Disscussion A&B

Pink Frilly Dresses and the Avoidance of All Things “Girly”: Children’s Appearance Rigidity and Cognitive Theories of Gender Development

May Ling Halim California State University, Long Beach

Diane N. Ruble and Catherine S. Tamis-LeMonda New York University

Kristina M. Zosuls Arizona State University

Leah E. Lurye and Faith K. Greulich New York University

Many young children pass through a stage of gender appearance rigidity; girls insist on wearing dresses, often pink and frilly, whereas boys refuse to wear anything with a hint of femininity. In 2 studies, we investigated the prevalence of this apparent hallmark of early gender development and its relation to children’s growing identification with a gender category. Study 1a examined the prevalence of this behavior and whether it would exhibit a developmental pattern of rigidity followed by flexibility, consistent with past research on identity-related cognitions. Interviews with 76 White, middle-class parents and their 3- to 6-year-old children revealed that about two thirds of parents of 3- and 4-year-old girls and almost half (44%) of parents of 5- and 6-year-old boys reported that their children had exhibited a period of rigidity in their gender-related appearance behavior. Appearance rigidity was not related to parents’ preferences for their children’s gender-typed clothing. Study 1b examined whether cognitive theories of identity development could shed light on gender appearance rigidity. The more important and positive children considered their gender and the more children understood that gender categories remain stable over time (gender stability), the more likely children were to wear gender-typed outfits. In Study 2, we extended this research to a more diverse population and found that gender appearance rigidity was also prevalent in 267 4-year-old children in the United States from African American, Chinese, Dominican, and Mexican immigrant low-income backgrounds. Results suggest that rigid gender-related appearance behavior can be seen among young children from different backgrounds and might reflect early developing cognitions about gender identity.

Keywords: gender identity, appearance, centrality, gender constancy, gender cognitions

Supplemental materials: http://dx.doi.org/10.1037/a0034906.supp

Our appearances are a symbolic representation of our self- concepts and convey messages to others about how we would like to be perceived. Clothing is a critical way we communicate our identity to others (e.g., Feinberg, Mataro, & Burroughs, 1992) and can signal membership in or separation from social groups (e.g., Freitas, Kaiser, & Hammidi, 1996). It is surprising, then, that the topic of children’s gender-typed appearances has been virtually

ignored in the gender identity development literature. The study of gender development has been dominated by a focus on gender stereotyping or activity preferences and behaviors (Zosuls, Miller, Ruble, Martin, & Fabes, 2011), despite numerous calls to focus on multiple domains (Huston, 1983; Ruble, Martin, & Berenbaum, 2006). Children’s gender-typed appearance is also unique from other aspects of gender-typing. During the course of a day, playing

This article was published Online First November 25, 2013. May Ling Halim, Department of Psychology, California State Univer-

sity, Long Beach; Diane N. Ruble, Department of Psychology, New York University; Catherine S. Tamis-LeMonda, Department of Applied Psychol- ogy, New York University; Kristina M. Zosuls, Department of Family and Human Development, Arizona State University; Leah E. Lurye and Faith K. Greulich, Department of Psychology, New York University.

This study was conducted at the Social Development Lab within the Department of Psychology in the New York University Graduate School of Arts and Sciences (GSAS) as well as at the Center for Research on Culture, Development, and Education (CRCDE), within the Department of Applied Psychology at New York University’s Steinhardt School of Culture, Edu- cation, and Human Development. This research was supported by the National Science Foundation Grants BCS 021859 and IRADS 0721383 to

Catherine S. Tamis-LeMonda. Preparation of this article was also sup- ported in part by National Institute of Child Health and Human Develop- ment Research Grant R01 HD04994 and an American Recovery and Reinvestment Act supplement to Diane N. Ruble.

We thank our colleagues and staff at the CRCDE and at GSAS, partic- ularly Celena Chong, Sandra Dias, Shuan-Ju Hung, Cristina Hunter, Cyndy Karras, Yana Kuchirko, Jennifer Ledesma, Catharine Lennon, Alana Paz, Julia Raufman, Jessica Senie, Emerald Shee, Irene Sze, Sheena Yap, Abigail Walsh, and Irene Wu, as well as the mothers and children who participated in our research.

Correspondence concerning this article should be addressed to May Ling Halim, Department of Psychology, California State University, Long Beach, 1250 Bellflower Boulevard, Long Beach, CA 90840-0901. E-mail: [email protected]

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Developmental Psychology © 2013 American Psychological Association 2014, Vol. 50, No. 4, 1091–1101 0012-1649/14/$12.00 DOI: 10.1037/a0034906

1091

with toys or with same- or other-gender peers is a transitory behavior, yet appearances are comparatively stable. Clothing thus allows a child to announce to the world, “This is who I am” as a girl or boy.

Appearance rigidity, adherence to conforming to gender norms in one’s appearance through gender-stereotypical dress, is one particular phenomenon that might elucidate this appearance– identity link in young children. Some young girls, it seems, ac- cording to parent anecdotes and informal observations, go through a phase in which they refuse to wear anything but pink, frilly dresses (Ruble, Lurye, & Zosuls, 2007). Parents have reported that this “rigidity” can be seen in the level of gender stereotypicality (e.g., wearing pink from head to toe) and in the frequency of its occurrence (e.g., insisting on wearing a dress every single day, rain or shine).

We might speculate that this kind of behavior is linked to socialization processes, especially pressures from advertising. Girls might copy Disney princesses (England, Descartes, & Collier-Meek, 2011), and others might reinforce their behavior with praise. However, such an interpretation seems incomplete. For example, in line with theories about the benefits of androgyny (Bem, 1981), the insistence on wearing ultra-feminine clothing might upset some parents, causing them to feel that they have failed to raise their daughters in a gender-neutral fashion, as the popular media has noted (Fine, 2010; Orenstein, 2010; Padawar, 2012). In addition, about a third of young children exhibit other forms of rigidity or “extremely intense interests” beyond those that involve gender identity (DeLoache, Simcock, & Macari, 2007). Moreover parents reported that these preoccupations originated from the child rather than being encouraged by others. These counterefforts by some parents and children’s intense interests across domains suggest that appearance rigidity might emanate, at least in part, from factors other than external socialization, such as cognitive–developmental and motivational processes, referred to here as self-socialization.

Cognitive Theories of Gender Development

Cognitive theories of gender development emphasize three key features (Martin, Ruble, & Szkrybalo, 2002). First, children are viewed as active, internally motivated agents who construct the meaning of gender categories (Tobin et al., 2010). Once children understand that they belong to a gender category, they embark on an investigation as “gender detectives,” attending to information about their own gender and about differences between girls and boys (Martin & Ruble, 2004). Second, children’s emerging under- standing of gender concepts motivates them to master gender categories by behaving in gender-appropriate ways (Stangor & Ruble, 1987). Third, there exists a developmental trajectory of gender typing. Once children recognize an important categorical distinction such as gender, they might exhibit phase-like shifts in the rigidity of category-relevant beliefs and behaviors, moving from a beginning awareness, to rigidity, to flexibility (Ruble, 1994), a trajectory supported by research on children’s endorse- ment of gender stereotypes (Miller, Trautner, & Ruble, 2006; Trautner et al., 2005).

Children’s gender appearance is an ideal behavior to test cog- nitive theories of gender development, as it is a clear marker of gender typing that is relatively permanent across situations. More-

over, appearances are both highly salient and important to young children as their person perception relies on peripheral character- istics such as physical appearance (Ruble & Dweck, 1995). We propose that increases in children’s adherence to wearing feminine or masculine clothing might be viewed as a manifestation of their passage through the rigidity phase in the trajectory of gender development (Halim, Ruble, & Amodio, 2011), reflecting cogni- tions about basic gender identity as a girl or boy, along with early-acquired and highly visible gender-stereotype knowledge (Miller, Lurye, Zosuls, & Ruble, 2009).

