Content uploaded by Eric Rasmussen
Author content
All content in this area was uploaded by Eric Rasmussen on Jul 07, 2016
Content may be subject to copyright.
Pretty as a Princess: Longitudinal Effects of Engagement With Disney
Princesses on Gender Stereotypes, Body Esteem, and Prosocial Behavior in
Children
Sarah M. Coyne
Brigham Young University
Jennifer Ruh Linder
Linfield College
Eric E. Rasmussen
Texas Tech University
David A. Nelson and Victoria Birkbeck
Brigham Young University
This study examined level of engagement with Disney Princess media/products as it relates to gender-
stereotypical behavior, body esteem (i.e. body image), and prosocial behavior during early childhood.
Participants consisted of 198 children (M
age
=58 months), who were tested at two time points (approxi-
mately 1 year apart). Data consisted of parent and teacher reports, and child observations in a toy prefer-
ence task. Longitudinal results revealed that Disney Princess engagement was associated with more female
gender-stereotypical behavior 1 year later, even after controlling for initial levels of gender-stereotypical
behavior. Parental mediation strengthened associations between princess engagement and adherence to
female gender-stereotypical behavior for both girls and boys, and for body esteem and prosocial behavior
for boys only.
“I am a princess. All girls are.”—Sara Crewe. (A
Little Princess; Burnett, 1905)
Girls and women are often stereotyped and sexual-
ized in the mass media (e.g., Collins, 2011; Smith,
Pieper, Granados, & Choueite, 2010). An examina-
tion of the media during early childhood is particu-
larly important, as these years lay down a
foundation for gender role development over time
(Bussey & Bandura, 1999), and media may act as a
key socializing agent for gender role development
(for a review, see Signorielli, 2011). A meta-analysis
of 30 studies found that television viewing can
develop and reinforce children’s attitudes regarding
gender stereotypes (Herrett-Skjellum & Allen, 1996).
Notably, female stereotypes in the media influence
both female and male attitudes about gender,
though effects are much stronger for girls than boys
(Bussey & Bandura, 1999). Additionally, exposure
to gender-stereotypical media can alter children’s
perceptions of the gender appropriateness of toys
(Pike & Jennings, 2005) and is associated with more
gender-typed play among preschoolers (Coyne, Lin-
der, Rasmussen, Nelson, & Collier, 2014).
One of the most popular types of media and
merchandise for young girls is the Disney Princess
line. The line is highly profitable, with sales in 2012
alone exceeding more than $3 billion (Goudreau,
2012). There is some public concern regarding the
effect that Disney Princesses may have on young
girls (e.g., Orenstein, 2011). However, there is little
empirical research examining how Disney Prin-
cesses are portrayed in the media and what effect
they might have on children’s gender role develop-
ment, attitudes, and behavior. Accordingly, the aim
of the current study is to examine longitudinal
associations between exposure to Disney Princess
media and gender-stereotypical behavior, body
esteem, and prosocial behavior for preschool and
kindergarten age children.
We would like to acknowledge the Women’s Research Initia-
tive at BYU for financially supporting this project. We would
also like to thank all the student research assistants for their help
throughout the project. Finally, Sarah M. Coyne would like to
thank Peggy Orenstein and her daughter, Hannah Coyne for
inspiring this research.
Correspondence concerning this article should be addressed to
Sarah M. Coyne, School of Family Life, Brigham Young Univer-
sity, JFSB 2087, Provo, UT 84602. Electronic mail may be sent to
smcoyne@byu.edu.
©2016 The Authors
Child Development ©2016 Society for Research in Child Development, Inc.
All rights reserved. 0009-3920/2016/xxxx-xxxx
DOI: 10.1111/cdev.12569
Child Development, xxxx 2016, Volume 00, Number 0, Pages 1–17
Disney Princesses and Gender
The remarkable profitability of the Disney Prin-
cess line is clear evidence of its popularity among
young girls (Disney Consumer Products, 2011). As
of 2015, the Disney Princess line involves 13 female
Disney characters identified by the corporation,
many of whom are princesses by birth or marriage
(e.g., Cinderella), though some are not (e.g., Mulan).
The princess franchise involves multiple ways for
children to engage with the Disney characters,
including films, toys, clothing, and more. According
to Orenstein (2011), parents generally like Disney
Princesses and view them as “safe”compared to
other, highly sexualized media models. Yet others
question this sense of safety with the Disney Prin-
cess brand and speculate that young girls may be
particularly vulnerable to potential negative effects
(Best & Lowney, 2009; Ehrenreich, 2007). The prin-
cess line, in particular, has drawn criticism for
glamorizing characters who essentially are passive
and need to be saved by men (Ehrenreich, 2007;
Orenstein, 2011).
A few content analyses reveal that Disney Prin-
cess films send a number of strong messages
regarding gender. One content analysis of Disney
Princess movies from 1937 to 2009 found that,
although portrayals of gender have become more
complex over time, there are still strong messages
of traditional gender role stereotypes for girls and
women (e.g., physically weak, affectionate, nurtur-
ing, helpful, fearful, submissive) in many of the
movies (England, Descartes, & Collier-Meek, 2011).
Although gender messages in Disney movies are
becoming more progressive and varied as time goes
on (England et al., 2011), older Disney movies (such
as Snow White, released in 1937, or Cinderella,
released in 1950) remain extremely popular with
children today (Do Rozar
ıa, 2004), and their mes-
sages are therefore still relevant to children’s learn-
ing and development. Additionally, newer, less
gender-stereotyped princesses (e.g., Merida) tend to
be feminized for Disney merchandise (Samakow,
2013).
