Race and the Workforce: Occupational Status, Aspirations, and Stereotyping Among African American Children

Developmental Psychology
Copyright 2003 by the American Psychological Association, Inc.
2003, Vol. 39, No. 3, 572–580
0012-1649/03/$12.00
DOI: 10.1037/0012-1649.39.3.572
Race and the Workforce: Occupational Status, Aspirations, and
Stereotyping Among African American Children
Rebecca S. Bigler and Cara J. Averhart
Lynn S. Liben
The University of Texas at Austin
The Pennsylvania State University
This study examined whether African American children’s perceptions of occupational status and their
own vocational interests are affected by racial segregation of the workforce. Children (N
92) rated
familiar occupations with respect to status, desirability, and stereotyping. Children also rated novel jobs
that had been depicted with African Americans, European Americans, or both African and European
Americans. As predicted, for familiar jobs, children’s judgments were linked to their knowledge of racial
segregation of these jobs. In addition, novel occupations that had been depicted with African Americans
were judged as lower in status than the identical occupations that had been depicted with European
Americans, demonstrating a causal influence of workers’ race on children’s judgments. Children’s age
and socioeconomic background moderated their occupational judgments.
Research indicates that African Americans have a long history
whether these stereotypes are related to children’s own occupa-
of occupational disadvantage in the United States relative to Eu-
tional aspirations. At the practical level, information about these
ropean Americans and that patterns of occupational inequality
factors is important for designing effective educational programs
between African Americans and European Americans continue
to expand individuals’ occupational opportunities. At the theoret-
today (Bound & Freeman, 1992; Danzinger & Gottschalk, 1991;
ical level, research on these topics is essential if the discipline is to
King, 1992). African Americans are underrepresented in high-
build a more complete database on normative development of
status skilled and managerial sectors and overrepresented in low-
African American children within “racialized” U.S. society (see
status service positions. For example, African Americans are em-
Garcia Coll et al., 1996; McLoyd & Randolph, 1985). In the
ployed at 40% the rate of European Americans in managerial,
present article, we report work on several issues pertaining to the
professional, and sales occupations but are employed at almost
ways in which the cultural context of work may affect African
twice the rate of European Americans in private household, ser-
American children. First, we explore whether African American
vice, and laborer occupations. Those middle-class African Amer-
children are sensitive to the racial segregation that characterizes
icans who do gain access to higher paying, prestigious jobs often
the U.S. workforce. In other words, we address the question of
face the “glass ceiling” effect, whereby access to the highest
whether African American children are knowledgeable about
echelons of most professions is blocked (Farley & Allen, 1987;
which occupations are more typically associated with European
Swinton, 1987; West, 1993). The Federal Glass Ceiling Commis-
Americans versus African Americans and, if so, whether they are
sion Report (1995), for example, revealed that African Americans
cognizant of the differential status of these jobs. Second, we
constitute 0.6% of senior-level managers in Fortune 1000 indus-
examine whether African American children form occupational
trial and Fortune 500 service companies.
stereotypes on the basis of race and whether racial distributions of
The occupational patterns of African Americans in the United
workers affect their own occupational judgments and aspirations.
States are likely to be relevant to the development of occupational
Little previous empirical or theoretical work has examined
aspirations in African American children and adolescents. Little
African American children’s understanding of the world of work
developmental research, however, has examined whether African
and whether—and if so, how—it is affected by racial group mem-
American children hold race-based occupational stereotypes or
bership. A statement written almost a decade ago remains true
today: “There has been almost no research on the way children
acquire, during childhood and adolescence, the values, attitudes,
Rebecca S. Bigler and Cara J. Averhart, Department of Psychology, The
and behavior patterns associated with ethnic group membership”
University of Texas at Austin; Lynn S. Liben, Department of Psychology,
(Phinney & Rotheram, 1993a, p. 291). Nearly all researchers
The Pennsylvania State University.
agree, however, that socialization processes lead African Ameri-
Portions of this work were presented at the biennial meeting of the
can children to absorb the culture’s ethnic attitudes unconsciously
Society for Research in Child Development, Washington, DC, March
(see Aboud, 1988; Phinney & Rotheram, 1993b; Williams &
1997.
Morland, 1976). Support for this contention comes from studies of
We thank the children and staff who graciously agreed to participate in
African American children’s knowledge and internalization of
this research.
European Americans’ broad stereotypes of African Americans.
Correspondence concerning this article should be addressed to Rebecca
S. Bigler, Department of Psychology, 1 University Station A8000, The
Considerable research indicates that African American children
University of Texas, Austin, Texas 78712. E-mail: bigler@psy.utexas.edu
are knowledgeable about the racial stereotypes of traits held by
572

