PARENTING: SCIENCE AND PRACTICE Copyright © 2007, Lawrence Erlbaum Associates, Inc.
October–December 2007 Volume 7 Number 4 Pages 357–366
HPAR1529-51921532-7922Parenting: Science and Practice, Vol. 7, No. 4, September 2007: pp. 1–15Parenting: Science and Practice
Do Mothers’ Gender-Related Attitudes
or Comments Predict Young Children’s
Predicting Children’s Gender BeliefsFRIEDMAN, LEAPER, BIGLER
Carly Kay Friedman, Campbell Leaper, and Rebecca S. Bigler
Objective. We tested relations among mothers’ gender attitudes, mothers’
comments about gender, and young children’s gender-stereotyped beliefs.
Design. Mothers (from mostly middle-class, European-American back-
grounds) read and discussed a gender-related story to their child (N = 74, M =
64 months). Mothers’ speech was coded as either endorsing or challenging
gender stereotypes. Results. Mothers with gender-egalitarian attitudes used
more counter-stereotypical comments. Mothers used more counter-
stereotypic comments with daughters than sons. Mothers’ gender attitudes
predicted gender stereotyping in younger children (3–5 years) but not older
children (6–7 years). However, mothers’ speech did not predict children’s gen-
der stereotyping. Conclusions. Mothers’ gender-related attitudes and com-
ments may not reliably predict young children’s gender-stereotyped beliefs.
Children endorse gender stereotypes concerning many traits and roles
between 3 and 7 years of age (see Martin, Ruble, & Szkrybalo, 2002). Dur-
ing this age period, mothers may be an especially influential source for
learning these beliefs (see Leaper, 2002). According to social cognitive
theory, gender socialization occurs through modeling, enactive experi-
ence, and direct tuition (Bussey & Bandura, 1999). Although many stud-
ies have examined the influence of modeling and enactive experience,
few studies have investigated direct tuition in relation to children’s gen-
der stereotyping (Friedman, Bigler, & Leaper, 2007; Friedman, Leaper, &
Bigler, 2004; Gelman, Taylor, & Nguyen, 2004). In one relevant study,
Gelman et al. (2004) observed mothers’ speech while reading a gender-
themed story to their preschool children. In general, the mothers
endorsed highly egalitarian attitudes. Consistent with their views, moth-
ers made few comments that explicitly endorsed gender stereotypes
358 FRIEDMAN, LEAPER, BIGLER
(“Only girls can play with dolls.”). In contrast, mothers commonly used
language that implicitly stereotyped gender through references to generic
gender categories (“Girls play with dolls.”) and gender labeling (“That’s a
boy racing the car.”). Furthermore, these highly egalitarian mothers
“rarely negated children’s gender stereotypes, and in fact primarily
affirmed what children said” (Gelman et al., 2004, p. 108). Thus, gender
attitudes and gender-related comments often may be contradictory—
which may help account for the weak average association between moth-
ers’ and children’s gender beliefs. Therefore, it may be helpful to consider
interrelations among mothers’ attitudes, mothers’ speech, and children’s
stereotypes at the same time.1
In the present investigation, we classified mothers’ gender-related
references to traits, behaviors, or roles as either endorsing (i.e., stereo-
typical) or challenging (i.e., counter-stereotypical) traditional gender
norms. Additionally, we distinguished between gender-related com-
ments made about females, males, or both females and males. Mothers
may be more likely to refer to characters that are relevant to the child’s
own gender (Gelman et al., 2004). Also, parents tend to be more rigid in
their stereotypes about males than females (Leaper, 2002). Lastly, we
coded for descriptive and prescriptive comments. Whereas descriptive
comments simply call attention to an example of a stereotypic or
counter-stereotypic trait or behavior, prescriptive comments suggest
approval or disapproval.
Our major research questions were as follows: First, are mothers’ gender
beliefs correlated with their young children’s gender stereotyping? Second,
are mothers’ gender-related comments correlated with their children’s gen-
der stereotyping? Third, do mothers’ gender beliefs predict their own com-
ments about gender? Finally, do mothers’ comments mediate the
association between mothers’ and children’s gender-stereotyped beliefs?
(Testing the last question presupposed that mothers’ beliefs, mothers’ com-
ments, and children’s stereotyping are significantly intercorrelated.)