Goals and Predictions for the Present Studies

Appearance rigidity among girls has been described previously (i.e., the “pink, frilly dress” phenomenon, or PFD; Ruble, Lurye, et al., 2007), but there has been no direct empirical inquiry to date regarding the phenomenon. In our present research, we aimed to (a) describe children’s appearance rigidity, (b) examine whether gender identity predicts children’s gender appearance rigidity, and (c) explore whether appearance rigidity generalizes to diverse populations.

Appearance Rigidity

We examined the alternative hypotheses that appearance rigidity might be seen in only a few, select young girls or that appearance rigidity might be relatively common in young girls. We were also interested to see whether boys would exhibit their own version of appearance rigidity. Because appearance rigidity is an appearance- based phenomenon, we expected a lower prevalence among boys, given past research suggesting that stereotypes of girls revolve around appearances and stereotypes of boys might revolve more around activities (Miller et al., 2009; Cristofaro & Tamis- LeMonda, 2008). Nevertheless, if appearance rigidity is found in boys, we expected it to include the embracing of superhero outfits (Neppl & Murray, 1997; Paley, 1986) or the donning of other masculine items such as suits and ties. We further anticipated that boys might exhibit appearance rigidity by avoiding feminine cloth- ing as gender-role behavior is sometimes more often defined and exhibited as something boys should not do, rather than what they should do (Chiu et al., 2006; Hartley, 1959; Pickering & Repa- choli, 2001).

We also hypothesized that the prevalence of gender-related appearance would show a trajectory of rigidity followed by flex- ibility across age as predicted by cognitive theories of gender development. Namely, because gender labeling and identity are evident in many children by age 2 (Zosuls et al., 2009) and gender stability usually emerges by age 3 or 4 (Ruble et al., 2006), we expected 3- and 4-year-old children to show more appearance rigidity than 5- and 6-year-old children. In addition, we anticipated that boys would show appearance rigidity later than girls because girls’ gender development sometimes precedes boys’ (Ruble et al., 2006; Zosuls et al., 2009).

Gender Identity and Gender-Typed Appearance

Our second aim was to examine the link between appearance rigidity and two aspects of gender identity development that cap- ture the motivational (gender centrality and evaluation) and knowl-

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1092 HALIM ET AL.

edge (gender constancy) components of cognitive theories of gen- der development. Gender centrality refers to the importance of gender to the self-concept and evaluation refers to personal regard for one’s own gender (see Egan & Perry, 2001).

There has been little research directly examining connections between gender centrality/evaluation and gender-typed behaviors in young children. We hypothesized that higher levels of identifi- cation with one’s own gender would be associated with higher levels of gender-typed appearances. Children who feel that their gender identities are important and positive might desire for others to recognize their gender identities and go to great lengths to convey the right message with their clothing.

Full gender constancy involves learning three increasingly so- phisticated gender category concepts by approximately the age of 6 or 7 years: (a) identity—that they and others are either boys or girls; (b) stability—that one’s sex remains stable over time (e.g., knowing that a baby girl will become a woman); and (c) consis- tency—that despite superficial changes (e.g., if a boy wears a dress), sex does not change (Slaby & Frey, 1975). Researchers examining the connection between gender constancy and gender- typed behaviors have reached mixed conclusions (Halim & Ruble, 2010). Some find positive associations, whereas others find no associations (see Martin et al., 2002). Recent analyses parsing the stages of constancy have suggested that stability might be a better predictor of gender rigidity compared with gender consistency (Smetana & Letourneau, 1984), which has been related to de- creased rather than increased gender rigidity (Ruble, Taylor, et al., 2007; Zucker et al., 1999). Thus, we hypothesized that gender stability would predict gender-typed appearance. We reasoned that if children have attained gender stability, they might feel more committed to their gender and thus dress in more gender- stereotypical ways.

Appearance Rigidity in Multiple Cultures

A third goal was to explore whether appearance rigidity would be seen in children from diverse ethnic backgrounds. There might be large variation in the gender attitudes, gender roles, and gender stereotyping of different cultural communities due to historical and philosophical influences (Kane, 2000). For example, machismo and marianismo, or, broadly, male dominance and female submis- siveness, might characterize Latino culture (Julian, McKenry, & McKelvey, 1994). Chinese Confucian teachings emphasize male dominance in a patriarchal clan system (Hofstede, 1980), while dictating that the female role is to serve (Tu, 1985). In contrast, some have argued that workforce participation by African Amer- ican women and the value of equality in American culture more broadly have created greater gender equality in African American families (Gutman, 1976; Tamis-LeMonda & McFadden, 2009). In light of the gendered contexts of different communities, examining gender appearance rigidity across ethnically diverse children from different social classes provides a stringent test of the prevalence and generality of the phenomenon.

Overview

We conducted two studies. Study 1a investigated whether gen- der appearance rigidity is apparent in early childhood in a middle- class, mostly White sample. We also investigated whether parents’

preferences are associated with children’s gender appearance ri- gidity. In Study 1b, we tested cognitive theories of gender devel- opment by examining whether appearance rigidity was connected to children’s gender identification and understanding of gender stability. In Study 2, we tested whether appearance rigidity was generalizable to populations of children from different ethnicities and from a different socioeconomic class.

Study 1a

Method. Participants and procedure. Parents of 76 children (39 girls

and 37 boys) were recruited from public and private schools in New York City and from university faculty and staff as part of a larger study on gender development. Children’s ages ranged from 3.13 to 6.98 years (M � 4.92, SD � 1.03; 16 three-year-olds, 22 four-year-olds, 26 five-year-olds, and 12 six-year-olds). The younger cohort consisted of 38 children (19 girls and 19 boys; M age � 4.08 years, SD � 0.56), as did the older cohort (20 girls and 18 boys; M age � 5.78 years, SD � 0.59). Families came from middle- to upper middle-class backgrounds (average income $120,000–$140,000; average education of parents: some graduate school) and were mostly White (about 90%), with a small number of Asian, Latino, and African American families. Female research- ers interviewed the primary caretaker of each child (74 mothers and two fathers) at the university and by phone.

Measures. Lifetime appearance rigidity. Because we assumed that chil-

dren’s extreme insistence on wearing gender-typed clothing would be a relatively short-lived phase, we asked about “lifetime” ap- pearance rigidity to assess whether appearance rigidity was ever exhibited (in the past or present). Interviewers asked parents, “Has your daughter (son) ever insisted on wearing traditional feminine (masculine) clothes whenever she (he) went out? Please describe what that was.” Three independent judges coded responses (� � .88) and discussed all disagreements until they reached agreement. Judges first rated responses on a 4-point scale: 1 � no interest in wearing traditional gender-typed clothes (e.g., “No”; “He does not care what he wears”); 2 � little interest/ambiguous (e.g., “No, [she] didn’t insist. [She] just likes to wear nice clothing—dresses or pants”; “No, he is greatly influenced by what boys in class wear”); 3 � pretty interested (e.g., “She likes to wear dresses”; “He wears boys’ clothing; color is important to him”); 4 � insistent (e.g., “Yes, pretty dress—pink and frilly,” “Yes, he wouldn’t be caught dead in girls’ clothing”; M � 2.51, SD � 1.36). Because we were interested in the most extreme cases and wanted to be conservative in our estimates of prevalence, we defined children who exhibited appearance rigidity as those coded as 4’s.

The ways in which children expressed appearance rigidity were also categorized. Categories included (a) dresses and skirts, (b) gender-typed colors (pink/purple versus blue; Chiu et al., 2006), (c) texture/feel/movement of a fabric or a piece of clothing (e.g., “frilly,” “flowing [and] silky,” “stuff that twirls,” “velvet”), (d) decorations/patterns (e.g., flowers, hearts), (e) fancy or formal dress (e.g., ties for boys), (e) avoidance of other-gender-typed clothing or colors (e.g., pants for girls, pink/purple colors for boys), (f) superhero costumes and graphics, and (g) other (e.g., ballet slippers, briefcases). A single child’s expression could fall

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1093APPEARANCE RIGIDITY

into more than one category (e.g., pink dresses would be catego- rized under gender-typed color and dresses/skirts).