---

RACE AND THE WORKFORCE
573
European Americans. When presented with the forced choice of
design are particularly important. First, in addition to studying
assigning traits to racial groups (or to African American and
children’s beliefs about occupational status, aspirations, and ste-
European American individuals), African American children are
reotyping of familiar jobs that vary with respect to racial compo-
much more likely to assign positive traits (e.g., smart, friendly) to
sition in U.S. society, we also studied children’s responses to novel
Whites and negative traits (e.g., lazy, mean) to Blacks than the
jobs. By varying (between children) the race of workers depicted
reverse (Aboud, 1988; Spencer & Markstrom-Adams, 1990).
in these novel occupations, we could examine whether race influ-
Unlike the large literature on trait-based racial stereotyping,
enced children’s judgments about and interest in jobs that were
there has been little work on occupation-related racial stereotyp-
otherwise identical. Second, we examined effects in relation to
ing. Yet it seems likely that African American children develop an
important variables within an African American sample. Specifi-
awareness of race-related occupational stereotypes. Stereotypes
cally, we included both lower and higher SES African American
are thought to reflect the distribution of social groups into societal
children to permit us to distinguish between effects that might be
roles (e.g., Eagly & Steffen, 2000), and African American children
linked to social class from those that might be linked to race, and
are exposed to a distribution of occupational roles in which race
we included both younger and older children.
and status are correlated. This is likely to be true within both
Participants were African American children aged 6 –7 and
children’s own environmental context and the broader U.S. cul-
11–12 years from a racially diverse school representing middle-
ture. High-status African American occupational models are un-
and lower class populations. The younger age was selected be-
likely to be available to many African American children in their
cause it allowed us to test children who are younger than those
own families and communities, especially among children from
included in previous samples (e.g., Cook et al., 1996) but who
lower socioeconomic backgrounds. Further, media portrayals in-
nevertheless already have considerable understanding of a wide
clude race-related information that reaches a broad range of chil-
variety of adult occupational roles and occupational status in
dren and can affect their racial attitudes (see Cortes, 2000; Graves,
general (McGee & Stockard, 1991; Simmons, 1962). The older
1996, 1999). Analyses of prime time and Saturday morning tele-
range was selected because it spans an age at which cognitive
vision programming indicate that despite increases in the number
skills can be expected to have developed significantly and is an age
and diversity of African American television characters, African
that marks a transitional period into curriculum and job choices
American actors and actresses are still shown primarily in low-
(e.g., part-time or summer work) that may privilege or constrain
status jobs (Children NOW, 1998; Greenberg & Brand, 1993;
later occupational choices.
Greenberg & Collette, 1997).
Although little work has examined occupational stereotyping,
Method
many studies have examined the occupational aspirations of Afri-
can American children and adolescents (e.g., Cook et al., 1996).
Participants
Findings suggest that African American students have relatively
Participants were 92 African American children (47 girls and 45 boys)
modest occupational goals compared with European American
recruited from the first and sixth grades of a racially heterogeneous (72%
students (e.g., Hale, 1980). Most African American students select
African American) school in the Midwest. For Grade 1, the mean age of
occupations in which African Americans traditionally have been
participants was 7 years 2 months (range
6 years 4 months to 7 years 11
well represented, such as military personnel, teacher, postal em-
months). For Grade 6, the mean age was 12 years 4 months (range
11
ployee, hair stylist, or cafeteria worker (Terrell, Terrell, & Miller,
years 4 months to 12 years 9 months). Approximately half of the partici-
1993). Among some samples, the tendency to aspire to lower
pants were from lower SES backgrounds and half were from upper-middle
status occupations is present by the end of second grade (Cook et
SES backgrounds, according to information provided by the schools about
al., 1996). African American students’ lower status occupational
students’ qualification for free and reduced-price meal programs.
aspirations have been attributed to students’ assessment of poten-
tial limitations afforded by their current socioeconomic status
Procedural Overview
(SES; e.g., Hale, 1980). Their lower status aspirations have also
been attributed to their awareness that they are likely to encounter
Parental permission forms were distributed to children in the first and
sixth grades. Only those children whose parents returned permission slips
prejudicial attitudes in occupational settings that are dominated by
and who themselves agreed to participate were included in the study
out-group members (Spencer, Dobbs, & Swanson, 1988; Spencer
(roughly 50% of the eligible children). Children were interviewed individ-
& Markstrom-Adams, 1990).
ually in a quiet room near their classroom by one of four African American
Previous empirical studies of occupational aspirations are char-
experimenters (two male and two female). In a first session, children were
acterized by several limitations, however. One problem is that race
given measures designed to assess their perceptions of occupational status
and SES have typically been confounded such that lower class
and their occupational aspirations. In a second session, they were given
African American children’s goals and expectations are compared
measures designed to assess individual differences in (a) knowledge of
with those of middle- and upper class European American children
racial stereotypes of occupations, (b) endorsement of racial stereotypes of
(e.g., Cook et al., 1996). In addition to these confounds between
occupations, and (c) endorsement of majority-culture trait stereotypes of
race and economic status, past researchers have paid relatively
African Americans. The racial stereotyping measures were always given
following children’s ratings of occupational status and aspirations. This
little attention to the ways in which occupational aspirations may
order was used because the stereotyping measures explicitly identify race
be shaped by social– cognitive variables (e.g., race-related stereo-
as a critical factor and thus, if given first, might have primed children to
types, differential availability of role models of different races).
attend to race in their status and aspiration judgments. The two measures
The primary purpose of the present study was to provide em-
of children’s endorsement of stereotypes showed no variability of respond-
pirical data that begin to examine the role of race in African
ing among older children and very little variability among younger children
American children’s occupational judgments. Two features of our
(virtually all children rejected racial stereotypes). In light of this finding,