Very little research has examined the effects of
viewing Disney Princess media on children’s inter-
nalization of gender stereotypes. Wohlwend (2009)
conducted an ethnographic study that followed
young girls who were avid Disney Princess fans
across 3 years. Results revealed that these girls not
only used the princess storylines (sometimes modi-
fied) in their own personal play but also showed
highly gendered expectations. This study gives
some evidence to the idea that Disney Princesses
can influence behaviors and attitudes; however, the
Disney Princess “group”only consisted of three
kindergarten girls, so it is difficult to generalize the
results with any confidence. Additionally, although
not examining Disney Princesses specifically,
Dinella (2013) found that emerging adult women
who self-identified themselves as a “princess”
reported less desire to work, expected more tradi-
tional divisions of household labor, and placed
greater value on superficial qualities, such as
appearance. They also put forth less effort in a chal-
lenging behavioral task. Although Dinella (2013)
did not specifically examine the role of the media
in her study, it suggests that there may be a long-
term effect of internalizing a princess culture early
in life.
Additionally, although Disney Princesses are pre-
dominantly popular with girls, many preschool
boys watch these programs. For example, Frozen,
the best-selling Disney Princess movie of all time,
was extremely popular with both girls and boys
(Gomez, 2014). Research has found that boys can
learn gender stereotypes from watching female her-
oines in the media and vice versa. For example,
Coyne et al. (2014) found that viewing superhero
programs was longitudinally associated with
engagement in some forms of male gender-stereo-
typical play (i.e., playing with weapons) for girls,
even though viewing superhero programs was rare
among girls. Accordingly, boys who frequently
view Disney Princess programs may engage in less
traditionally masculine gender-stereotypical play,
relative to other boys, particularly if they identify
with the leading female characters (Bussey & Ban-
dura, 1999). Alternatively, boys who frequently
watch these programs may instead identify with
the male characters (e.g., the princes), who typically
exhibit a mixture of feminine and masculine traits
(England et al., 2011). Therefore, in order to exam-
ine associations with princess identification and
involvement, the current study focused on partici-
pants’identification and involvement with Disney
Princess characters (rather than Disney media and
products in general).
Given Disney’s wide marketing and product
campaign, children also have frequent opportunities
to rehearse the gender stereotypes they view in the
movies through playing with princess toys, dress-
ing up in princess costumes, and more. To our
knowledge, no research has been conducted on Dis-
ney Princess toys and gender stereotypes. However,
playing with Barbie dolls (which are in recent years
connected to Barbie movies) is associated with girls
seeing fewer career options for themselves in the
2 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
future (Sherman & Zurbriggen, 2014). This study
suggests that playing with gendered toys, particu-
larly those associated with a movie franchise, may
also promote internalization of gender-stereotypical
expectations in early childhood.
Social Cognitive Theory of Gender Development
The social cognitive theory of gender develop-
ment explains how exposure to gender-stereotyped
behavior in the media and playing with gendered
toys may have an influence on the development of
gender stereotypes and gendered behavior in young
children (Bussey & Bandura, 1999). According to
this theory, gender development is a complex pro-
cess that is shaped by two different mechanisms.
The first is modeling, whereby children learn gen-
dered behavior by watching others, including par-
ents, peers, teachers, and the media. In this process,
children tend to learn best from powerful, salient,
attractive, same-gender models. Social cognitive
theory also suggests that models who are rewarded
for their behavior and who are attractive role mod-
els should be particularly influential. Disney Prin-
cesses represent salient, powerful, attractive
characters who tend to be portrayed as conforming
to gender stereotypes and are rewarded for their
gendered behavior (England et al., 2011). Accord-
ingly, this theory would suggest that Disney Prin-
cesses may be a particularly potent model for the
learning of gendered behavior in children. Given
that Disney Princesses are female, and are predomi-
nantly popular among young girls, we would
expect the effects of viewing such media on behav-
iors to be stronger for girls than boys, and espe-
cially for those girls who highly identify with
Disney Princesses. Although boys could learn gen-
der stereotypes and norms from viewing Disney
Princess media, this learning may not readily trans-
fer into behavior. In particular, conformity to tradi-
tional gender stereotypes is more pronounced and
valued for boys than for girls, and viewing, identi-
fying with, and playing with Disney Princess toys
represent a fairly large departure from traditional
masculine norms (Bussey & Bandura, 1999). In
addition, boys may be less likely to identify with
and therefore learn from female characters. Accord-
ingly, we expect any effects on gender-stereotypical
behavior to be weaker for boys than girls.
A second process outlined by social cognitive
theory that is important for gender development of
young children is direct tuition, in which children
are encouraged and rewarded for engaging in
behaviors considered appropriate for their sex, and
are discouraged from gender-inappropriate activi-
ties. In early childhood, access to toys and media
content is primarily under the control of parents;
therefore, high exposure to Disney Princess prod-
ucts is likely to reflect parental endorsement and
encouragement of gender-typed behavior. The
degree to which parents engage in direct tuition
has been shown to be associated with children’s
gender-typed toy preferences and gender stereo-
types (e.g., Fagot, Leinbach, & O’Boyle, 1992).
Beginning in the preschool period and continuing
through childhood, peers also engage in direct tui-
tion by reinforcing gender-typed toy selection and
affiliation with same-sex play partners (Fagot,
1977). The widespread popularity of Disney Prin-
cesses among preschool and kindergarten girls sug-
gests that peers likely reinforce involvement with
princess toys and media products. Therefore, direct
tuition through parents and peers may be an
additional mechanism beyond modeling by which
Disney Princess exposure influences gender devel-
opment.