In addition to these research questions, we considered three potential
moderators. First, gender-typing pressures tend to be stronger for boys
than girls (Leaper & Friedman, 2007); therefore, mothers may make more
stereotypical comments with sons than daughters. Second, prior research
1The present research was carried out prior to the publication of Gelman et al.’s (2004)
study. It built upon Carly Friedman’s 2002 undergraduate honors thesis in psychology at
the University of Texas at Austin (Friedman, Bigler, & Leaper, 2007). Although our current
study is similar to Gelman et al.’s (2004) research in many ways, the two studies employ dif-
ferent parent attitude measures, verbal coding methods, counter-balancing procedures, and
types of analyses.
PREDICTING CHILDREN’S GENDER BELIEFS 359
indicates a larger association between mother-child gender attitudes with
older (6-year-old and above) than younger (3- to 5-year-old) children
(Tenenbaum & Leaper, 2002); and therefore, mothers’ comments may
have more influence on older children. Finally, by counter-balancing the
order of administering the story and the attitude surveys, it is possible to
assess the effect of each on children’s beliefs as well as mothers’ beliefs
Seventy-four children, from 3 to 7 years of age (M = 64.09 months,
SD = 18.29), participated with their mothers. The mothers were predomi-
nantly White European American (91.5%) and middle to upper-middle
class; and all resided in a California coastal city.
The stated purpose of the study was to examine parental influences on
children’s story understanding. The research occurred at a university
research laboratory in two parts: (1) Mothers and children completed sep-
arate surveys measuring their gender attitudes and (2) mothers read a
story to their child. (The order of survey and story-reading was counter-
balanced across participants.) Mothers were asked to read the story as
they normally would with their child, and to discuss the book for 5 min
after reading the story. The reading and the discussion were videotaped
for later transcription and coding. At the end, mothers were debriefed.
Mothers read and discussed with their children an illustrated book
entitled, Fun at the Farm (Leaper & Friedman, 2002). The story was written
to include 9 gender-typed and 9 cross-gender-typed representations of
traits or behaviors in girls and boys. It focused on two mixed-gender sib-
ling pairs who are cousins with one another. One pair (Jenny and David)
was gender-typed, whereas the other pair (Catherine and Michael) was
cross-gender-typed. For example, in the first part of the story, each sibling
pair is portrayed in their homes playing and doing chores that are either
gender-typed (e.g., boy playing with blocks) or cross-gender-typed (e.g.,
girl mowing the lawn).
360 FRIEDMAN, LEAPER, BIGLER
Children completed the activity subscale of the Preschool Occupation,
Activity, and Trait Attitude Measure scale (POAT-AM; based on Liben &
Bigler, 2002). Respondents were asked if they think “only boys,” “only
girls,” or “both boys and girls” should do 18 different activities (12 gender-
stereotypical and 6 neutral). Internal reliability for the measure was high
(α = .81).
Mothers completed the Pacific Attitudes Towards Gender Scale
(PATG; Vaillancourt & Leaper, 1997), which included 21 statements rated
on a 7-point scale assessing views regarding dating, family, and occupa-
tional roles; personality traits and activities; and overall gender equality.
Internal reliability on the PATG was high (α = .81).
Mothers’ Gender-Related Comments
Mothers’ gender-related comments about the story were transcribed
from the videotapes and coded. Gender-related comments were defined
as statements or questions referring to any traits, behaviors, or roles attrib-
uted to girls or boys. These gender-related comments were subsequently
coded along three dimensions: First, target refers to whether the comments
pertained to females, males, or generally both genders. Second, valence
included three categories: Stereotypic comments acknowledged or
supported stereotypic roles (“Catherine was brushing her doll.”); counter-
stereotypic comments negated a stereotype or emphasized counter-stereo-
typic roles (“Basketball is boys’ stuff and girls’ stuff.”); and neutral
gender-related comments neither supported nor negated stereotypes
(“Which chore would you rather do?”). Finally, mode refers to whether
statements were prescriptive or descriptive. Prescriptive comments explic-
itly endorsed a stereotypic or counter-stereotypic gender role (“I like that
the girl is doing the dishes.”), whereas descriptive comments only called
attention to a stereotypic or counter-stereotypic gender role (“Those girls
are playing ball.”). The inter-coder reliability was high across three
researchers based on a comparison of 14 transcripts (minimum κ = .70).
RESULTS AND DISCUSSION
Effects of Child Gender, Child Age, and Survey Order on Outcome
Children’s gender stereotyping. In the first set of analyses, we carried out
2 (child gender) × 2 (child age level: 3- to 5-year-olds versus 6- to 7-year-olds)
PREDICTING CHILDREN’S GENDER BELIEFS 361
× 2 (survey order) ANOVAs to explore if any of these factors were related
to children’s gender stereotyping on the POAT. Children scored lower on
gender stereotyping if they were surveyed after the story rather than
before the story, F(1, 66) = 8.31, η2 = .11, p <.01. However, the order effect
was qualified by two interactions. First, there was a significant Survey
Order X Child Age Level interaction, F(1, 66) = 6.16, η2 = .09, p = .02.