---

574
BIGLER, AVERHART, AND LIBEN
and given that these two measures had been given after all other measures
a(n) _____?” Participants responded using a 5-point scale, ranging from 1
and could thus not have affected other results, we omit any further
(not at all) to 5 (very much), again supplemented by the thermometer
discussion of these individual-differences measures.
graphics for the younger children.
Knowledge of occupational stereotypes.
To assess children’s knowl-
edge of the racial composition of familiar occupations, as well as to elicit
Occupational Materials
their beliefs about the racial composition of the novel occupations, we
asked them, “Who usually does the job of being a(n) ______?” Response
Each of the occupational scales (i.e., status, aspirations, and stereotyp-
options included only Black people, only White people, and both Black and
ing) contained 39 occupations. Of these, 27 were familiar occupations
White people.
sampled from Swinton’s (1987) categories of high- and medium-status
occupations in which African Americans are underrepresented and from
lower status occupations in which African Americans are overrepresented
Results
in comparison with European Americans (see Appendix). The remain-
Overview of Analyses
ing 12 were novel occupations, that is, occupations that participants would
not have known about previously. The job titles and job descriptions for the
We present results in two major sections, covering ratings of
novel occupations were taken from Liben, Bigler, and Krogh (2001).
familiar occupations and ratings of novel occupations. Because
Included were 6 obscure job titles that testing had shown were unknown to
cell sizes would become unacceptably small if all potential
children and even to most adults (e.g., higgler, “a person who sells items
between-group variables were included in the analysis of variance
such as clothing, watches, or candy on the street. They carry the things they
(ANOVA), we conducted preliminary ANOVAs to test for possi-
are selling along with them in a cart so that they can move up and down the
ble effects of participant sex. Results indicated no significant main
street selling their goods”) and 6 newly coined job titles used to refer to
some work activity not already explicitly identified as a job in our culture
effects or interactions involving sex of participant, and thus, data
(e.g., tenic, “a person in charge of creating handicapped parking places for
were pooled for boys and girls in the analyses reported below. In
city buildings and stores. They decide how many handicapped parking
all cases, significant ANOVA effects were followed by compari-
places there should be, where they should go, and supervise to make sure
sons across means using the Newman–Keuls method.
that the signs are put up correctly”).
Pictorial illustrations of these jobs were based on those used by Liben et
Familiar Occupations
al. (2001), except that rather than varying the depicted sex of workers as
had been done in that research, we varied the depicted race of workers. By
Occupational
status.
As
explained
earlier,
participants
manipulating the race of the workers who were depicted in the novel jobs,
rated 27 familiar occupations for (a) difficulty to learn, (b) diffi-
we could experimentally test the effect of a worker’s race on children’s
culty to perform, (c) pay, and (d) importance. As a preliminary
perceptions of jobs without the extant confounds between jobs and race in
analysis, we began by examining correlations between all pairs of
the United States.
questions. As expected, all pairs were significantly correlated (all
At the start of the session, children were told that the experimenter was
ps
.01). Thus, we averaged responses to each of the four
interested in their opinions about various occupations. The child was told
questions to create a composite status score that yielded a high
that although many of the jobs would be familiar, some would be unfa-
miliar and that the experimenter had thus brought pictures to explain these.
Cronbach’s alpha (.82). This composite score then served as the
Familiar and novel jobs were interspersed in one randomly ordered list, and
dependent measure in a 3 (job type: high, medium, and low
whenever one of the 12 novel occupations was encountered, children were
status)
2 (age: 6 –7 vs. 11–12)
2 (SES: higher vs. lower)
read a standard job description (as in the higgler and tenic examples given
repeated measures ANOVA, with job type as a within-subject
above) and shown colored drawings of two men and two women perform-
variable. Because the Mauchley sphericity test was significant
ing the job. Children were randomly assigned to conditions in which the
(i.e., the homogeneity of variance assumption was violated), the
workers shown in the drawings were either (a) four African Americans, (b)
Geisser–Greenhouse procedure was used to adjust the degrees of
four European Americans, or (c) two African Americans and two European
freedom for the F tests involving the within-subject variable (i.e.,
Americans. Job descriptions (minus pictures) were also prepared for all
job type). Table 1 presents means and standard deviations for
familiar jobs in case children expressed uncertainty about them. For
status ratings of familiar jobs, broken down by participant age and
younger children, this procedure was typically necessary two or three times
per child, most often for the jobs of bank teller, business executive, and
SES.
politician. For older children, no descriptions of familiar jobs were needed.
As predicted, the main effect of job type was significant,
F(1.27, 99.14)
1,889.62, p
.01. Comparisons among means
indicated that children rated high-status occupations as higher in
Dependent Measures
status than the medium-status occupations, which in turn were
Ratings of occupational status.
To assess children’s ratings of occu-
rated as higher in status than low-status occupations. Also signif-
pational status, we asked them four questions about each of the 39 jobs, in
icant was the two-way interaction of participant age and SES, F(1,
the following order: (a) “How hard do you think it is to learn to be a(n)
78)
9.73, p
.01. Post hoc analyses showed that within the
_____?”; (b) “How hard do you think it is to do the job of being a(n) _____
older age group, familiar occupations were rated as lower in status
every day?” (c) “How much money do you think a(n) _____ gets paid?”;
by children from higher SES families than by children from lower
and (d) “How important is the job of being a(n) _____?” Children re-
SES families. Within the younger age group, however, ratings did
sponded using a 5-point scale. For younger children, the numerical scale
not differ by SES. Subsumed by the interaction were significant
was supplemented by a graphic response format (thermometers indicating
main effects of participant age, F(1, 78)
17.77, p
.01, and
increasing levels of mercury). Response options included 1 (none or not at
all), 2 (a little. . . or a little bit), 3 (medium. . . or a medium amount), 4
participant SES, F(1, 78)
5.49, p
.05. Younger children
( pretty. . . or pretty much), and 5 (very. . . or very much).
(compared with older children) and children from lower SES
Occupational aspirations.
To assess children’s desire to perform each
backgrounds (compared with children from higher SES back-
of the 39 occupations, we asked them, “How much would you like to be
grounds) rated familiar occupations as higher in status.