Theories of gender development emphasize the
early childhood period as an important develop-
mental time period for the acquisition of gender
stereotypes and development of gendered patterns
of behavior. Although knowledge of gender stereo-
types and some gender-typed play patterns (e.g.,
toy preferences) first emerge in the toddler years,
gender stereotypes become more rigid between the
ages 3 and 6, and gender-typed play patterns and
toy preferences also become stronger during this
age period (e.g., Ruble & Martin, 1998). As such,
this is an important developmental period for the
socialization of gender-typed behaviors and stereo-
types, and therefore the current study focused on
children of preschool and kindergarten age.
Body Esteem
The physical appearance of Disney Princesses
has also been a common topic in the news media,
popular press, and research literature (Orenstein,
2011). The typical princess is portrayed as young
and attractive with large eyes, small nose and chin,
moderately large breasts, prominent cheekbones,
lustrous hair, and good muscle tone and skin com-
plexion (Gangestad & Scheyd, 2005; Lacroix, 2004).
Additionally, princesses generally embody a form
of “thin-ideal”media, meaning that they represent
an unrealistically thin female figure as the most
positive and desirable (Lacroix, 2004). Studies have
shown that as early as preschool, children begin to
express a preference for thin body types, and girls
Disney Princesses and Child Development 3
as young as 5 years old express fears of getting fat
or show problems with body esteem, a self-evalua-
tion of one’s body and appearance (Tremblay,
Lovsin, Zecevic, & Lariviere, 2011). Although
preschoolers may not understand the sexual impli-
cations, many are aware that body fat is undesir-
able for women (Tremblay et al., 2011), and that it
is important for women to be pretty (Smolak &
Murnen, 2011). Disney Princess movies, their por-
trayal of the thin ideal, and emphasis on the prin-
cess’beauty may be an early context in which girls
are taught that attractiveness is a necessary compo-
nent of female identity.
According to research on the thin-ideal internal-
ization, a belief that “thin is good”results when
individuals internalize perceptions that are voiced
by significant or respected individuals, including
parents, peers, and the media (Thompson & Stice,
2001). Media messages regarding the thin ideal are
both pervasive and powerful and may negatively
influence body esteem in viewers (Hohlstein, Smith,
& Atlas, 1998). Indeed, a number of meta-analyses
show that heightened exposure to the thin ideal in
media is associated with internalization of this stan-
dard in girls (Grabe, Ward, & Hyde, 2008; Want,
2009). Although most research examines the effect
of media on older girls and women, longitudinal
studies suggest that early exposure to the thin ideal
in media (around the age of 5 years old) predicts
appearance-related concerns in the future (Dohnt &
Tiggemann, 2006). Thin-ideal media is also associ-
ated with disordered eating in early elementary-
aged children both concurrently and longitudinally
(Harrison, 2000; Moriarty & Harrison, 2008). Not
surprisingly, playing with thin-ideal dolls (e.g.,
Barbie) is also associated with lower body esteem,
especially in younger girls (e.g., Anschutz & Engels,
2010). Accordingly, Disney Princess movies and the
associated marketing are concerning because they
potentially represent some of the earliest exposure
to thin-ideal media in a girl’s life.
To our knowledge, there is only one study exam-
ining how viewing Disney Princesses influences
body esteem in children. Hayes and Tantleff-Dunn
(2010) showed 3- to 6-year-old girls a series of
appearance-related clips, including several from
Disney movies (e.g., Cinderella,Little Mermaid).
Compared to a control group, these girls did not
experience any body dissatisfaction or engage in
any appearance-related play behaviors. Although
this study suggests that effects of short-term expo-
sure to thin-ideal media may not be particularly
strong on young girls, the study does not examine
the cumulative effects over time. It may be that
early exposure to thin-ideal media may influence
later concerns with body esteem. Accordingly, the
current study examined the effect of exposure to
Disney Princess media on body esteem over time.
Although we examine the potential effects of Dis-
ney Princesses on body esteem in boys, we do not
expect the body esteem of boys to be affected by
the thin ideal. Body esteem in boys tends to be
influenced by exposure to media characters with a
“muscular ideal”(e.g., large muscles, small waist)
as opposed to the “thin ideal”epitomized by Dis-
ney Princesses (e.g., Hargreaves & Tiggemann,
2009).
Prosocial Behavior in Disney Films
Although media research often focuses on prob-
lematic content and effects, it is important to note
that media also have the potential to be a positive
socializing agent for children. For example, a meta-
analysis of 34 studies regarding prosocial television
viewing found that children who watched prosocial
content were more likely to behave positively or
have more positive attitudes (Mares & Woodard,
2005). Consistent with social cognitive theory,
media portrayals of rewarded acts of prosocial
behavior by attractive characters are especially
likely to result in learning of these behaviors by
viewers.
Although research has not examined Disney
Princesses specifically, Padilla-Walker, Coyne,
Fraser, and Stockdale (2013) found that Disney
movies portray extremely high levels of prosocial
behaviors, at the rate of nearly one act per minute.
Anecdotally, many of the Disney Princesses are
extremely prosocial. For example, in Beauty and the
Beast, Belle sacrifices herself in order to save her
father. Indeed, Padilla-Walker et al. (2013) found
that two of the three very most “prosocial”Disney
films were Pocahontas and Mulan, two princesses
from Disney’s line.