Among older children, those who were surveyed after reading the story
indicated less stereotyping than those surveyed before the story, F(1, 32)
= 15.04, η2 = .32, p <.01. Among younger children, there was no difference
in POAT scores between those surveyed after the story and those sur-
veyed before the story, F(1, 34) < 1, ns. In addition, there was a marginal
Survey Order × Child Gender interaction, F(1, 66) = 3.03, η2 = .04, p = .09.
Boys demonstrated significantly lower stereotyping if surveyed after the
story (M = .14, SD = .20) than before the story (M = .38, SD = .28), F(1, 30)
= 9.06, η2 = .23, p <.01. In contrast, there was no difference in POAT scores
between girls who were surveyed after (M = .24, SD = .20) versus those
who were surveyed before the story (M = .29, SD = .22), F(1, 36) < 1, ns.
Earlier intervention studies have demonstrated that presenting
counter-stereotypical stories can reduce children’s gender stereotypes
(e.g., Flerx, Fidler, & Rogers, 1976). In our study, the story had an imme-
diate effect only for older children and for boys. First, older children may
have been influenced more due to their greater cognitive flexibility (see
Martin et al., 2002). Boys may have been more responsive to this influence
because — as some studies suggest — boys tend to be more concerned
with gender stereotypes than are girls (see Leaper & Friedman, 2007).
Although some readers may find it encouraging that a story could reduce
children’s gender stereotyping, this type of effect is usually weak and
transitory (see Bigler, 1999).
Mothers’ gender attitudes. The same 3-way ANOVA design previously
described was used with mothers’ mean PATG score. Results indicated a
significant Child gender × Age level interaction, F(1, 65) = 6.00, η2 = .08,
p = .02. Simple effects tests revealed that mothers of older daughters had
significantly more egalitarian attitudes than did either mothers of older
sons or mothers of younger daughters; but mothers of younger sons did
not significantly differ from the other three groups.
Mothers’ gender-related comments. We performed a mixed-design ANOVA
with survey order, child gender, and child age level as between-group
factors and mode of remark (prescriptive versus descriptive), valence of
remark (stereotypic versus counter-stereotypic), and target (females ver-
sus males versus general) as repeated measures. Overall, descriptive
362 FRIEDMAN, LEAPER, BIGLER
comments (M = 23.77, SD = 13.17) were more common than prescriptive
comments (M = 6.41, SD = 5.79), F(1, 66) = 154.13, η2 = .70, p < .01. In addition,
counter-stereotypical comments (M = 18.26, SD = 10.27) were more com-
mon than stereotypical comments (M = 11.92, SD = 8.17), F(1, 66) =
37.69, η2 = .36, p < .01.
A target main effect also occurred, F(2, 66) = 12.86, η2 = .16, p < .001, but
it was qualified in a Target × Gender interaction, F(2, 66) = 13.89, η2 = .17, p <
.001. Pairwise comparison tests indicated that mothers were more likely
to refer to the child’s own gender. This meant mothers with daughters
made more comments about females than males; and mothers with sons
made more comments about males than females (all ps <.05). But the tar-
get effect was further qualified by a 4-way Speech Mode × Valence × Tar-
get × Child Gender interaction, F(2, 66) = 3.55, η2 = .05, p <.05. Follow-up
pairwise comparisons indicated the following pattern: Among mothers of
daughters, there was a target gender difference for descriptive counter-
stereotypical comments — which were more commonly about females
than males (p <.001). In contrast, among mothers of sons, there was a tar-
get gender difference for descriptive stereotypical comments; they were
more frequently about males than females (p <.05). Thus, when making
descriptive comments about gender roles, mothers tended to highlight
counter-stereotypical female roles with daughters and stereotypical male
roles with sons.
Many of our findings are consistent with those reported by Gelman
et al. (2004). In both studies, mothers made few explicit statements
about gender roles, but they commonly did make generic statements
about gender. Both studies also found that mothers made more refer-
ences to counter-stereotypical story content. Given that counter-
stereotypical images are unusual in most children’s books, perhaps
mothers were especially likely to take notice of them. Like Gelman
et al. (2004), we also found that mothers referenced characters that
matched their child’s own gender. Consistent with prior reports that
parents tend to allow more gender-role flexibility in daughters than
sons (see Leaper, 2002), we also found mothers used more counter-ste-
reotypical speech with daughters and used more stereotypical speech
Relations Among Variables
One of the unique contributions of our study was to test how well chil-
dren’s gender beliefs, mothers’ gender beliefs, and mothers’ gender-
stereotypical or counter-stereotypical speech were interrelated.