Current gender-typed appearance. To capture current gender- typed appearance, interviewers asked parents, “What is a typical outfit for your child?” The same three independent judges coded all of the responses (� � .81) for the degree of femininity and the degree of masculinity of the typical outfit for both boys and girls on a scale from 1 to 5 (see online supplemental materials), r(70) � �.57, p � .001. We reverse-coded how other-gender- typed the outfit was and then averaged the two items to create a scale. Current gender-typed appearance was correlated with life- time appearance rigidity, r(70) � .25, p � .038. Six children were missing data due to an interview error.

Parent preferences for child’s clothing. To measure parents’ preferences toward their children’s clothing, we asked parents of girls (boys), “How comfortable do you feel when your child dresses up in “girlish” (“boyish”) clothes?” (1 � very uncomfort- able, 7 � very comfortable). We also asked, “What are your preferences regarding your child’s clothing?” (for girls [boys]: 1 � very masculine [very feminine], 5 � very feminine [very mascu- line]). Four parents did not complete the interview due to time constraints and had missing data. Parents of boys (M � 3.71, SD � 0.83) and girls (M � 3.47, SD � 0.74) did not significantly differ in their preference for gender-typed clothing, t(69) � 1.31, ns. However, parents of boys (M � 6.88, SD � 0.41) compared with parents of girls (M � 6.46, SD � 0.84) reported being more comfortable if their child wore gender-typed clothing, t(53.3) � 2.74, p � .008, d � 0.75. These items were correlated, r(70) � .34, p � .004, and thus were standardized around their means and averaged together.

Results. Plan of analyses. We first describe the distributions of chil-

dren who have ever exhibited (“lifetime”) appearance rigidity, then provide details about how many children displayed current appear- ance rigidity. Next, to explore trajectories of gender-typed appear- ance, we examine gender by cohort interactions for lifetime and current gender-typed appearance using chi-squares and analyses of variance (ANOVAs). Finally, we investigate whether parent pref- erences concerning clothing were associated with children’s cur- rent gender-typed appearance using multiple regressions.

Prevalence and extremity. Analyses of the lifetime appear- ance rigidity measure indicated that 54% of the girls (68% in the younger cohort and 40% in the older cohort) were reported as having ever exhibited appearance rigidity (this finding also means that 46% either had never exhibited appearance rigidity or had shown a little bit or some interest; see Table 1). In contrast, 27% of the boys (11% in the younger cohort and 44% in the older cohort) were reported to have been insistent on wearing gender- typed clothing at some point in their early childhood. This result also means that 73% of boys either had never exhibited appearance rigidity or had shown a little bit or some interest.

Parents’ comments indicated that this lifetime appearance rigid- ity was quite extreme for some of the girls. About a quarter of the parents of girls who had exhibited appearance rigidity (24%) used expressions such as “has to,” “no option,” “won’t wear anything else,” “not a choice,” or “that’s it.” Parents also pointed to chil- dren’s refusals or rejections of certain items of clothing (80% of boys and 19% of girls who exhibited appearance rigidity).

Ways in which children express lifetime appearance rigidity. For girls, parent responses clustered around the adherence to dresses and skirts and the avoidance of pants (see Table 2 and Figure 1). For boys, parent responses converged on the avoidance of feminine clothing and in some cases the donning of superhero costumes and formal menswear. Children insisted on wearing gender-typed clothing on a daily basis and often refused other, less gender-typed, suggested options. About 17% of children showed lifetime appearance rigidity in only one category (i.e., dresses); 32% showed rigidity in two or three ways (i.e., dresses and pink); and 12% showed rigidity in between four and six ways. About 40% of parents did not mention specific clothing elements, usually answering with unequivocal yes’s or no’s.

Differences by gender and cohort. We examined gender and cohort differences in both lifetime and current appearance rigidity. Because the distribution of lifetime appearance rigidity was non- parametric (79% of children at the ends of the distribution), we contrasted the most rigid children (4’s) with everyone else. Results revealed that, across cohorts, more girls than boys showed lifetime appearance rigidity, �2(1, N � 76) � 5.66, p � .017, � � .27. In addition, there was a gender by cohort interaction, G2(df � 4) � 14.70, p � .005. Among girls, lifetime appearance rigidity was marginally more prevalent in the younger than in the older cohort, �2(1, N � 39) � 3.17, p � .075, � � .29 (see Figure 2). It seems possible that given the high frequency of lifetime appearance rigidity among 3- to 4-year-olds, some parents of 5- to 6-year-old girls might have underreported past gender-rigid behavior among their daughters. In contrast, lifetime appearance rigidity was more prevalent in boys in the older (44%) than in the younger (11%) cohort, �2(1, N � 37) � 5.39, p � .020, � � .38.

For current gender-typed appearance, a two-way ANOVA re- vealed that although the main effects of gender, F(1, 66) � 1.69, and cohort, F(1, 66) � 0.19, were not significant, a gender by cohort interaction was found, F(1, 66) � 7.11, p � .010, �2 � .10. As expected, girls in the younger (M � 4.32, SD � 0.69) com- pared with the older (M � 3.78, SD � 0.79) cohort were rated as exhibiting a higher level of current gender-typed appearance, t(35) � 2.21, p � .034, d � 0.75. In contrast, boys in the older (M � 4.46, SD � 0.66) compared with the younger (M � 4.08, SD � 0.71) cohort did not significantly differ. Overall, children

Table 1 Percentage of Children Who Have Shown Appearance Rigidity (Lifetime Appearance Rigidity)

Gender/age groups No

interest Little

interest Pretty

interested Insistent

Girls 3- & 4-year-olds 16 11 5 68 5- & 6-year-olds 35 15 10 40

All girls 26 13 8 54 Boys

3- & 4-year-olds 68 21 0 11 5- & 6-year-olds 33 6 17 44

All boys 51 14 8 27

All 3- & 4-year-olds 16 11 5 68 All 5- & 6-year-olds 35 15 10 40 All children 38 13 8 41

Note. Percentages are calculated for each row.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1094 HALIM ET AL.

currently dressed in rather gender-typed ways (overall M � 4.14 on a scale from 1 to 5, SD � .75), especially girls in the younger cohort (M � 4.32 on a scale from 1 to 5, SD � .71).

The role of parents. Finally, we examined the possible asso- ciation between parents’ preferences and children’s appearance rigidity. We did this in several ways. First, we examined the parents’ open-ended responses for spontaneous reactions to this

behavior. In their open-ended responses concerning lifetime ap- pearance rigidity, none of the parents of children who insisted on wearing gender-typed clothing reported actively encouraging or supporting their children to wear gender-typed clothing. Indeed, some parents reported having to resort to some sort of negotiation or persuasion to dress their daughters in a manner that deviated from their daughters’ wishes (e.g., “[She] always prefers a dress. I have to convince her if it’s cold to wear leggings”). Second, we

70 80

ig id

ity

40 50 60 70

pp ea

ra nc

e R

i

Girls

Boys

0 10 20 30

Ex hi

bi te

d A

p

0 3- & 4-yr-olds 5- & 6-yr-olds%

E

Figure 2. Study 1a: Prevalence of lifetime appearance rigidity by gender and cohort.

Table 2 Study 1a: Characteristic Examples of Lifetime Appearance Rigidity

Gender/clothing item Parent quotes

Girls Dresses and skirts “No option, she wears dresses and won’t wear anything else, always stockings even when it is cold out.”

“All the time, skirts—pastel colors, [she] has to wear skirts, [she’s] always done this since [she] turned 3.”

“[She] always prefers a dress. I have to convince her if it’s cold to wear leggings.” “[Her typical outfit is] a dress or a skirt. We do negotiate if [we’re] going to the park, but [she] doesn’t

like jeans unless [they] have embroidered flowers [on them].” “Pretty dress—pink and frilly.” “Dress, tights, Mary Jane shoes; pink is [her] favorite color now.”