Both social cognitive theory (Bandura, 2001) and
the general learning model (Buckley & Anderson,
2006) suggest that exposure to prosocial media may
influence prosocial behavior among viewers. Social
cognitive theory highlights the process of modeling
of characters by viewers, resulting in the learning
of both behaviors and scripts. Similarly, the general
learning model highlights the development of atti-
tudes, beliefs, and goals as a result of exposure to
media. Specifically, the general learning model sug-
gests that, in both the short term and the long term,
exposure to prosocial media leads to the develop-
ment of cognitive scripts supportive of prosocial
4 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
behavior among viewers (Buckley & Anderson,
2006). Very little research has examined how view-
ing prosocial behavior in the media influences
young children, and none to our knowledge focus
specifically on Disney Princesses. Accordingly, the
current study will examine effects of exposure to
Disney Princesses on prosocial behavior, both in the
short term and long term.
Parental Mediation of Media
The social cognitive theory of gender develop-
ment also suggests that the process of direct tuition
occurs when parents verbally discuss appropriate
gender behavior with children (Bussey & Bandura,
1999). Direct tuition about media depictions of gen-
der roles and behavior likely takes place within the
broader context of parental mediation of children’s
media exposure. Accordingly, in the current study,
we will examine whether parental mediation moder-
ates the effect of exposure to Disney Princess media.
Parental mediation refers to parents’efforts to
“provide guidance about and control over exposure
to certain media offerings”(Hogan, 2001, p. 664).
One form of parental mediation that parents fre-
quently engage in is active mediation, which is
defined as parent–child discussion of the child’s
exposure to certain media content. Through active
mediation, parents seek to impact the extent to
which children are influenced by the messages
received via media exposure (Rasmussen, 2013).
Past research shows that active mediation is
positively associated with children’s acceptance of
nontraditional gender roles (Corder-Bolz, 1980),
self-esteem, and body satisfaction (Schooler, Kim, &
Sorsoli, 2006), and negatively associated with accep-
tance of gender stereotypes (Nathanson, Wilson,
McGee, & Sebastian, 2002). Other research, how-
ever, shows that active mediation is related to
greater body image disturbance (Nathanson &
Botta, 2003). In addition, past research has also
demonstrated that active mediation with preschool-
ers, who have limited cognitive abilities, may have
a minimal or negative impact on media effects
(Coyne et al., 2014; Wilson & Cantor, 1987).
Accordingly, we will assess whether parental medi-
ation practices moderate any associations between
Disney Princess engagement and outcomes.
The Current Study
In summary, although existing research and the-
ory suggest that engagement with Disney Princesses
may potentially influence the development of
children’s gender stereotypes, body esteem, and
prosocial behavior, research is scarce concerning
such effects. Given the ubiquitous nature of Disney
Princess media and toys, many have suggested that
this specific brand of media socialization will by
itself affect young children’s developing conceptions
of the social world and their place in it. The current
study examines this proposal for the first time, using
a multimethod approach in a sample of preschool
and kindergarten children. An additional strength of
this study is that the associations between variables
are considered both concurrent and over time, in
order to examine both short- and long-term effects.
We offer the following hypotheses, along with an
exploratory research question:
Hypothesis 1: On the basis of previous research and
the social cognitive theory of gender
development, we predict that expo-
sure to Disney Princesses will be
associated with heightened adher-
ence to female gender stereotypes
both concurrently and over time. We
also predict associations will be
stronger for girls than for boys.
Hypothesis 2: According to research on thin-ideal
internalization in media, we would
expect that exposure to Disney Prin-
cesses would be related to poorer
body esteem in girls, both in the
short term and long term. However,
given that research suggests that
body esteem socialization effects are
either nonexistent or weak in chil-
dren under the age of 6 years old, we
would predict rather modest effects.
Given that children are particularly
influenced by models of the same
sex, we would expect any associa-
tions to be stronger among girls than
among boys.
Hypothesis 3: Given that prosocial behavior is
extremely common in Disney
movies, we predict that engage-
ment with Disney Princesses will
be associated with heightened pro-
social behavior both concurrently
and over time. Again, we predict
effects to be stronger among girls
than boys given girls’potentially
higher identification with same-
gender princess characters.
Disney Princesses and Child Development 5
Research Question 1 RQ1 In order to clarify pre-
vious research, we will
explore the role of active
mediation in altering the
long-term relation between
Disney Princess engagement
and gender stereotyping,
body esteem, and prosocial
behavior.
Method
Participants
Participants consisted of 198 children (47%
male, M
age
=58 months, SD =7.52 at Time 1, age
range 36–78 months) and their parents (97.5%
were maternal report, M
age
=33.50 years,
SD =5.39) who participated in a larger study on
children and media. Children were recruited from
preschools and kindergartens at four different
sites. Two were in a midsize city in the midwes-
tern United States, one of which was at the uni-
versity, one at a Headstart school. The other two
were from a smaller city in the Pacific Northwes-
tern United States, again one at the university
school, the other at a community school. Participa-
tion rates at all schools exceeded 70%. Time 2 con-
sisted of only parent reports and took place
approximately 1 year after the initial data collec-
tion. There was an 82.5% retention rate from Time
1(n=240) to Time 2 (n=198). Only participants
who completed both parts of the study are
included in the following analyses. Testing took
place between 2012 and 2014.
For ethnicity, approximately 87% of participants
were Caucasian, 10% were Hispanic, and 3% were
another ethnicity. In terms of relationship status,
86% of parents were married, with 6% divorced,
and 8% in other family circumstances. Parental
education ranged fairly substantively, with 38%
not holding a college degree, 41% holding a bache-
lor’s degree, and 21% holding a higher degree
(e.g., MSc or PhD). Approximately 36% of parent
respondents (97.5% mothers) worked outside the
home. For income, 29% of families earned an
annual income less than $30,000, 13% earned
between $30,000 and $50,000, 26% earned between
$50,000 and $80,000, and 33% earned more than
$80,000. Child age, parent education level, and
family income were used as controls in the final
structural equation modeling (SEM) models we
describe below.
