) x 2 (Low/High Job Task Importance) x 2 (Low/High Job Relations Importance) mixed-model ANOVAs yielded main effects for TP Task Skills, TP Relations Skills, TP Masculinity and TP Femininity on both supervisors' and clients' hiring decisions (See Table 3), indicating that TPs scoring high on each of these dimensions were more likely to be hired than were TPs scoring low on these dimensions. However, numerous interaction effects indicated that a combination of high task-high masculinity was rated most favorably by supervisors, especially for candidates in the high-task job of computer specialist. Further, a combination of high relations-high femininity was rated most favorably by clients, especially for candidates in the high-relations job of sales clerk. Effect sizes indicated that TP task skills were the major determinant of supervisors’ hiring decisions (w² = .43), while TP relations skills were the major determinant of clients’ decisions (w² = .22).

    Conclusions: The cognitive links among components of the schema-matching model were clearly demonstrated in Study 1: TP masculinity/femininity triggered inferences regarding TP task/relations skills and TP sex, and job task/relations importance triggered inferences regarding job masculinity/femininity, TP sex and percentage of male incumbents. Hiring decisions were mediated primarily by rated TP task/relations skills. Study 2 also supported the schema-matching model, and indicated that exhibited task/relations behaviors are more important than masculine/feminine traits or TP sex in hiring decisions (cf.. Locksley et al., 1980), but that cognitive links do exist between these behavior/trait components of occupational gender schemas, especially when they are exhibited in stereotypic "appropriate" combinations in certain occupations. Finally, the links appear to vary according to the relative perspective of the perceiver (cf. Cann & Siegfried, 1987), with supervisors emphasizing task skills and clients emphasizing relations skills.

Arkkelin, D. & O'Connor, R. (1992). The "good" professional: Effects of trait-profile gender type, androgyny, and likableness on impressions of incumbents of sex-typed occupations. Sex Roles, 27, 351-358.

Cann, A., & Siegfried, W. D., Jr. (1987). Sex stereotypes and the leadership role. Sex Roles, 17(7/8), 401-408.

Deaux, K., & Lewis, L. L. (1984). Structure of gender stereotypes: Interrelationships among components and gender label. Journal of Personality and Social Psychology, 46, 991-1004.

Glick, P. (1991). Trait-based and sex-based discrimination in occupational prestige, occupational salary, and hiring. Sex Roles, 25, 351-378.

Locksley, A., Borgida, E., Brekke, N., & Hepburn, C. (1980). Sex stereotypes and social judgment. Journal of Personality and Social Psychology, 39(5), 821-831.

Table 1

Summary of Significant Main Effects and Relevant Marginal Means--Study 1

_______________________________________________________________________

I.V.s TP Masculinity TP Femininity

High Low F_1,289 High Low F_1,289

D.V.s

TP Task Skills 5.33 > 3.40 188.69^c 4.24 < 4.51 4.46^a

TP Rels Skills 4.06 < 4.54 13.24^c 5.56 > 3.05 334.36^c

TP Masculinity 4.41 > 3.76 23.83^c 3.64 < 4.53 44.67^c

TP Femininity 3.71 < 4.20 12.15^c 4.60 > 3.31 83.47^c

Hiring Decision 4.58 > 3.42 68.79^c 4.52 > 3.51 52.32^c

Prob (TP Male) 61.39 > 49.33 35.21^c 47.80 < 62.89 57.10^c

I.V.s Job Task Importance Job Relations Importance

High Low F_1,289 High Low F_1,289

D.V.s

Job Masculinity 4.11 > 3.81 4.43^a 3.87 = 4.05 1.61^ns

Job Femininity 3.56 < 3.93 7.45^b 4.06 > 3.45 21.39^c

% Men in Job 71.75 > 50.82 166.23^c 46.11 < 65.88 34.02^c

Prob (TP Male) 59.63 > 51.34 15.45^c 52.00 < 58.69 10.41^c

_______________________________________________________________________

Note.Low rating = 1; high rating = 7 for all D.V.s except p(TP is Male) and % Men in Job,

where low = 0; high = 100%. ^ap < .05 ^bp < .01 ^cp < .001 ^nsnonsignificant (p > .05)
 

TP Masc --
TP Fem -.51b --
TP Fem -.51b --
TP Task .26b -.12a --
TP Rels -.34b .48b -.05ns --
p(Male) .53b -.62b .24b -.38b --
Hire .04ns .17b .53b .50b .02ns -

 

Table 2

Intercorrelations Among Perceptions of TPs Across Jobs--Study 1

_______________________________________________________________________

TP Masc TP Fem TP Task TP Rels p(Male) Hire

TP Masc --

TP Fem -.51b --

TP Task .26b -.12a --

TP Rels -.34b .48b -.05ns --

p(Male) .53b -.62b .24b -.38b --

Hire .04ns .17b .53b .50b .02ns --

_______________________________________________________________________

Note. Two-tailed test, df = 303. ^ap < .05 ^bp < .01 ^nsnonsignificant (p > .05)
 
 

Table 3

Summary of Significant Main Effects and Relevant Marginal Means--Study 2

________________________________________________________________________

I.V.s Masculinity Femininity

High Low F_1,86 High Low F_1,86

D.V.s

Superior Hiring Decision 4.18 > 3.42 91.50 4.13 > 3.47 110.09

Client Hiring Decision 3.99 > 3.52 81.57 4.44 > 3.07 183.28

I.V.s Task Skills Relations Skills

High Low F_1,86 High Low F_1,86

D.V.s

Superior Hiring Decision 4.33 > 2.56 353.87 4.32 > 3.27 217.99

Client Hiring Decision 4.37 > 3.14 119.42 4.67 > 2.84 246.65
 
 

________________________________________________________________________

Note. Low rating = 1; high rating = 7 for all D.V.s. p < .001 for all F values