---

RACE AND THE WORKFORCE
575
Table 1
way interactions. First, there was a significant interaction of job
Mean Composite Status Ratings (and Standard Deviations) for
type and participant SES, F(1.45, 117.73)
11.58, p
.01. Post
Familiar Occupations by Participant Age and Socioeconomic
hoc comparisons indicated that although overall, children showed
Status (SES)
greater interest in high-status than in medium- and low-status
occupations and higher interest in medium- than in low-status
Occupational status
occupations, this pattern was more pronounced among children
Age and
from higher than from lower SES families. Second, there was a
SES group
High
Medium
Low
significant two-way interaction of age and participant SES, F(1,
Age 6–7
81)
11.67, p
.01. Post hoc comparisons indicated that among
Higher SES
4.69 (0.28)
3.67 (0.21)
2.04 (0.35)
older (but not younger) children, higher SES children expressed
Lower SES
4.63 (0.22)
3.63 (0.22)
2.11 (0.25)
Combined
4.66 (0.25)
3.65 (0.22)
2.07 (0.31)
greater interest in occupations, overall, than did lower SES chil-
Age 11–12
dren. Further subsumed by these two-way interactions was a
Higher SES
4.42 (0.25)
3.36 (0.15)
2.04 (0.36)
significant main effect of job type, F(1.45, 117.73)
467.35, p
Lower SES
4.70 (0.19)
3.49 (0.12)
2.03 (0.32)
.01, with participants showing greater interest in high-status than
Combined
4.56 (0.26)
3.42 (0.15)
2.03 (0.33)
Combined ages
4.61 (0.26)
3.53 (0.22)
2.05 (0.32)
in medium- or low-status occupations and greater interest in
medium- than in low-status occupations.
Note.
Scores ranged from 0 to 5, with higher numbers indicating higher
Knowledge of occupational stereotypes.
To examine whether
status.
participants were knowledgeable about which racial group was
typically associated with the high-, medium-, and low-status jobs,
Occupational aspirations.
Participants’ ratings of their own
we counted the number of times that children stated that African
interest in performing familiar jobs were analyzed using a 3 (job
Americans usually performed each of the familiar occupations. A 3
type: high, medium, and low status)
2 (age: 6 –7 vs. 11–12)
2
(job type: high, medium, and low status)
2 (age: 6 –7 vs.
(SES: higher vs. lower) repeated measures ANOVA, with job type
11–12)
2 (SES: higher vs. lower) ANOVA indicated a signifi-
as a within-subject variable. Results indicated a significant three-
cant
interaction
between
job
type
and
SES,
F(1.79,
way
interaction
of
job
type,
age,
and
SES,
F(1.45,
146.86)
4.62, p
.05. Means and standard deviations are shown
117.73)
11.06, p
.01. Means and standard deviations are
in Table 3. Post hoc comparisons among means indicate that
shown in Table 2. Post hoc comparisons among means indicated
children from lower SES backgrounds were more likely than those
that among younger children, children from both higher and lower
from higher SES backgrounds to believe that African Americans
SES families showed significantly more interest in high-status than
performed low- and medium-status jobs; means (and standard
in medium- or low-status jobs and more interest in medium- than
deviations) were 4.77 (2.76) and 2.02 (1.50), respectively, for
in low-status jobs. Among older children, the pattern differed by
lower SES children and 1.84 (1.96) and 0.34 (0.68) for higher SES
SES. Specifically, children from higher SES families showed the
children. In contrast, both higher and lower SES children agreed
same pattern just described for younger children, although the
that African Americans were unlikely to perform high-status oc-
differential interest by job status was less pronounced. Children
cupations; means (and standard deviations) were 0.03 (0.15)
from lower SES families showed no significant difference in their
and 0.02 (0.15), respectively. Subsumed by the interaction was a
interest in high- and medium-status jobs. Furthermore, these chil-
significant main effect of job type, F(1.29, 105.01)
137.47, p
dren exhibited a much smaller (although still significant) differ-
.01. Children reported that African Americans were significantly
ential interest in medium- versus low-status jobs.
less likely to perform high-status than medium- and low-status
Subsumed by the three-way interaction were significant two-
jobs and less likely to perform medium-status than low-status jobs.
Table 2
Table 3
Mean Ratings (and Standard Deviations) of Interest in Familiar
Mean Ratings (and Standard Deviations) of “Who Usually”
Occupations by Participant Age and Socioeconomic Status
Performs Familiar Occupations by Participant Age and
(SES)
Socioeconomic Status (SES)
Occupational status
Occupational status
Age and
Age and
SES group
High
Medium
Low
SES group
High
Medium
Low
Age 6–7
Age 6–7
Higher SES
4.01 (0.42)
2.63 (0.45)
1.16 (0.20)
Higher SES
0.04 (0.21)
0.31 (0.64)
1.28 (1.79)
Lower SES
3.98 (0.44)
2.63 (0.42)
1.14 (0.20)
Lower SES
0.00 (0.00)
2.19 (1.33)
4.61 (3.09)
Combined
4.00 (0.43)
2.63 (0.44)
1.15 (0.20)
Combined
0.02 (0.15)
1.23 (1.39)
2.95 (3.01)
Age 11–12
Age 11–12
Higher SES
3.18 (0.57)
2.11 (0.38)
1.14 (0.16)
Higher SES
0.00 (0.00)
0.73 (0.36)
2.36 (2.01)
Lower SES
1.85 (1.32)
1.71 (0.55)
1.09 (0.23)
Lower SES
0.04 (0.21)
1.86 (1.67)
4.91 (2.47)
Combined
2.49 (1.22)
1.90 (0.51)
1.11 (0.20)
Combined
0.02 (0.14)
1.11 (1.48)
3.64 (2.57)
Combined ages
3.26 (1.17)
2.26 (0.60)
1.13 (0.20)
Combined ages
0.02 (0.15)
1.17 (1.43)
3.30 (2.80)
Note.
Scores ranged from 0 to 5, with higher numbers indicating greater
Note.
Scores are the number of “only Black people” responses of 9
interest.
possible.