Materials/Procedures
Parents completed all questionnaires (at both
time points) at home, either online or on paper.
Children were tested during normal school hours
in a quiet, separate room. Teachers completed
questionnaires outside of school hours. Child and
teacher data were only obtained at Time 1. Multi-
ple informants were used where possible on the
variables described below as one way to reduce
shared method variance in analyses (e.g., Orth,
2013).
Disney Princess Engagement
Parents reported on three aspects of their child’s
engagement with Disney Princesses: identification,
toys, and media (each was measured with one
question). For identification, they were first shown a
picture of all the Disney Princesses as specified by
the Walt Disney Company. They were then asked
to choose the princess that their child most identi-
fied with and then rate how much their child iden-
tified with that princess on a 7-point Likert-type
scale (1 =not at all to 7 =highly identifies with). For
toys, parents were asked how frequently their child
plays with toys relating to any Disney Princess.
This measure was rated on a 7-point Likert-type
scale (1 =less than once a month to 7 =6 or more
times per week). For media, parents reported how fre-
quently their child viewed television shows or
movies (including DVDs) portraying Disney Prin-
cesses, also measured on a 7-point Likert-type scale
(1 =never to 7 =2–3 times per week). Cronbach’s
alpha for the combined measure was acceptable,
a=.78.
Media Time
Consistent with prior studies assessing media
effects, overall time spent viewing television was
used as a control variable in this study. It was mea-
sured by a single item that asked parents to rate
how frequently their child viewed television on an
average day. This measure was rated using a 6-
point Likert-type scale (1 =none to 6 =more than
3hr; Vandewater et al., 2007).
Children’s Gender-Stereotypical Behavior
Children’s male and female gender-stereotypical
behavior at Time 1 consisted of two separate latent
variables consisting of a child behavioral task (toy
6 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
preference task conducted with each child), parent
reports of their child’s gender-stereotypical toys,
activities, and characteristics and a teacher report of
gender-stereotypical behavior of the same child.
Parent reports only were used at Time 2 as the chil-
dren had scattered to many different schools fol-
lowing preschool. Higher scores indicate stronger
adherence to gender-stereotypical behavior.
Toy preference task (child observation). Children
were engaged in a toy preference task to measure
gendered toy preferences (Spence & Helmreich,
1978). Children were given a number of toys and
were asked to sort them into boxes of which toys
they like to play with “a lot,”“a little,”or “not at
all,”which were given 3, 2, and 1 points, respec-
tively, in the scoring system. Four toys were female
gender-stereotypical (e.g., doll, tea set), four were
male gender-stereotypical (e.g., action figure, tool
set), and four were gender neutral (e.g., puzzle,
paint set). Gender-neutral toys were not used in
analyses below. Female and male toy preference
scores were computed by averaging the preference
scores for the four female and the four male stereo-
typical toys. Cronbach’s alpha was acceptable for
both female stereotypical and male stereotypical toy
preferences (as=.91 and .79, respectively).
Parent report. Parents completed a modified ver-
sion of the Preschool Activities Inventory (Golom-
bok & Rust, 1993), which measures gender-
stereotypical behavior in preschool children. Several
items were modified to be more specific (e.g.,
“playing with girls”was changed to “playing with
girls [other than siblings]”). All items were mea-
sured on a 5-point Likert-type scale from 1 =never
to 5 =very often. Separate scales were computed for
male and female gender-stereotypical behavior.
There were three subscales, which, in the main ana-
lyses, are combined with child observations and
teacher reports (see below) into a latent variable.
Gender-stereotypical toy preference was assessed with
a seven-item measure that asked parents to report
how frequently their child played with gender-
stereotypical toys in the previous month. (e.g., male
items: toy guns, tool set; female items: dolls, tea
set). Cronbach’s alpha for this subscale was accep-
table (male items: a=.75; female items, a=.85).
Gender-stereotypical activities were a 12-item scale
that asked parents to rate how frequently their
child engaged in a number of activities during the
past month (e.g., male activities: “sports and ball
games,”“climbing [fences, trees, gym equipment,
etc.]”; female activities: “playing house [e.g., clean-
ing, cooking],”“playing dress up”). Cronbach’s
alpha for this subscale was acceptable (male
activities: a=.75; female activities: a=.82). Gender-
stereotypical characteristics were a seven-item scale
that asked parents to rate how often their child
showed a number of stereotypical characteristics
(e.g., male characteristics: “enjoys rough and tumble
play”; female characteristics: “likes pretty things”).
Cronbach’s alpha for this subscale was moderate
(male characteristics, a=.67; female characteristics,
a=.50).
Teacher report. Teachers also completed a mea-
sure of gender-stereotypical behavior for each indi-
vidual child, loosely based on the Preschool
Activities Inventory (Golombok & Rust, 1993). They
were asked to read six characteristics and rate how
often each child in the study exhibited these beha-
viors during school. Items were rated on a 5-point
Likert-type scale from 1 =never to 5 =almost always.
Again, female stereotypical items (e.g., “engages in
quiet play with other children”) were averaged
together as one scale, whereas male items (e.g.,
“engages in rough and tumble play with children”)
were averaged into another scale. This measure has
some overlap with the parent’s questionnaire but
was substantially shorter. Cronbach’s alpha for this
scale was acceptable (male stereotypical behavior:
a=.89; female stereotypical behavior: a=.79).