Avoidance of pants “Pants are not a choice. [She says,] ‘I want to wear a dress, and that’s it.’ “Dress over skort.” “I bought two pairs of corduroy pants [for her]. She won’t wear them because [they’re] too masculine.” “[She] refuses to wear pants–I have to persuade her. [My daughter says,] ‘I want to wear a dress–I don’t

like pants.’ ” Other “She’d wear [her pink ballet slippers] all the time if she could.”

Boys Avoidance of feminine clothing “He wouldn’t be caught dead in girls’ clothing.”

“Negated burgundy pants because ‘red is for girls.’” “He wouldn’t wear a sweater that is feminine—he won’t wear hand-me-downs from his big sister.” “At [his] grandmother’s, he had no underwear. She put hers on him. He got upset.” “Won’t wear his sister’s pink hand-me-downs.” “Won’t wear pink or purple. Says boys’ stuff is better in general.” “He refuses colors ‘for girls.’” “Wouldn’t wear sandals because they looked like girls’ [sandals].”

Superhero “Superhero costumes—Superman, Spiderman.” “Batman outfits.”

Formal menswear “Suit and tie and shoes and briefcase! [He says to me,] ‘Don’t I look handsome!’” “Loves to wear shirts and ties.”

Other “[He] loves to wear baggy cargo pants.”

Note. Reponses to the question, “Has your daughter/son ever insisted on wearing traditional feminine/masculine clothes whenever she/he went out? Please describe.”

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

Dres se

s/s kir

ts

Colo rs

Te xtu

re

Dec ora

tio ns

Fan cy

vo id

oth er-

ge nd

er

Sup erh

ero

Othe r

% o

f C hi

ld re

n

Girls

Boys

A

Figure 1. Study 1a: Categorization of the way girls and boys manifest appearance rigidity based on parents’ reports.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1095APPEARANCE RIGIDITY

asked parents, “During an average week, how frequently do you attempt to influence your child’s clothing?” (0 � never, 1 � rarely, 2 � a couple times, 3 � often, 4 � a lot). The average response indicated low parental attempts to influence children’s outfits (M � 1.51, SD � 1.20).

Third, we examined whether parents’ preferences for their chil- dren’s clothing in terms of femininity/masculinity were associated with their child’s current appearance rigidity. We conducted a hierarchical multiple regression, in which children’s current ex- pression of appearance rigidity served as the dependent variable with age and gender entered on the first step, parent preferences entered on the second step, the 3 two-way interactions entered on the third step, and the three-way interaction entered on the fourth step. No significant effects were found. The largest coefficient for parents’ preferences predicting children’s current appearance ri- gidity was found on the second step but was not significant, � � �.18, t(65) � 1.36, ns. See supplemental online materials for more details.

Discussion. These data are the first to document the preva- lence of gender appearance rigidity in young children. We used strict criteria for calculating the prevalence of lifetime appearance rigidity, and using these criteria, we found that more than half of girls and a little more than a quarter of boys had at some point strongly insisted on wearing gender-typed clothing (Table 2). On the other end of the spectrum, about a quarter of girls and about half of boys had never shown any appearance rigidity over their lifetimes. These data point to the variability in children’s appear- ance rigidity.

As expected, lifetime gender appearance rigidity was reported to be more prevalent in girls than in boys. In addition, we found that lifetime appearance rigidity was marginally reported more often for 3- and 4-year-old girls (68%) compared with 5- and 6-year-old girls (40%), whereas lifetime appearance rigidity was reported more often for 5- and 6-year-old boys (40%) compared with 3- and 4-year-old boys (11%). Current appearance rigidity showed simi- lar, and significant, patterns for girls, but no differences by cohort for boys. Together these results tentatively suggest a later onset of appearance rigidity for boys. This delay would mirror research showing that girls are often a little ahead of boys in gender development (Zosuls et al., 2009). Alternatively, because boys most frequently expressed appearance rigidity by avoiding any- thing feminine, it might be that it takes longer to learn what to avoid about the other gender (e.g., Chiu et al., 2006), than to learn what to embrace about one’s own gender.

The qualitative data on lifetime appearance rigidity suggested that some children were quite insistent on wearing or avoiding feminine clothing. Qualitatively, parent-reported reactions ranged from tolerance to reasoning with or making compromises with their children. Interestingly, only parents of daughters explicitly mentioned conflict. Perhaps parents of sons were very comfortable with their boys wearing ‘boyish’ clothing. Somewhat surprisingly, parents’ preferences did not predict children’s current gender ap- pearance rigidity.

Study 1b

In Study 1b, we used child interviews to test whether two aspects of self-socialization in gender identity development—(a) feelings of importance and positive evaluation of one’s gender and

(b) understanding of the permanence of gender categories—could help explain why young children want to dress like girly girls and masculine boys. We tested these hypotheses by examining the association between these aspects of gender identity development with their current gender appearance rigidity as reported by their parents in Study 1a.

Method. Participants and procedure. Participants included the 76 chil-

dren of the parents in Study 1a. While parents were being inter- viewed, children were interviewed separately in two sessions at the university or in their schools.

Measures. Current expression of appearance rigidity. Appearance rigid-

ity was measured using the parent reports of current expressions of appearance rigidity described in Study 1a.

Centrality and evaluation. Interviewers asked children about how important and positive being a girl or boy was to them, using 10 items adapted from adult identity measures of centrality and evaluation (e.g., Luhtanen & Crocker, 1992) asked in a Harter-type format (Harter, 1982; Ruble, Taylor, et al., 2007; e.g., to a girl, “Some girls feel that [being a girl is important to them/girls are great], but other girls do not feel that [being a girl is important to them/girls are great]”). Children chose which group they were more like and whether it was “sort of true” or “really true” for them (1 � really not at all true, 2 � sort of not true, 3 � sort of true, 4 � really true). Higher numbers indicated greater centrality and evaluation. We combined centrality and evaluation items because these two components of identification were correlated, r(72) � .59, p � .001. Overall, centrality and evaluation of one’s own gender were moderately high ( � .72; M � 3.41, SD � 0.60; range: 1.60–4.00). No gender differences were found.

Stability constancy. There were two parts of the stability mea- sure. The first part consisted of seven forced-choice questions (e.g., “When you grow up, will you be a man or a woman?”; 0 � incorrect, 1 � correct; Ruble, Taylor, et al., 2007; Slaby & Frey, 1975). The second part consisted of seven sets of drawings where children had to match children and adults by gender (e.g., “What did this adult look like as a child?”; for more details, see Hirschfeld, 1996; Ruble, Taylor, et al., 2007). Children needed to match by gender for responses to be counted as correct. The stability scale was a sum of the 14 total items ( � .71; M � 12.72, SD � 2.02; range: 7.00–14.00). Girls (M � 13.11, SD � 1.49) showed somewhat more understanding of gender stability than did boys (M � 12.31, SD � 2.41), t(57.6) � 1.71, p � .093.

Results. Plan of analyses. We examined whether gender identification

would be associated with current gender-typed appearance by conducting hierarchical multiple regressions. We treated children’s current gender-typed appearance as the dependent variable with age and gender entered on the first step, the identity variable of interest entered on the second, the 3 two-way interactions entered on the third, and the three-way interaction entered on the fourth. All continuous predictors were mean-centered. Gender was dummy coded (girls � 0; boys � 1).

Gender centrality and evaluation. Consistent with our hy- pothesis, gender centrality/evaluation positively predicted chil- dren’s current appearance rigidity adjusting for age and gender, � � .25, t(62) � 2.08, p � .042. We found no other significant effects.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1096 HALIM ET AL.

Gender stability. Confirming our hypothesis, an understand- ing of gender stability was associated with children’s current appearance rigidity, � � .32, t(64) � 2.24, p � .028. We found no other significant effects. See supplemental materials about gender consistency as a predictor.

Discussion. The data lend support to cognitive theories of gender development regarding children’s appearance rigidity. First, we found that gender centrality and evaluation positively predicted children’s current appearance rigidity across age and gender. These findings suggest that children exhibiting appearance rigidity might be doing so, in part, because they consider their gender identity to be an important and positive aspect of them- selves. In addition, we found that a greater understanding of gender stability predicted children’s current appearance rigidity across gender. These data indicate that children’s current appear- ance rigidity might mark a solidified commitment to their gender identities, once children know that gender is relatively permanent over time. These associations are consistent with predictions from the phase model of transitions (Ruble, 1994) with children’s cog- nitive development predicting their identity-displaying behavior.