---

576
BIGLER, AVERHART, AND LIBEN
Ratings of Novel Occupations
European Americans were rated significantly higher in status than
jobs depicted with only African Americans. Among older children,
Occupational status.
As was the case for familiar jobs, we
jobs depicted with only European Americans were rated signifi-
began by examining the relation among ratings on the four status
cantly higher in status than jobs depicted with only African Amer-
questions for novel jobs. As expected, all pairs were significantly
icans (but not higher than jobs depicted with both African Amer-
correlated (all ps
.01), and thus, we again created a single
icans and European Americans).
composite score that had a high Cronbach’s alpha (.87). This score
The two-way interaction of SES and age was also significant,
then served as the dependent measure in a 3 (picture condition:
F(1, 72)
8.17, p
.01. Comparisons among means indicated
only African American, only European American, or both African
that among younger children, higher and lower SES children rated
American and European American)
2 (age: 6 –7 vs. 11–12)
2
the novel occupations similarly, whereas among older children,
(SES: higher vs. lower) ANOVA. Means and status ratings for
lower SES children rated the novel occupations as higher in status
novel jobs are presented in Table 4, again divided by participant
than did higher SES children.
age and SES.
Subsumed by the two-way interactions were main effects of
Results indicated a significant three-way interaction of picture
picture condition, F(2, 72)
79.97, p
.01; age, F(1,
condition, age, and SES, F(2, 72)
9.33, p
.01. Post hoc
72)
37.11, p
.01; and SES, F(1, 72)
15.25, p
.01.
comparisons among means revealed that among younger children,
Comparisons among means indicated that children rated jobs de-
both higher and lower SES children rated jobs depicted with only
European Americans as significantly higher in status than jobs
picted with only European Americans and jobs depicted with both
depicted with only African Americans. Among older children,
African Americans and European Americans as significantly
lower (but not higher) SES children rated jobs depicted with only
higher in status than jobs depicted with only African Americans. In
European Americans as significantly higher in status than jobs
addition, children rated jobs depicted with only European Amer-
depicted with only African Americans. The identical pattern
icans as significantly higher in status than jobs depicted with both
emerged for comparisons between jobs depicted with only Euro-
African Americans and European Americans. Overall, younger
pean Americans and those jobs depicted with both European
children rated the novel occupations as higher in status than did
Americans and African Americans. That is, in the younger age
older children, and lower SES children rated the novel occupations
group, both lower and higher SES children rated jobs depicted with
as higher in status than did higher SES children.
only European Americans as higher in status than jobs depicted
Occupational aspirations.
Data from participants’ ratings of
with both African Americans and European Americans. In the
their own interest in performing novel jobs were analyzed with a 3
older age group, lower (but not higher) SES children rated jobs
(picture condition: only African Americans, only European Amer-
depicted with European Americans as significantly higher in status
icans, or both African Americans and European Americans)
2
than jobs depicted with both African Americans and European
(age: 6 –7 vs. 11–12)
2 (SES: higher vs. lower) ANOVA.
Americans.
Results indicated a significant interaction of picture condition and
Subsumed by the three-way interaction was the two-way inter-
SES, F(2, 72)
14.23, p
.01. Means and standard deviations are
action of picture condition and age, F(2, 72)
12.97, p
.01.
presented in Table 5. Post hoc comparisons among means indi-
Post hoc comparisons indicated that among younger children, jobs
cated that children from higher SES backgrounds showed signif-
depicted with only European Americans were rated significantly
icantly more interest in occupations depicted with only European
higher in status than jobs depicted with only African Americans or
Americans and jobs depicted with both African Americans and
jobs depicted with both African Americans and European Amer-
European Americans than they did in occupations depicted with
icans. In addition, jobs depicted with both African Americans and
only African Americans; means (and standard deviations)
were 2.94 (0.51), 2.88 (0.75), and 2.09 (0.48), respectively. In
contrast, children from lower SES backgrounds showed no signif-
Table 4
icant differences in their level of interest across the three picture
Mean Composite Status Ratings (and Standard Deviations) for
conditions, with means (and standard deviations) for the only
Novel Occupations by Participant Age and Socioeconomic
European American, both African American and European Amer-
Status (SES)
ican, and only African American conditions, respectively, 2.19
(1.30), 2.41 (1.02), and 2.45 (0.81).
Picture condition
The interaction of picture condition and age was also significant,
Age and
African
European
F(2, 72)
13.64, p
.01. Comparisons among means indicated
SES group
American
American
Both
that younger (but not older) children showed significantly more
interest in occupations depicted with only European Americans
Age 6–7
Higher SES
2.50 (0.39)
4.01 (0.19)
3.45 (0.73)
and in occupations depicted with both African Americans and
Lower SES
2.77 (0.14)
4.03 (0.29)
3.36 (0.29)
European Americans than they did in occupations depicted with
Combined
2.62 (0.33)
4.01 (0.23)
3.41 (0.34)
only African Americans. The two-way interaction of SES and age
Age 11–12
was also significant, F(2, 72)
45.50, p
.01. Post hoc com-
Higher SES
2.75 (0.40)
2.85 (0.20)
2.61 (0.19)
Lower SES
2.62 (0.41)
3.75 (0.16)
3.16 (0.09)
parisons indicated that among younger children, lower SES chil-
Combined
2.68 (0.39)
3.26 (0.50)
2.84 (0.32)
dren showed more interest in the novel occupations than higher
Combined ages
2.65 (0.36)
3.67 (0.54)
3.15 (0.43)
SES children; means (and standard deviations) were 3.37 (0.33)
and 2.93 (0.75), respectively. Among older children, lower SES
Note.
Scores ranged from 0 to 5, with higher numbers indicating higher
status.
backgrounds showed less interest in the novel occupations than did