Body Esteem
Child body esteem. Parents reported on the per-
ceived body esteem of children using a five-item
questionnaire created for the purposes of the study,
though loosely based on adult body esteem mea-
sures (Mendelson, Mendelson, & White, 2001). Par-
ents were asked to read each statement (“my child
likes his/her body,”“my child would like to be
thinner,”“my child talks about his/her weight
often,”“my child wishes he/she were better look-
ing,”“my child feels good about the way he/she
looks”) and then rate how much they agreed with
each one using a 5-point Likert-type scale
(1 =completely disagree to 5 =completely agree).
Higher scores indicate a better body esteem. Cron-
bach’s alpha was .76.
Parent body esteem. Parental body esteem was
used as a control variable in the current study as it
tends to be a good predictor of child body esteem
(Guiney & Furlong, 1999). This measure was rated
using a 14-item scale: Body Esteem Scale for Ado-
lescents and Adults–MODIFIED (Mendelson et al.,
2001). Participants were asked to indicate how
much they agreed with a number of statements
(e.g., “I like what I look like in pictures”) using a 5-
point Likert-type scale (1 =never to 5 =always).
Disney Princesses and Child Development 7
Higher scores indicate a better body esteem. Cron-
bach’s alpha was .90.
Prosocial Behavior
Prosocial behavior was constructed as a latent
variable consisting of both parents and teacher
reports of prosocial behavior at Time 1. Parent
report only was used at Time 2.
Parent report. Parents completed a four-item
questionnaire from the Parent Adaptation of the
Preschool Social Behavior Survey that asked them
to report how frequently their child engages in pro-
social behavior (e.g., “your child is helpful to
peers”; Crick, Casas, & Mosher, 1997; Crick, Casas,
& Ku, 1999). Items were rated using a 5-point
Likert-type scale (1 =never or almost never true to
5=always or almost always true). Higher scores indi-
cate higher levels of prosocial behavior. Cronbach’s
alpha was .80.
Teacher report. Teachers also rated each child on
their prosocial behavior, using a four-item measure
from the Social Skills Questionnaire (Evans, 2009).
Teachers were asked to rate how frequently each
child engaged in a number of prosocial behaviors
(e.g., “offers to help other children who are having
difficulty with a task in the classroom”) using a 5-
point Likert-type scale (1 =never to 5 =almost
always). Higher scores indicate higher levels of pro-
social behavior. Cronbach’s alpha was .91.
Parental Active Mediation
Parental mediation of media was measured using
three items (Padilla-Walker & Coyne, 2011) on a 5-
point Likert-type scale (1 =never to 5 =very often)
that asked parents how frequently they discussed
media content with their children (e.g., “How often
do you try to help your child understand what he/
she sees in the media?”). The measure was a gen-
eral parental mediation of media scale and was not
specific to Disney Princesses. Cronbach’s alpha was
acceptable (a=.88).
Results
Preliminary Analyses
Gender Differences
We first examined gender differences in all main
variables at Time 1. For Disney Princess engagement,
22% of girls and 8% of boys viewed Disney Princess
media at least once a week, with 50% of girls and
29% of boys viewing such media at least once a
month. Only 4% of girls and 13% of boys reported
never viewing Disney Princess media. In addition,
over 61% of girls played with Disney Princess toys at
least once a week, compared with only 4% of boys
(indeed, 90% of boys rarely, if ever, played with such
toys, according to their parents).
Table 1
Gender Differences in All Main Study Variables at Time 1
Boys Girls
Fvalue Partial g
2
MSDMSD
Princess identification (PR) 1.30 0.67 3.69 1.64 113.95*** .46
Princess toys (PR) 1.23 0.82 4.00 1.97 104.55*** .44
Princess media (PR) 3.41 1.40 4.47 1.54 17.28*** .11
Female gender-stereotypical behaviors (PR) 2.32 0.44 3.52 0.48 225.17*** .63
Male gender-stereotypical behaviors (PR) 3.57 0.56 2.77 0.50 74.66*** .36
Female gender-stereotypical behaviors (TR) 2.48 0.59 4.08 0.53 281.85*** .68
Male gender-stereotypical behavior (TR) 3.81 0.70 2.05 0.66 224.01*** .63
Female gendered toy preference (CO) 1.58 0.64 2.89 0.21 273.06*** .67
Male gendered toy preference (CO) 2.68 0.38 1.67 0.57 137.01*** .51
Prosocial behavior (PR) 3.85 0.49 4.11 0.46 10.08** .07
Prosocial behavior (TR) 2.79 0.89 3.38 0.93 13.83*** .09
Body esteem (PR) 4.70 0.37 4.59 0.48 1.93 .01
Media time (PR) 3.15 1.30 3.31 1.33 0.49 .01
Active mediation (PR) 3.09 0.54 3.16 .53 0.66 .01
Note. PR =parent report; TR =teacher report; CO =child observation.
**p<.01. ***p<.001.
8 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
Table 1 shows means and standard deviations
by gender for all variables. For simplicity, in all
remaining preliminary analyses, we averaged the
three subscales of parent-reported male and female
gender-stereotypical behavior (to indicate an aggre-
gate view of parent perceptions of child male and
female gender-stereotypical behavior). These sepa-
rate items/subscales are later used as observed
indicators of latent variables in the main analyses.