A limitation of Study 1 was the homogeneity of our sample. We suggest that appearance rigidity could be found in any culture in which gender is an important and salient categorical distinction and in which certain appearance characteristics are closely con- nected with gender. Hence, to examine the generalizability of gender appearance rigidity during early childhood to other cultural communities, we conducted a second study on a different, diverse population.

Study 2

Study 2 examined whether the identified patterns of gender appearance found in Study 1 would generalize to 4-year-old chil- dren from ethnically diverse backgrounds (Mexican, Chinese, Do- minican, and African American) and from low-income neighbor- hoods. Based on the responses of parents in Study 1, we constructed closed-ended questions to assess children’s appear- ance rigidity. We examined mothers of 4-year-olds, in particular, because Study 1 showed that appearance rigidity was prevalent at age 4 and cognitive theories of gender development would also predict a peak in gender-typed appearance at this time. We chose these four particular cultural backgrounds based on their increas- ing presence in the United States and because they each have unique histories and traditions that might affect children’s gender typing. In terms of ethnic group differences, African American children might be expected to display less appearance rigidity than the three immigrant groups because gender distinctions might be less salient among African American families in light of high employment and economic responsibilities of mothers (Hill, 2002; Jarrett, Roy, & Burton, 2002).

Method. Participants and procedure. Participants included 267 moth-

ers of 4-year-olds and their children (129 girls, 138 boys: 59 African American [25 girls, 34 boys], 90 Chinese American [45 girls, 45 boys], 61 Dominican American [27 girls, 34 boys], and 57 Mexican American [32 girls, 25 boys]). Among these participants, 100% of the Chinese, 96% of the Mexican, 79% of the Dominican, and 0% of the African American mothers were born outside the United States. Average household annual income was approxi-

mately $21,823 (SD � $14,576), and 84% were currently receiv- ing government assistance (75% of Dominican, 90% of African American, 86% of Chinese, and 86% of Mexican families). Twenty-five percent of mothers completed some college, 38% completed only high school or received a GED, and 37% did not complete high school. Participants were recruited at the maternity wards of New York City hospitals in low-income areas to take part in a larger longitudinal study on culture and school readiness. Participants were interviewed in their dominant language.

Gender appearance rigidity. We read mothers of sons two statements: “My son avoids wearing feminine clothing and colors like pink,” and “My son loves to wear really masculine things like baseball caps, basketball shoes, and/or sports jerseys” (1 � not at all true, 2 � a little bit true, 3 � somewhat true, 4 � very true, 5 � extremely true). We averaged the responses to these two state- ments to make a scale, M � 3.66, SD � 1.09—for all boys: r(137) � .30, p � .001; by ethnicity: Chinese: r(44) � .58, p � .001; Mexican: r(25) � �.13, ns; Dominican: r(34) � .35, p � .043; African American: r(25) � .10, ns. For mothers of daughters, we read: “My daughter loves to wear dresses and skirts,” and “My daughter loves to wear pink clothing and accessories.” We also averaged the responses to make a scale for girls, M � 3.95, SD � 0.88—all girls: r(129) � .47, p � .001; by ethnicity: Chinese: r(45) � .42, p � .004; Mexican: r(32) � .47, p � .007; Domin- ican: r(27) � .08, ns; African American: r(25) � .64, p � .001 (see the online supplemental materials for more information on the cross-cultural equivalence of the measure). We selected these items from a larger questionnaire on general gender rigidity be- cause they directly asked about the appearance rigidity qualities apparent in Study 1.

Results. Prevalence of gender appearance rigidity by gender. We

analyzed girls’ and boys’ appearance rigidity separately. Among girls, gender appearance rigidity was very high (M � 3.95, SD � 0.88). In fact, 69% of girls were reported to exhibit appearance rigidity, scoring 4 (very true) or 5 (extremely true) on the response scale. This also means that 31% of girls exhibited some (23%), little (5%), or no (3%) appearance rigidity, indicated by scores of 1 (not at all true), 2 (a little bit true), or 3 (somewhat true). Among boys, gender appearance rigidity was also high (M � 3.66, SD � 1.09). A little more than half (56%) of boys were reported to exhibit appearance rigidity (scoring 4’s or 5’s). Thus, 44% of boys exhibited some (20%), little (17%), or no (7%) gender appearance rigidity.

Ethnic group differences in prevalence of gender appearance rigidity. Because the distributions leaned toward being nega- tively skewed (for girls:�1.25; for boys: �0.58), to examine whether rigidity varied by ethnic group, we conducted chi square tests contrasting children who exhibited appearance rigidity (4’s or 5’s) with children who did not (from 1’s to 3’s). Among girls, there were significant differences by ethnicity, �2 (3, N � 129) � 7.87, p � .049, � � .25. Fewer Mexican American girls (50%) showed appearance rigidity compared with Dominican (82%; p � .012), Chinese (73%; p � .036), and African American (72%; p � .093) girls, who did not significantly differ from each other, as deter- mined by follow-up pairwise chi square tests. Hence, appearance rigidity was generally prevalent in these 4-year-old girls from low-income, ethnically diverse backgrounds.

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1097APPEARANCE RIGIDITY

Among boys, there was a marginally significant effect for eth- nicity, �2 (3, N � 138) � 7.37, p � .061, � � .24. More Dominican American boys (77%) were reported to express appear- ance rigidity than did boys from the other ethnic groups—Mexican (48%), Chinese (51%), African American (50%; all ps � .05)— who did not significantly differ from each other. Thus, appearance rigidity was found among more than half of these 4-year-old boys across a diverse number of cultural groups and was even more prevalent among Dominican American boys.

Discussion. Study 2 showed that appearance rigidity is not limited to White, middle-class samples. Two thirds of 4-year-old girls and more than half of 4-year-old boys from diverse ethnic backgrounds, including Chinese, Dominican, and Mexican immi- grant children, as well as African American children, on average showed a moderately high to very high degree of gender appear- ance rigidity. As in Study 1, the data also point to variability in children’s appearance rigidity. One third of ethnic minority 4-year- old girls and 44% of ethnic minority 4-year-old boys showed a more tempered interest or no interest in gender-typed clothing. Perhaps these children might be low on gender typing generally; perhaps their gender identities are not as central to their general self-identities; or perhaps gender is made less salient in their peer environments.

We found a few ethnic differences. Unexpectedly, Mexican girls showed the least amount of gender appearance rigidity, which contrasts with assumptions that Latino culture might endorse more gender typing. Moreover, African American girls showed as much gender appearance rigidity as girls in the other ethnic groups, contrary to the idea that African American girls might endorse gender roles less because of African American family structures (e.g., Hill, 2002). Gender is multidimensional (Ruble et al., 2006); thus, gender rigidity might be expressed in different ways depend- ing on what is emphasized in one’s culture. Perhaps as an alter- native to appearance, Mexican girls express their gender in their play or peer preferences (Halim, Ruble, Tamis-LeMonda, & Shrout, 2013). And perhaps African American girls learn that appearance is particularly important in order to ensure a sense of respectability and social status (Wolcott, 2001). In the case of the ethnic group difference among boys, Dominican boys showed marginally more appearance rigidity than the boys in all the other ethnic groups. Scholars have noted that gender roles are often clearly defined in Dominican culture (Lopez, 2002; Cristofaro & Tamis-LeMonda, 2008). Perhaps gender is particularly salient to Dominican boys, and they are choosing appearance as the choice avenue of gender identity expression.