---

RACE AND THE WORKFORCE
577
Table 5
were significantly more likely to say that jobs were performed by
Mean Ratings (and Standard Deviation) of Interest in Novel
“only Black people” when the jobs asked about were low status
Occupations by Participant Age and Socioeconomic Status
(and, indeed, are held by a high proportion of African Americans
(SES)
in this society) than when the jobs were high status (performed by
high proportions of European Americans).
Picture condition
It seems likely that awareness of the different statuses of occu-
pations that are dominated by European Americans rather than
Age and
African
European
African Americans is one aspect of a broad race-related occupa-
SES group
American
American
Both
tional schema developed by African American children. Such a
Age 6–7
schema is likely to influence children’s predictions about the status
Higher SES
2.07 (0.57)
3.34 (0.26)
3.39 (0.50)
and desirability of both familiar and unfamiliar jobs. Consistent
Lower SES
3.25 (0.31)
3.58 (0.40)
3.29 (0.25)
with this notion, a study by Averhart and Bigler (1997) demon-
Combined
2.56 (0.76)
3.45 (0.34)
3.33 (0.37)
Age 11–12
strated that African American children had better memory for
Higher SES
2.11 (0.47)
2.48 (0.27)
2.17 (0.26)
stories in which light-complexioned African Americans performed
Lower SES
1.95 (0.58)
1.14 (0.35)
1.40 (0.36)
high-status occupations (and dark-complexioned African Ameri-
Combined
2.03 (0.49)
1.77 (0.75)
1.72 (0.50)
cans performed low-status occupations) than stories in which
Combined ages
2.26 (0.67)
2.58 (1.03)
2.62 (0.92)
dark-complexioned African Americans performed high-status oc-
Note.
Scores ranged from 0 to 5, with higher numbers indicating greater
cupations (and light-complexioned African Americans performed
interest.
low-status occupations).
Findings from children’s ratings of their own interest in per-
forming the familiar occupations are highly consistent with previ-
higher SES children; means (and standard deviations) were 1.50
ous studies (e.g., Cook et al., 1996). The present data showed that
(0.55) and 2.25 (0.36), respectively.
younger children from both higher and lower SES backgrounds
Subsumed by the two-way interactions were main effects of
showed high levels of interest in the high-status occupations.
picture condition, F(2, 72)
3.95, p
.05, and age, F(1,
Among 11–12-year-old children, however, there was evidence of
72)
3.16, p
.01. Post hoc comparisons among means indicated
lower occupational goals among participants with lower SES back-
that children showed greater interest in occupations depicted with
grounds. The significant effect of SES among older children pro-
only European Americans and in occupations depicted with both
vides support for the theory that economic variables may be
African Americans and European Americans than they did in
important in shaping older children’s occupational plans and as-
occupations depicted with only African Americans and that, over-
pirations (Hale, 1980).
all, younger children expressed greater interest in novel occupa-
Interestingly, the occupational interests of these African Amer-
tions than did older children (Ms [SDs]
3.11 [0.6] and 1.84
ican children do not appear to be affected by stereotypic beliefs
[0.6]).
concerning the appropriateness of various occupations for Euro-
pean American and African American individuals. When children
were asked which racial group(s) “should” perform the familiar
Discussion
occupations, participants responded in a highly unbiased manner,
Results from the present study offer a number of important
almost invariably answering that “both Whites and Blacks” should
contributions to understanding the effects of occupational segre-
perform all occupations.
gation of the workforce on African American children’s judgments
The most compelling evidence of the role played by race in
of occupational status and occupational aspirations. Overall, the
African American children’s assessment of occupational status,
data suggest that race has consistent and powerful effects on
however, is the set of data derived from novel occupations. These
African American children’s perceptions of occupations. In the
data are particularly useful because they avoid the usual problem
discussion below, we first address implications of children’s rat-
of separating effects of workers’ race from many other job qual-
ings of familiar occupations and then consider implications of
ities. Most critically, even though the jobs themselves were iden-
findings from the experimental manipulation of race in the depic-
tical across conditions, children’s ratings of the status of novel
tion of novel occupations.
occupations differed in relation to whether the jobs were depicted
Data from the familiar occupations indicated that children ac-
with only European American, only African American, or both
cord higher status to those jobs that have high concentrations of
European American and African American workers. Specifically,
European Americans (and low concentrations of African Ameri-
African American children rated occupations that had been de-
cans) than to those jobs with low concentrations of European
picted with only European American workers as being higher in
Americans (and high concentrations of African Americans). This
status than the identical occupations depicted with only African
finding held for both lower and higher SES African American
American workers. Across age and socioeconomic backgrounds,
children. Furthermore, the finding of differential status ratings was
children rated occupations that were performed solely by African
obtained even in children as young as 6. It is possible, of course,
Americans as particularly low in status. These data indicate clearly
that race per se is not the primary factor responsible for the
that race has an independent effect on occupational judgments and
differential ratings of these jobs. However, children’s responses to
thus that it cannot be only the qualities inherent in occupations
questions about who “usually does” these jobs revealed that even
themselves that affect children’s judgments about job status. In
6-year-olds were highly knowledgeable about which racial group
short, the data support the conclusion that solely as a function of
members are typically associated with which jobs. That is, children
workers’ race, jobs performed by African Americans are viewed