A multivariate analysis of variance (MANOVA)
revealed an overall significant effect for gender on
most variables, F(14, 121) =66.58, p<.001, partial
g
2
=.89 (see Table 1). Girls had significantly higher
levels of princess identification, playing with prin-
cess toys, and viewing more princess media than
boys. Girls also showed higher levels of female gen-
der-stereotypical behaviors than boys as rated by
parents, teachers, and observers in the child obser-
vation task. Conversely, boys showed higher levels
of male gender-stereotypical behaviors than girls as
rated by all three informants. Girls also had higher
levels of prosocial behavior as rated by teachers
and parents. Finally, there were no gender differ-
ences on body esteem, overall media time, and
active mediation. We also conducted similar ana-
lyses for Time 2. The pattern of gender differences
did not differ, and analyses can be obtained from
the first author on request.
Ethnicity
A MANOVA was also conducted to examine
any differences in Hispanic and non-Hispanic chil-
dren on any princess variable or any outcome at
both time points. A significant multivariate effect
was found, F(16, 129) =1.94, p<.05, partial
g
2
=.19. Hispanic children viewed more Disney
Princess media than non-Hispanic children at both
time points (ps<.05), though there were no signifi-
cant differences on any other variable. Accordingly,
ethnicity will be controlled for in all major analyses.
Missing Data
A MANOVA was conducted comparing those
who dropped out of the study after Time 1 with
those that completed both waves of the study, with
no significant differences on any outcome or predic-
tor variable being found, F(8, 199) =1.51, p>.05.
Correlations
Bivariate correlations for all study variables are
shown in Table 2. Although this is not included in
Table 2
Bivariate Correlations for Main Variables at Time 1, Including Control Variables
1234567891011121314
1. Princess identification (PR) —.45*** .31** .30** .07 .01 .16 .13 .02 .21* .01 .06 .23* .23*
2. Princess toys (PR) .22
†
—.49*** .57*** .01 .11 .19
†
.21* .10 .10 .17 .10 .23* .01
3. Princess media (PR) .23* .45*** —.28** .03 .11 .02 .05 .11 .08 .07 .04 .26** .08
4. Female gender-stereotypical behaviors (PR) .23* .25* .15 —.09 .26** .14 .29** .03 .16 .18
†
.02 .03 .11
5. Male gender-stereotypical behaviors (PR) .12 .02 .18
†
.38** —.10 .11 .10 .36** .08 .09 .09 .03 .02
6. Female gender-stereotypical behaviors (TR) .04 .16 .16 .33** .05 —.52*** .04 .03 .13 .03 .05 .13 .01
7. Male gender-stereotypical behavior (TR) .07 .14 .01 .16 .01 .57*** —.11 .05 .04 .25* .05 .01 .03
8. Female gendered toy preference (CO) .26* .25* .27* .29** .12 .08 .05 —.16 .04 .15 .13 .10 .08
9. Male gendered toy preference (CO) .18 .12 .01 .15 .21* .07 .02 .14 —.01 .14 .10 .04 .03
10. Prosocial behavior (PR) .21* .04 .08 .22* .24* .21
†
.09 .21
†
.23* —.38*** .07 .14 .05
11. Prosocial behavior (TR) .01 .16 .01 .35*** .08 .31** .06 .09 .27* .32** —.01 .01 .18
†
12. Body esteem (PR) .03 .07 .11 .04 .02 .09 .01 .14 .21
†
.25* .03 —.13 .08
13. Media time (PR) .12 .16 .16 .05 .06 .18
†
.06 .01 .12 .05 .12 .08 —.01
14. Active mediation (PR) .05 .06 .08 .19
†
.18 .10 .18 .07 .15 .12 .08 .06 .01 —
Note. Correlations for girls are above the diagonal, boys below. PR =parent report; TR =teacher report; CO =child observation.
†
p<.10. *p<.05. **p<.01. ***p<.001.
Disney Princesses and Child Development 9
the table, child body esteem and parental body
esteem were moderately positively correlated,
r=.20, p<.01.
Main Analyses
Measurement Model
In prelude to the structural models tested below, a
measurement model was first examined using SEM
via the Analysis of Moments Structure Software (ver-
sion 23; IBM Corporation, Armonk, New York,
USA). There were five constructs (shown in Table 3):
princess engagement (three items), male gender-
stereotypical behavior (five scales), female gender-
stereotypical behavior (five scales), prosocial beha-
vior (two scales), and body esteem (one scale). Good-
ness-of-fit indices suggested that the model
adequately represented the data, v
2
(170) =208.44,
p=.05, comparative fit index (CFI) =.92, root mean
square error of approximation (RMSEA) =.03.
A multigroup model was examined comparing boys
and girls. When the model was constrained by
gender, the fit significantly decreased (p<.01). How-
ever, partial metric invariance was achieved by free-
ing up one factor loading (princess toys), resulting in
no significant difference between the unconstrained
and constrained models, v2
differenceð10Þ¼18:00,
p>.05.
Structural Model
Cross-Sectional Results
The first model examined cross-sectional associa-
tions between engagement with Disney Princesses,
gender-stereotypical behavior, body esteem, and
prosocial behavior. The structural model used the
maximum likelihood data estimation method to
account for missing data. The analysis modeled
Disney Princess media engagement (consisting of a
latent variable of watching princess media, identify-
ing with princesses, and playing with princess toys)
on male and female gender-stereotypical behavior
(latent variable composed of parent and teacher
report, as well as child observation), body esteem
(parent report), and prosocial behavior (latent vari-
able composed of parent and teacher report). The
parent’s own body esteem, the child’s overall time
spent viewing media, child age, parental education,
ethnicity, and parental income were used as con-
trols. The figure is not included due to overlap with
the longitudinal results.
A multiple group analysis was conducted to
examine potential gender differences, comparing a
fully constrained model to a fully unconstrained
model, which resulted in a decrease in model fit,
v2
differenceð55Þ¼482:04, p<.001. Analyses suggested
that model fit was best when intercepts and struc-
tural paths were allowed to vary across groups.