General Discussion

Across multiple studies, mixed methods, and diverse samples, we found that most girls were reported to show a keen interest in dressing in gender-typed ways at some point in early childhood (68% of 3- and 4-year-olds in Study 1; 69% of 4-year-olds in Study 2). Similarly, but to a lesser degree, a number of boys (more than half of ethnic minority boys, about a quarter of White middle- class boys) also were reported to show an affinity for especially masculine clothing. We consider this prevalence among boys to be remarkable given our expectations that appearance rigidity would be a phenomenon found primarily among girls. Although gender might be more or less salient in different cultural com-

munities, children from multiple ethnic backgrounds—particu- larly girls— on average exhibited a penchant for gendered clothing, thereby demonstrating the robustness of gender ap- pearance rigidity in early childhood.

Gender Differences

Girls generally were reported to exhibit more gender appearance rigidity than were boys across both studies. This contrast might be due to gender differences in the construction of the meaning of gender identity. For girls, being a girl might mean looking like a girl. For boys, we speculate that being a boy might largely mean something else, such as acting like a boy. Children’s stereotypes about girls are largely defined by appearance, whereas children’s stereotypes about boys are more often defined by behavior and activities (Miller et al., 2009). It is also possible that boys do not have to be obsessed with appearances because their wardrobe choices are more constrained, as boys’ clothing already usually excludes feminine options. Boys might have less choice in what they wear and thus find other avenues for the expression of their gender identities.

Although appearance rigidity was less prevalent among boys, when boys did show appearance rigidity, it often revolved around avoiding other-gender-typed clothing, as in Study 1. We speculate that knowledge of status differences could be emerging; thus, boys might desire to avoid looking feminine because females have lower status than males (Rudman & Glick, 2012). It is also possible that boys might be punished more for looking feminine than vice versa (Smetana, 1986). Perhaps also, at an early age, boys’ avoidance of femininity can be seen in their avoidance of feminine clothing, and later on in development this type of avoid- ance is extended to other domains such as boys’ self-censorship in expressing certain emotions (Adler, Kless, & Adler, 1992). Be- cause the avoidance of femininity was a noticeable theme among boys, it would be beneficial in future research to directly and systematically assess the avoidance of cross-gender-typed cloth- ing.

Ethnic Similarities and Differences

Studies 1 and 2 showed that appearance rigidity was comparably prevalent in both White (68% of 3- and 4-year-olds) and ethnic minority (69% of 4-year-olds) young girls. Thus across five dif- ferent ethnic groups, appearance rigidity was the norm among young girls, although in each sample there were girls who did not express the highest levels of appearance rigidity as well (31%– 32% in 3- or 4-year-olds). In contrast, cross-cultural comparisons among boys revealed an interesting finding. In Study 1, only 11% of 3- and 4-year-old White boys showed appearance rigidity. Based on these results, it was unexpected to find that more than half (56%) of 4-year-old ethnic minority boys expressed appear- ance rigidity (and among Dominican 4-year-old boys, the preva- lence rate was 77%). This finding suggests that appearances might be more integral to the gender identities of ethnic minority boys (see Archer & Yamshita, 2003). Alternatively, as 44% of 5- and 6-year-old White boys in Study 1 showed appearance rigidity, nearing the 56% of the 4-year-old ethnic minority boys, it is possible that appearance rigidity has an earlier developmental course for ethnic minority boys than White boys. A future, direct

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1098 HALIM ET AL.

cross-cultural comparison and a longitudinal study on the devel- opmental trajectories of appearance rigidity could elucidate this finding. If other dimensions of gender typing (sex segregation, play) also showed earlier developmental courses for ethnic minor- ity boys than for White boys, this might suggest that general gender identity development might begin earlier for them than for White boys. We speculate that perhaps gender is more salient in the environments of ethnic minority boys compared with the environments of White boys.

Another interesting cultural difference was that in the White, Chinese, and African American groups, appearance rigidity was more prevalent in girls than in boys. However, both Latino groups showed parity in appearance rigidity between girls and boys (82% and 77% of Dominican girls and boys, respectively; 50% and 48% of Mexican girls and boys, respectively). We speculate that the importance of looking feminine or masculine might be stressed equally for girls and boys in Latino groups.

Testing Cognitive Theories of Gender Development

Supporting cognitive theories of gender development, several key findings point to the role of children’s self-socialization as an important addition or alternative to socialization factors in explain- ing gender appearance rigidity. First, a greater understanding of gender stability was associated with currently appearing more gender-typed in Study 1b. Second, currently looking more gender- typed was linked to motivational underpinnings such as positive feelings for and evaluations of one’s own gender group in Study 1b. Third, consistent with predictions from the phase model of transitions (Ruble, 1994), we found in Study 1a that girls’ gender- typed appearances followed the trajectory of rigidity followed by flexibility (marginally for the lifetime appearance rigidity measure, significantly for the current appearance rigidity measure). These cohort differences suggest that appearance rigidity might change over the course of early childhood and follow a phase-like pattern similar to gender-related cognitions, such as gender stereotyping. However, as to the specific timing of rigidity followed by flexi- bility, prior research has shown that the rigidity of gender stereo- typing peaks around ages 5 and 6 and then becomes more flexible (Miller et al., 2006; Trautner et al., 2005). Thus, these data suggest that, for girls, a somewhat earlier period of rigidity might occur for gender-related behavior than for gender-related cognition. It re- mains to be seen if rigid gendered behavior becomes more flexible after age 6 for boys. If so, then boys’ trajectories of gender-typed behavior might align with the trajectories of their gender-related cognitions. Overall, these data suggest that cognitive processes might underlie children’s gender appearance rigidity, though lon- gitudinal research is needed to establish the causal direction of this relation. Future studies should also test whether the same cognitive processes are linked to appearance rigidity in ethnically diverse groups.

Limitations and Future Directions

In the current study, parent preferences were not associated with children’s gender-typed appearances. However, because our pri- mary focus concerned testing cognitive theories of gender devel- opment, our measures of parent preferences were limited in the number of items used. Also, it is possible that some parents might

have been biased in their reports, either being unaware of or downplaying how much they actually encourage their children to dress in stereotypical ways. In addition, it is possible that parent preferences interact with some other factors, such as children’s cognitive development, in affecting their gender-typed dress, but we lacked the statistical power to adequately examine this kind of interaction. Nevertheless, our parent measures had the benefit of being very specific (Zosuls, Ruble, Tamis-LeMonda, & Martin, 2013). More research is needed to understand the intersection between parent preferences and children’s gender-typed dress. Perhaps a more nuanced measure would capture a bidirectional influence between children and parents in determining children’s gendered appearances. Alternatively, parent preferences might not be as strongly connected to appearance rigidity at this age, but at earlier ages, when children are not as attuned to gender categories and are not as invested in their gender identities.

Another limitation was that our measure of appearance rigidity in Study 2 showed low correlations for some of the ethnic by gender groups. However, sample sizes were relatively small when split by ethnic and gender group. Thus, we lacked the power to draw firm conclusions about the cross-cultural equivalence of the measure (Cohen, 1992). In addition, because the items in the measure were very concrete (“skirts,” “baseball caps”) and did not ask about abstract concepts (like “appearance rigidity”), we are confident that mothers from all ethnic groups understood the questions. Nevertheless, it would be interesting for investigators in a future study to directly examine the particular ways gender manifests itself in the adornments and dress of different cultural groups.

The present findings suggest a number of other important di- rections for future research. One issue concerns the social identity implications of gender-related appearance rigidity. For example, if appearance rigidity represents strong gender identification, as our research suggests, then children who are rigid might show more ingroup favoritism than those who are more flexible. Our finding that girls showed more gender appearance rigidity than boys is consistent with literature sometimes reporting that girls show more intergroup bias than do boys (Leroux, 2008; Powlishta, 1995; Susskind & Hodges, 2007; Zosuls et al., 2011).

Another important direction concerns the causes and conse- quences of individual differences in appearance rigidity. Although we have emphasized how prevalent appearance rigidity is for girls, it is important to remember that there was also a subset of girls who exhibited no appearance rigidity. And although, depending on the sample, about a quarter to almost half of boys exhibited appearance rigidity, many boys never did. What makes these individuals different from one another? Some children could be generally more gender-typed than others due to experiencing dif- ferent hormonal environments in utero (e.g., girls exposed to high levels of androgens could prefer more masculine dress, activities, and peers; Berenbaum & Snyder, 1995) or experiencing different amounts of felt pressure to conform to gender norms from family, peers, or the media (Egan & Perry, 2001). Alternatively, perhaps some children are defining their gender through other avenues besides their appearance.