PREDICTING CHILDREN’S GENDER BELIEFS 363
Relations between children’s and mothers’ gender beliefs. Children’s gen-
der stereotypes and mothers’ gender attitudes were not significantly
correlated (r = .17). We carried out additional analyses to test if the corre-
lations differed by child gender or child age. Age level — but not gender —
moderated the associations: There was a significant correlation between
mothers’ and children’s gender beliefs among younger children, r (37) =
.38, p < .05, but not among older children, r (36) = −.03, ns. Although this
pattern is contrary to the age-related differences indicated in Tenenbaum
and Leaper’s (2002) meta-analysis, it is consistent with the idea that extra-
familial influences — such as peers and the media — may increasingly
shape children’s gender beliefs as they get older (see Leaper & Friedman,
Relations between children’s gender stereotypes and mothers’ speech. There
were no significant correlations between children’s gender stereotypes
and either mothers’ total stereotypical comments (r = .02) or total
counter-stereotypical comments (r = .03). When more specific speech
codes were analyzed, this also included mothers’ prescriptive stereo-
typical comments (r = .07), descriptive stereotypical comments (r = .01),
prescriptive counter-stereotypical comments (r = .02), and descriptive
counter-stereotypical comments (r = .04). Additional tests indicated no
association when either child gender or age level was taken into
account. Furthermore, testing a quadratic function did not affect the
Relations between mothers’ gender attitudes and mothers’ speech. Mothers’
gender attitudes were unrelated to either their total stereotypical (r =
−.05) or total counter-stereotypical (r = .03) comments. Next, we distin-
guished between comments that were either prescriptive or descriptive.
Mothers’ traditional gender attitudes were negatively correlated with
their prescriptive counter-stereotypical comments (r = −.24, p <.05), and
positively correlated with their descriptive counter-stereotypical com-
ments at a marginal level (r = .21, p <.08). But mothers’ attitudes were
unrelated to their prescriptive stereotypical comments (r = −.14) or
descriptive stereotypical comments (r = −.02).
We found a small correlation between younger (but not older) chil-
dren’s gender stereotypes and mothers’ gender attitudes. We also
observed a small correlation between mothers’ gender attitudes and
364 FRIEDMAN, LEAPER, BIGLER
counter-stereotypic speech. However, mothers’ comments did not predict
children’s gender stereotyping. Therefore, the idea that mothers’ speech
might mediate the association between mothers’ attitudes and children’s
stereotyping was not supported. Perhaps this relationship could be identi-
fied in future studies through the use of a different methodology.
When a significant association between mothers’ gender attitudes and
behavior was found, it was specifically with regard to gender-egalitarian
mothers’ greater use of prescriptive counter-stereotypical speech. Elicit-
ing counter-stereotypical comments may have depended largely on our
presentation of counter-stereotypical and stereotypical story content. In
most children’s everyday lives, however, this would be relatively uncom-
mon. It is especially rare for children to see boys demonstrating feminine-
stereotyped traits or roles. In contrast, gender stereotyped models prolif-
erate in the media, children’s peer groups, and other sources (Leaper &
Friedman, 2007). Mothers’ influence on children’s gender stereotyping
may simply pale compared to the impact of peers and the media. In addi-
tion, mothers generally do not challenge gender stereotyping in these
contexts. Lastly, it appears that gender labeling alone — even in neutral
statements — is a cause of gender stereotyping among children (Bigler &
Liben, 2007). Possibly for these and other reasons, there is a low-to-negli-
gible average correspondence between mothers’ and young children’s
gender beliefs (Tenenbaum & Leaper, 2002). At the same time, if parents
provide clear and consistent messages, perhaps they could directly influ-
ence young children’s gender stereotyping (see Bigler & Liben, 2007). In
addition, factors such as childrearing style (e.g., Ex & Janssens, 1998) and
children’s identification with their parents (e.g., Sinclair, Dunn, & Lowery,
2005) may moderate how strongly children adopt their parents’ attitudes.