---

578
BIGLER, AVERHART, AND LIBEN
as lower in status (e.g., less difficult to learn and perform, receiv-
The relation between race of workers and ratings of status found
ing lower pay, and being less important) than the very same
here is troubling given the potential for two types of vicious cycles
occupations performed by European Americans.
with respect to race and jobs. First, African American children,
The data derived from the experimental manipulations of race of
especially those from disadvantaged backgrounds, may preferen-
worker also showed that depicted race affected African American
tially seek out low-status jobs in which minorities are well repre-
children’s own desires to perform various occupations. Overall,
sented and thereby ensure that such jobs remain overpopulated by
children rated their own level of interest in performing the depicted
minorities, thus perpetuating the skewed models for new genera-
novel occupations as significantly lower when the occupations had
tions of poor African American children. Second, those medium-
been depicted as performed exclusively by African Americans
and high-status jobs that do attract an increasing proportion of
than when the occupations had been depicted as performed exclu-
African American workers (perhaps from more advantaged house-
sively by European Americans or by both African Americans and
holds) may, across time, be viewed as lower in status simply as a
European Americans. Again, these data are particularly informa-
function of the race of the worker, and consequently show decreas-
tive in that they clearly indicate that the race of occupational
ing levels of pay and prestige. Indeed, many people have argued
models—apart from the content and activities of the occupations
that a similar cycle of decreasing pay and prestige characterizes
themselves (which were held constant)—affects children’s own
those occupations that show increasing proportions of female
occupational interests.
versus male workers (see, e.g., American Psychological Associa-
In addition to demonstrating the importance of racial cues in
tion, 1995).
American children’s judgments about occupations at the group
These data also attest to the importance of attending to hetero-
level, the present data also contribute to our understanding of
geneity within African American children. Beliefs about race and
developmental and group differences in African American chil-
the workforce seem to change with age. Among older children,
dren’s occupational stereotyping, judgments, and goals. Consistent
beliefs about race and the status and desirability of occupations are
with Hale’s (1980) reality theory, socioeconomic factors do appear
especially diverse. The present findings suggest that it may be
to influence children’s occupational judgments, particularly at
useful to devise interventions that address children’s knowledge
adolescence. Among the 11–12-year-olds, those children from
about race and the workforce, as well as how racial schemas affect
lower SES backgrounds showed less interest in performing famil-
their aspirations to perform certain occupations in the future (see,
iar occupations that were populated primarily by European Amer-
e.g., Aboud & Levy, 1999). For example, such interventions might
icans and hence that were associated with higher status. It is
explicitly address the absence or low percentage of African Amer-
possible that economic factors are important in this decision, with
icans among high-status jobs presented routinely in school settings
(e.g., president of the United States).
lower SES children realizing that they lack the financial resources
Finally, it is important to note that this study represents only one
to pursue jobs that require a good deal of education beyond high
step in the process of understanding the effects of race on occu-
school.
pational judgments. SES is only a rough marker of the family
Importantly, significant interactions between participants’ SES
experiences that may contribute to occupational stereotyping and
background and age were found for the way that children rated the
aspirations. Furthermore, the marker of SES used here (i.e., qual-
desirability of novel jobs. Consistent with the pattern for familiar
ification for free or reduced-priced school lunches) is not ideal for
occupations, older children from higher— but not lower—SES
tapping family income (see Hauser, 1994). Future work should
backgrounds were significantly more interested in performing
examine more closely how financial constraints shape vocational
novel occupations that they saw performed solely by European
goals and expectations, including consideration of how families
Americans or by both African Americans and European Ameri-
discuss occupations, the types of role models provided by parents’
cans than they were in occupations that they saw performed solely
own work, and how school and community contexts contribute to
by African Americans. Obviously, it is impossible to determine the
occupational stereotyping. It will also be important to examine
cause of the differential effect of race of occupational worker
these issues among other samples of African American children
among children from different age and economic groups on the
(ideally with higher rates of participation than obtained here) and
basis of these data alone. As Spencer (1985) suggested, the occu-
among children from other racial and ethnic groups. Clearly,
pational goals of older African American children may be influ-
further research is needed to understand more completely the
enced by their consciousness of racial discrimination within soci-
long-term implications of children using racial cues as a factor in
ety, in general, and within occupational settings, in particular.
determining their judgments about and interests in occupations.
Economically disadvantaged youths may be especially ill prepared
to cope with the perceived racial biases of the majority culture
(Spencer & Markstrom-Adams, 1990). It is also possible that the
References
perception of economic constraints among older African American
children leads to decreased interest in what they perceive to be
Aboud, F. E. (1988). Children and prejudice. New York: Blackwell.
high-status, and thus difficult-to-attain, occupations. What is clear
Aboud, F. E., & Levy, S. R. (Eds.). (1999). Reducing racial prejudice,
from these data is that by the age of 6, African American children
discrimination, and stereotyping: Translating research into programs.
Journal of Social Issues, 55(4).
have developed racial schemas that incorporate beliefs about oc-
American Psychological Association. (1995). Task force report on the
cupations and that these schemas affect their perceptions of jobs
changing gender composition of psychology. Washington, DC: Author.
and occupational aspirations in significant ways. Further, these
Retrieved February 27, 2003, from http://www.apa.org/pi/taskforce/
schemas differ in important ways across development and across
homepage.html
children from different socioeconomic backgrounds.
Averhart, C. A., & Bigler, R. S. (1997). Shades of meaning: Skin tone,