Whatever the explanation for these individual differences, it would be interesting to track these differences over time and understand the developmental trajectory of appearance rigidity. Journalists have increasingly discussed the negative impact of the

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1099APPEARANCE RIGIDITY

princess culture on girls’ development (e.g., Fine, 2010; Orenstein, 2010). On the one hand, in light of the lower levels of appearance rigidity among older girls in Study 1, which suggests a dearth of elementary-school-aged girls who wear pink, frilly dresses (Halim et al., 2011), we posit that appearance rigidity might be a short- lived phase. If so, then perhaps appearance rigidity need not elicit stress around parent–child interactions, potentially damaging parent–child relationships and making children perceive that ex- pressing one’s gender identity is bad. On the other hand, if ap- pearance rigidity is more long-term and stable, then it might lead children, and especially girls, to focus too much on their physical appearance, especially if they are continually praised by others for looking pretty. Appearance rigidity might then feed into defining one’s self and one’s self-worth in terms of how one looks. In turn, one’s self-esteem might become contingent on self-perceived at- tractiveness, which can contribute to psychological distress (Crocker & Wolfe, 2001; Eccles, Barber, Jozefowicz, Malenchuk, & Vida, 1999). A hyper-focus on one’s physical appearance might also feed into girls’ self-objectification, which has also been as- sociated with poor psychological adjustment and math perfor- mance (e.g., Fredrickson, Roberts, Noll, Quinn, & Twenge, 1998).

Conclusion

In conclusion, through multiple methods and across studies in a large and diverse sample, our study emphasizes that the clothing that children put on each day has significance and is a central, but previously missing, piece in the study of gender and identity development. Whether a girl dons a pink, frilly dress or a boy wears a red and blue Spiderman T-shirt might reflect changing understandings of gender categories and developing motivations to master these categories. Thus, the present findings support the view of young children as active self-socializing agents, picking up clues on what gender looks like and doggedly following their deductions.

References

Adler, P. A., Kless, S. J., & Adler, P. (1992). Socialization to gender roles: Popularity among elementary school boys and girls. Sociology of Edu- cation, 65, 169–187. doi:10.2307/2112807

Archer, L., & Yamashita, H. (2003). Theorising inner-city masculinities: “Race,” class, gender and education. Gender and Education, 15, 115– 132. doi:10.1080/09540250303856

Bem, S. L. (1981). Gender schema theory: A cognitive account of sex typing. Psychological Review, 88, 354–364. doi:10.1037/0033-295X.88 .4.354

Berenbaum, S. A., & Snyder, E. (1995). Early hormonal influences on childhood sex-typed activity and playmate preferences: Implications for the development of sexual orientation. Developmental Psychology, 31, 31–42. doi:10.1037/0012-1649.31.1.31

Chiu, S. W., Gervan, S., Fairbrother, C., Johnson, L. L., Owen-Anderson, A. F. H., Bradley, S. J., & Zucker, K. J. (2006). Sex-dimorphic color preference in children with gender identity disorder: A comparison to clinical and community controls. Sex Roles, 55, 385–395. doi:10.1007/ s11199-006-9089-9

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. doi:10.1037/0033-2909.112.1.155

Cristofaro, T. N., & Tamis-LeMonda, C. S. (2008). Lessons in mother– child and father–child personal narratives in Latino families. In A. McCabe, A. Bailey, & G. Melzi (Eds.) Spanish-language narration and

literacy: Culture, cognition, and emotion (pp. 54–91). Cambridge, MA: Cambridge University Press. doi:10.1017/CBO9780511815669.006

Crocker, J., & Wolfe, C. T. (2001). Contingencies of self-worth. Psycho- logical Review, 108, 593–623. doi:10.1037/0033-295X.108.3.593

DeLoache, J. S., Simcock, G., & Macari, S. (2007). Planes, trains, auto- mobiles—and tea sets: Extremely intense interests in very young chil- dren. Developmental Psychology, 43, 1579–1586. doi:10.1037/0012- 1649.43.6.1579

Eccles, J., Barber, B., Jozefowicz, D., Malenchuk, O., & Vida, M. (1999). Self-evaluations of competence, task values, and self-esteem. In, N. G. Johnson, M. C. Roberts, & J. Worell (Eds.), Beyond appearance: A new look at adolescent girls (pp. 53–83). Washington, DC: American Psy- chological Association. doi:10.1037/10325-002

Egan, S. K., & Perry, D. G. (2001). Gender identity: A multidimensional analysis with implications for psychosocial adjustment. Developmental Psychology, 37, 451–463. doi:10.1037/0012-1649.37.4.451

England, D. E., & Descartes, L., & Collier-Meek, M. A. (2011). Gender role portrayal and the Disney princesses. Sex Roles, 64, 555–567. doi: 10.1007/s11199-011-9930-7

Feinberg, R. A., Mataro, L., & Burroughs, W. J. (1992). Clothing and social identity. Clothing & Textiles Research Journal, 11, 18–23. doi: 10.1177/0887302X9201100103

Fine, C. (2010). Delusions of gender. New York, NY: Norton. Fredrickson, B. L., Roberts, T., Noll, S. M., Quinn, D. M., & Twenge, J. M.

(1998). That swimsuit becomes you: Sex differences in self- objectification, restrained eating, and math performance. Journal of Personality and Social Psychology, 75, 269–284. doi:10.1037/0022- 3514.75.1.269

Freitas, A., Kaiser, S., & Hammidi, T. (1996). Communities, commodities, cultural space, and style. Journal of Homosexuality, 31, 83–107.

Gutman, H. G. (1976). The Black family in slavery and freedom. New York, NY: Vintage.

Halim, M. L., & Ruble, D. N. (2010). Gender identity and stereotyping in early and middle childhood. In J. Chrisler & D. McCreary (Eds.), Handbook of gender research in psychology (pp. 495–525). New York, NY: Springer. doi:10.1007/978-1-4419-1465-1_24

Halim, M. L., Ruble, D. N., & Amodio, D. M. (2011). From pink frilly dresses to “one of the boys”: A social–cognitive analysis of gender identity development and gender bias. Social and Personality Psychol- ogy Compass, 5, 933–949. doi:10.1111/j.1751-9004.2011.00399.x

Halim, M. L., Ruble, D. N., Tamis-LeMonda, C. S., & Shrout, P. E. (2013). Rigidity in gender-typed behaviors in early childhood: A longitudinal study of ethnic minority children. Child Development, 84, 1269–1284.

Harter, S. (1982). The Perceived Competence Scale for Children. Child Development, 53, 87–97. doi:10.2307/1129640

Hartley, R. E. (1959). Sex-role pressures and socialization of the male child. Psychological Reports, 5, 457–468. doi:10.2466/PR0.5.457-468

Hill, S. A. (2002). Teaching and doing gender in African American families. Sex Roles, 47, 493–506. doi:10.1023/A:1022026303937

Hirschfeld, L. A. (1996). Race in the making: Cognition, culture, and the child’s construction of human kinds. Cambridge, MA: MIT Press.

Hofstede, F. (1980). Culture’s consequences. London, United Kingdom: Sage.

Huston, A. C. (1983). Sex-typing. In E. M. Hetherington (Ed.) & P. H. Mussen (Series Ed.), Handbook of child psychology: Vol. 4. Socializa- tion, personality, and social development (pp. 387–467). New York, NY: Wiley.

Jarrett, R., Roy, K., & Burton, L. (2002). Fathers in the ’hood: Qualitative research on African American men. In C. Tamis-LeMonda & N. Cabrera (Eds.), Handbook of father involvement: Multidisciplinary perspectives (pp. 211–248). Hillsdale, NJ: Erlbaum.