AFFILIATIONS AND ADDRESS
Carly Kay Friedman, Department of Psychology, University of California
Santa Cruz, 1156 High Street, Santa Cruz, CA 95064. E-mail: cam@
ucsc.edu. Campbell Leaper is at University of California Santa Cruz.
Rebecca S. Bigler is at University of Texas at Austin.
The research was supported by a National Science Foundation fellowship
to the first author at the University of California Santa Cruz; and by
grants from the University of California Santa Cruz Social Sciences
PREDICTING CHILDREN’S GENDER BELIEFS 365
Division and Academic Senate to the second author. The authors thank
the families who participated in the study for their time and cooperation.
Preliminary findings were presented at the Gender Development
Research Conference, San Francisco, April 2004.
Bigler, R. S. (1999). Psychological interventions designed to counter sexism in children:
Empirical limitations and theoretical foundations. In W. B. J. Swann, & J. H. Langlois
(Eds.), Sexism and Stereotypes in Modern Society: The Gender Science of Janet Taylor Spence.
(pp. 129–151). Washington, DC: American Psychological Association.
Bigler, R. S., & Liben, L. S. (2007). Developmental intergroup theory: Explaining and reduc-
ing children’s social stereotyping and prejudice. Current Directions in Psychological Science,
Brooks-Gunn, J. (1986). The relationship of maternal beliefs about sex typing to maternal
and young children’s behavior. Sex Roles, 14, 21–35.
Bussey, K., & Bandura, A. (1999). Social cognitive theory of gender development and differ-
entiation. Psychological Review, 106, 676–713.
Ex, C. T. G. M., & Janssens, J. M. A. M. (1998). Maternal influences on daughters’ gender role
attitudes. Sex Roles, 38, 171–186.
Flerx, V. C., Fidler, D. S., & Rogers, R. W. (1976). Sex role stereotypes: Developmental
aspects and early intervention. Child Development, 47, 998–1007.
Friedman, C. K., Bigler, R. S., & Leaper, C. (2007). Mothers’ Guidance of Young Children’s Gen-
der Attitudes During Shared Storybook Reading. Manuscript in preparation.
Friedman, C. K., Leaper, C., & Bigler, R. S. (2004, April). The Influence of Parental Discussion
on Children’s Gender Stereotyping. Presentation at the Gender Development Research
Conference, San Francisco.
Gelman, S., Taylor, M., & Nguyen, S.P. (2004). Mother-child conversations about gender.
Monographs of the Society for Research in Children Development, 69(1).
Goodnow, J. (1997). Parenting and the transmission and internalization of values: From
social-cultural perspectives to within-family analyses. In J.E. Grusec & L. Kuczynski
(Eds.), Parenting and Children’s Internalization of Values: A Handbook of Contemporary Theory
(pp. 333–361). Hoboken, NJ: John Wiley & Sons, Inc.
Leaper, C. (2002). Parenting girls and boys. In M. H. Bornstein, Handbook of Parenting:
Children and Parenting (2nd ed., Volume 1, pp. 189–225). Mahwah, NJ, US: Erlbaum.
Leaper, C., & Bigler, R. S. (2004). Gendered language and sexist thought. Monographs of the
Society for Research in Child Development, 69(1), 128–142.
Leaper, C., & Friedman, C. K. (2002). Fun at the Farm. Unpublished document available from
Leaper, C., & Friedman, C.K. (2007). The socialization of gender. In J. Grusec & P. Hastings
(Eds.), The Handbook of Socialization: Theory and Research (pp. 561–587). New York:
Liben, L. S., & Bigler, R. S. (2002). The developmental course of gender differentiation: Con-
ceptualizing, measuring, and evaluating constructs and pathways. Monographs of the
Society for Research in Child Development, 67(2).
Martin, C. L., Ruble, D. N., & Szkrybalo, J. (2002). Cognitive theories of early gender devel-
opment. Psychological Bulletin, 128, 903–933.
366 FRIEDMAN, LEAPER, BIGLER
Sinclair, S., Dunn, E., & Lowery, B. S. (2005). The relationship between parental racial attitudes
and children’s implicit prejudice. Journal of Experimental Social Psychology, 41, 283–289.
Tenenbaum, H. R., & Leaper, C. (2002). Are parents’ gender schemas related to their
children’s gender-related cognitions?: A meta analysis. Developmental Psychology, 38,
Vaillancourt, T., & Leaper, C. (1997). Pacific Attitudes Toward Gender (PATG) Scale. Retrieved
Weitzman, N., Birns, B., & Friend, R. (1985). Traditional and nontraditional mothers’ com-
munication with their daughters and sons. Child Development, 56, 894–898.