---

RACE AND THE WORKFORCE
579
racial attitudes, and constructive memory in African American children.
Hauser, R. M. (1994). Measuring socioeconomic status in studies of child
Journal of Experimental Child Psychology, 67, 363–388.
development. Child Development, 65, 1541–1545.
Bound, J., & Freeman, R. (1992). What went wrong? The 1980’s erosion
King, M. C. (1992). Occupational segregation by race and sex, 1940 –1988.
of the economic well-being of Black men. Quarterly Journal of Eco-
Monthly Labor Review, 115, 30 –37.
nomics, 107, 201–232.
Liben, L. S., Bigler, R. S., & Krogh, H. R. (2001). Pink and blue collar
Children NOW. (1998). A different world: Children’s perceptions of race
jobs: Children’s judgments of job status and job aspirations in relation to
and class in the media. Oakland, CA: Children NOW.
sex of worker. Journal of Experimental Child Psychology, 79, 346 –363.
Cook, T. D., Church, M. B., Ajanaku, S., Shadish, W. R., Kim, R., &
McGee, J., & Stockard, J. (1991). From a child’s view: Children’s occu-
Cohen, R. (1996). The development of occupational aspirations and
pational knowledge and perceptions of occupational characteristics. In S.
expectations among inner-city boys. Child Development, 67, 3368 –
Cahill (Ed.), Sociological studies of child development (Vol. 4, pp.
3385.
113–136). Greenwich, CT: JAI Press.
Cortes, C. E. (2000). The children are watching: How the media teach
McLoyd, V. C., & Randolph, S. (1985). Secular trends in the study of
about diversity. New York: Teachers College Press.
Afro-American children: A review of Child Development, 1936 –1980.
Danzinger, S., & Gottschalk, D. (1991). Uneven tides. New York: Russell
In A. B. Smuts & J. W. Hagen (Eds.), History and research in child
Sage.
development. Monographs of the Society for Research in Child Devel-
Eagly, A. H., & Steffen, V. J. (2000). Gender stereotypes stem from the
opment, 50(4 –5, Serial No. 211), 78 –92.
distribution of women and men into social roles. In C. Stangor (Ed.),
Phinney, J. S., & Rotheram, M. J. (1993a). Children’s ethnic socialization:
Stereotypes and prejudice: Essential readings (pp. 142–160). Philadel-
Themes and implications. In J. S. Phinney & M. J. Rotheram (Eds.),
phia: Psychology Press.
Children’s ethnic socialization: Pluralism and development (pp. 274 –
Farley, R., & Allen, W. R. (1987). The color line and the quality of life in
292). Newbury Park, CA: Sage.
America. New York: Russell Sage Foundation.
Phinney, J. S., & Rotheram (Eds.). (1993b). Children’s ethnic socializa-
Federal Glass Ceiling Commission. (1995). Good for business: Making full
tion: Pluralism and development. Newbury Park, CA: Sage.
use of the nation’s human capital. Washington, DC: U.S. Government
Simmons, D. D. (1962). Children’s rankings of occupational prestige.
Printing Office.
Personnel and Guidance Journal, 41, 332–336.
Garcia Coll, C., Lamberty, G., Jenkins, R., McAdoo, H. P., Crnic, K.,
Spencer, M. B. (1985). Cultural cognition and social cognition as identity
Wasik, B. H., & Vazquez Garcia, H. (1996). An integrative model for
factors in Black children’s personal-social growth. In M. B. Spencer,
the study of developmental competencies in minority children. Child
G. K. Brookins, & W. R. Allen (Eds.), Beginnings: The social and
Development, 67, 1891–1914.
Graves, S. B. (1996). Diversity on television. In T. M. MacBeth (Ed.),
affective development of Black children (pp. 215–230). Hillsdale, NJ:
Tuning in to young viewers: Social science perspectives on television
Erlbaum.
(pp. 61– 86). Newbury Park, CA: Sage.
Spencer, M. B., Dobbs, B., & Swanson, D. P. (1988). Afro-American
Graves, S. B. (1999). Television and prejudice reduction: When does
adolescents: Adaptational processes and socioeconomic diversity in be-
television as a vicarious experience make a difference? Journal of Social
havioral outcomes. Journal of Adolescence, 11, 117–137.
Issues, 55, 707–725.
Spencer, M. B., & Markstrom-Adams, C. (1990). Identity processes among
Greenberg, B. S., & Brand, J. E. (1993). Cultural diversity on Saturday
racial and ethnic minority children in America. Child Development, 61,
morning television. In G. Berry & J. K. Asamen (Eds.), Children and
290 –310.
television: Images in a changing sociocultural world (pp. 132–142).
Swinton, D. (1987). The state of African American America 1987. New
Newbury Park, CA: Sage.
York: National Urban League.
Greenberg, B. S., & Collette, L. (1997). The changing faces on TV: A
Terrell, F., Terrell, S. L., & Miller, F. (1993). Level of cultural mistrust as
demographic analysis of network television’s new seasons, 1966 –1992.
a function of educational and occupational expectations among African
Journal of Broadcasting & Electronic Media, 41, 1–13.
American students. Adolescence, 28, 573–578.
Hale, J. E. (1980). De-mythicizing the education of Black children. In R. L.
West, C. (1993). Race matters. Boston: Beacon Press.
Jones (Ed.), The Black psychology (pp. 221–230). San Francisco: Harper
Williams, J. E., & Morland, J. K. (1976). Race, color, and the young child.
& Row.
Chapel Hill: The University of North Carolina Press.
(Appendix follows)

---

580
BIGLER, AVERHART, AND LIBEN
Appendix
Familiar Occupations Rated by Participants
Occupational status
High
Medium
Low
Airline pilot
Bank teller
Car washer
Business executive
Bus driver
Cashier
College professor
Electrician
Fast food worker
Dentist
Fire fighter
Garbage collector
Doctor
Hair dresser
Grocery bagger
Politician
Mail carrier
Janitor
Scientist
Police officer
Laundry worker
Surgeon
School teacher
Maid
Veterinarian
Secretary
Waiter
Received December 7, 2000
Revision received November 7, 2002
Accepted December 30, 2002

---