Julian, T. W., McKenry, P. C., & McKelvey, M. W. (1994). Cultural variations in parenting: Perceptions of Caucasian, African-American,

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1100 HALIM ET AL.

Hispanic, and Asian-American parents. Family Relations, 43, 30–37. doi:10.2307/585139

Kane, E. W. (2000). Racial and ethnic variations in gender-related atti- tudes. Annual Review of Sociology, 26, 419–439. doi:10.1146/annurev .soc.26.1.419

Leroux, A. (2008). Do children with gender identity disorder have an in-group or an out-group gender-based bias? (Unpublished master’s thesis). Ontario Institute for the Study in Education of the University of Toronto, Toronto, Ontario, Canada.

Lopez, N. (2002). Hopeful girls, troubled boys: Race and gender disparity in urban education. New York, NY: Routledge.

Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self- evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18, 302–318. doi:10.1177/0146167292183006

Martin, C. L., & Ruble, D. N. (2004). Children’s search for gender cues: Cognitive perspectives on gender development. Current Directions in Psychological Science, 13, 67–70. doi:10.1111/j.0963-7214.2004 .00276.x

Martin, C. L., Ruble, D. N., & Szkrybalo, J. (2002). Cognitive theories of early gender development. Psychological Bulletin, 128, 903–933. doi: 10.1037/0033-2909.128.6.903

Miller, C. F., Lurye, L. E., Zosuls, K. M., & Ruble, D. N. (2009). Accessibility of gender stereotypes domains: Developmental and gender differences in children. Sex Roles, 60, 870–881. doi:10.1007/s11199- 009-9584-x

Miller, C. F., Trautner, H., & Ruble, D. N. (2006). The role of gender stereotypes in children’s preferences and behavior. In L. Balter & C. S. Tamis-LeMonda (Eds.), Child psychology: A handbook of contemporary issues (pp. 293–323). New York, NY: Psychology Press.

Neppl, T. K., & Murray, A. D. (1997). Social dominance and play patterns among preschoolers: Gender comparisons. Sex Roles, 36, 381–393. doi:10.1007/BF02766654

Orenstein, P. (2010). Cinderella ate my daughter. New York, NY: Harper. Padawar, R. (2012, August 12). Boygirl. New York Times Magazine, pp.

18–23, 36, 46. Paley, V. G. (1986). Boys and girls: Superheroes in the doll corner.

Chicago, IL: University of Chicago Press. Pickering, S., & Repacholi, B. (2001). Modifying children’s gender-typed

musical instrument preferences: The effects of gender and age. Sex Roles, 45, 623–643. doi:10.1023/A:1014863609014

Powlishta, K. K. (1995). Intergroup processes in childhood: Social cate- gorization and sex role development. Developmental Psychology, 31, 781–788. doi:10.1037/0012-1649.31.5.781

Ruble, D. N. (1994). A phase model of transitions: Cognitive and motivational consequences. In M. Zanna (Ed.), Advances in experimental social psy- chology (pp. 163–214). New York, NY: Academic Press. doi:10.1016/ S0065-2601(08)60154-9

Ruble, D. N., & Dweck, C. S. (1995). Self-perceptions, person concep- tions, and their development. In N. Eisenberg (Ed.), Social development (pp. 109–139). Thousand Oaks, CA: Sage.

Ruble, D. N., Lurye, L. E., & Zosuls, K. M. (2007). Pink frilly dresses (PFD) and early gender identity. Princeton Report on Knowledge (P–ROK), 2(2). Retrieved from http://www.princeton.edu/prok/issues/2- 2/pink_frilly.xml

Ruble, D. N., Martin, C., & Berenbaum, S. (2006). Gender development. In N. Eisenberg (Vol. Ed.), W. Damon, & R. M. Lerner (Series Eds.), Handbook of child psychology: Vol. 3. Personality and social develop- ment (6th ed., pp. 858–932)). New York, NY: Wiley.

Ruble, D. N., Taylor, L. J., Cyphers, L., Greulich, F. K., Lurye, L. E., & Shrout, P. E. (2007). The role of gender constancy in early gender

development. Child Development, 78, 1121–1136. doi:10.1111/j.1467- 8624.2007.01056.x

Rudman, L. A., & Glick, P. (2012). Social psychology of gender: How power and intimacy shape gender relations. New York, NY: Guilford Press.

Slaby, R. G., & Frey, K. S. (1975). Development of gender constancy and selective attention to same-sex models. Child Development, 46, 849– 856. doi:10.2307/1128389

Smetana, J. G. (1986). Preschool children’s conceptions of sex-role trans- gressions. Child Development, 57, 862–871. doi:10.2307/1130363

Smetana, J. G., & Letourneau, K. J. (1984). Development of gender constancy and children’s sex-typed free play behavior. Developmental Psychology, 20, 691–696. doi:10.1037/0012-1649.20.4.691

Stangor, C., & Ruble, D. N. (1987). Development of gender role knowl- edge and gender constancy. In L. S. Liben & M. L. Signorella (Eds.), New directions for child development: Vol. 39. Children’s gender sche- mata (pp. 5–22). San Francisco, CA: Jossey–Bass.

Susskind, J. E., & Hodges, C. (2007). Decoupling children’s gender-based in-group positivity from out-group negativity. Sex Roles, 56, 707–716. doi:10.1007/s11199-007-9235-z

Tamis-Lemonda, C. S., & McFadden, K. E. (2009). The United States of America. In M. H. Bornstein (Ed.), Handbook of cultural developmental science (pp. 299–322). New York, NY: Psychology Press.

Tobin, D. D., Menon, M., Menon, M., Spatta, B. C., Hodges, E. V. E., & Perry, D. G. (2010). The intrapsychics of gender: A model of self- socialization. Psychological Review, 117, 601– 622. doi:10.1037/ a0018936

Trautner, H. M., Ruble, D. N., Cyphers, L. Kirsten, B., Behrendt, R., & Hartmann, P. (2005). Rigidity and flexibility of gender stereotypes in children: Developmental or differential? Infant and Child Development, 14, 365–381. doi:10.1002/icd.399

Tu, W. M. (1985). Selfhood and otherness in Confucian thought. In A. J. Marsella, G. De Vos, & F. L. K. Hsu (Eds.), Culture and the self (pp. 231–251). New York, NY: Tavistock.

Wolcott, V. W. (2001). Remaking respectability: African American women in interwar Detroit. Chapel Hill, NC: University of North Carolina Press.

Zosuls, K. M., Miller, C. F., Ruble, D. N., Martin, C. L., & Fabes, R. A. (2011). Gender development research in sex roles: Historical trends and future directions. Sex Roles, 64, 826–842. doi:10.1007/s11199-010- 9902-3

Zosuls, K. M., Ruble, D. N., Tamis-LeMonda, C. S., & Martin, C. L. (2013). Does your infant say the words “girl” and “boy”? How gender labels matter in early gender development. In M. R. Banaji & S. A. Gelman (Eds.), Navigating the social world: What infants, children, and other species can teach us (pp. 301–305). Oxford, United Kingdom: Oxford University Press.

Zosuls, K. M., Ruble, D. N., Tamis-LeMonda, C. S., Shrout, P. E., Bornstein, M. H., & Greulich, F. K. (2009). The acquisition of gender labels in infancy: Implications for sex-typed play. Developmental Psy- chology, 45, 688–701. doi:10.1037/a0014053

Zucker, K. J., Bradley, M. D., Kuksis, M., Pecore, K., Birkenfeld-Adams, A., & Doering, R. W. (1999). Gender constancy judgments in children with a gender identity disorder: Evidence for a developmental lag. Archives of Sexual Behavior, 28, 475–502. doi:10.1023/A: 1018713115866

Received March 10, 2011 Revision received July 8, 2013

Accepted August 21, 2013 �

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an Ps

yc ho

lo gi

ca l

A ss

oc ia

tio n

or on

e of

its al

lie d

pu bl

is he

rs .

T hi

s ar

tic le

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

1101APPEARANCE RIGIDITY