The final model resulted in adequate fit,
v2
differenceð281Þ¼347:26, p<.01; CFI =.90,
RMSEA =.04. For both boys (b=.77, p<.01) and
girls (b=.72, p<.001), princess engagement was
associated with higher levels of female gender-
stereotypical behavior but not male gender-stereo-
typical behavior, prosocial behavior, or body image
concurrently.
Longitudinal Results
We also examined longitudinal results of prin-
cess engagement on child outcomes (as reported by
parents at Time 2) using a cross-lagged structural
equation model. In these analyses, princess
Table 3
Standardized Factor Loadings for the Measurement Model of Princess
Engagement, Gender-Stereotypical Behavior (Male and Female), and
Prosocial Behavior at Time 1
Factor Items/scale Loadings
Princess
engagement
Princess identification (PR) .76
Play with princess toys (PR) .91
Watch Princess media (PR) .59
Male gender-
stereotypical
behavior
Gender-stereotypical toy
preference (PR)
.73
Gender-stereotypical
activities (PR)
.79
Gender-stereotypical
characteristics (PR)
.42
Gender-stereotypical toy
preference task (CO)
.73
Gender-stereotypical
behaviors (TR)
.73
Female gender
stereotypical
behavior
Gender-stereotypical toy
preference (PR)
.93
Gender-stereotypical
activities (PR)
.80
Gender-stereotypical
characteristics (PR)
.52
Gender-stereotypical toy
preference task (CO)
.82
Gender-stereotypical
behaviors (TR)
.79
Prosocial
behavior
Prosocial behavior (PR) .52
Prosocial behavior (TR) .71
Note. PR =parent report; TR =teacher report; CO =child
observation.
10 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
engagement at Time 1 (again, a latent variable of
watching princess media, identifying with prin-
cesses, and playing with princess toys) was mod-
eled to predict both male and female gender-
stereotypical behavior (latent variable of parents’
reports on gender-stereotypical activities, toys, and
characteristics), prosocial behavior, body esteem,
and princess engagement at Time 2. All previous
control variables and all outcome variables at Time
1 were used as controls. We did not have the
power to use the full latent variables as modeled in
the cross-sectional model at Time 1. However, to
increase validity, we have used an aggregate
measure of teacher and parent reports for the gen-
der stereotyping and prosocial measures as control
variables at Time 1. All other variables in the model
are parent report.
Similar to the first model, a multiple group
analysis of gender was conducted that also
resulted in a substantial decrease in model fit,
v2
differenceð56Þ¼212:62, p<.001, suggesting that we
should allow intercepts and structural paths to vary
across child gender. The final model showed ade-
quate fit, v2
differenceð281Þ¼299:74, p<.001; CFI =.90,
RMSEA =.04. Autoregressive paths were highly
stable over time (see Figure 1; see Table S1, for all
the coefficient paths for the model). For both girls
(b=.65, p<.001) and boys (b=.41, p=.07), prin-
cess engagement at Time 1 was associated with
higher female gender-stereotypical behavior at Time
2, even after controlling for initial levels of gender-
stereotypical behavior. Additionally, body esteem
(b=.14, p≤.05) at Time 1 negatively predicted
princess engagement at Time 2 for girls only. Proso-
cial behavior was not longitudinally associated with
princess engagement.
Parental Mediation
Next, parental mediation was examined as a
potential moderator for long-term effects on both
male and female gender-stereotypical behavior,
body esteem, and prosocial behavior. For male and
female gender-stereotypical behavior, the three par-
ent measures (activities, toys, and characteristics)
were averaged to produce overall levels of adher-
ence to male and female stereotypes. The first anal-
ysis examined the effect of princess engagement on
female gender-stereotypical behavior depending on
both the child’s gender and the frequency of active
mediation each gender received from parents. The
moderated moderation model was estimated via
OLS regression analysis where the effect of
princess engagement on female gender-stereotypical
behavior was estimated as a three-way interaction
between gender, active mediation, and princess
engagement using Hayes’(2013) PROCESS SPSS
macro (Model 3). As part of this analysis, PROCESS
generated conditional effects of princess engage-
ment on female gender-stereotypical behavior for
various values of gender (boys and girls) and active
mediation. Child age, parent income, parent educa-
tion, TV/media time, and female gender-stereotypi-
cal behavior at Time 1 were entered into the model
as covariates. Results showed that princess engage-
ment was positively associated with female gender-
stereotypical behavior at Time 2 for both girls
(b=.104, t=2.41, p<.05, 95% CI [0.0186, 0.1890])
and boys (b=.229, t=2.64, p<.01, 95% CI
[0.0572, 0.4009]) at high amounts of active media-
tion. A similar analysis was conducted with male
gender-stereotypical behavior at Time 2 as the
dependent variable (controlling for male gender-
stereotypical behavior at Time 1, along with the
other covariates). Results revealed a nonsignificant
three-way interaction between princess engagement,
active mediation, and gender on male gender-
stereotypical behavior at Time 2 (b=.006, t=.05,
p>.05, 95% CI [0.2210, 0.2322]).
Next, the same moderated moderation OLS
regression model with prosocial behaviors at Time
2 (controlling for prosocial behaviors at Time 1 and
other control variables) and active mediation and
gender as moderators revealed that princess
engagement predicted prosocial behaviors at Time
2 for boys (b=.210, t=2.16, p<.05, 95% CI
[0.0173, 0.4024]) at high levels of active mediation.
Finally, the same regression model with body
esteem at Time 2 (controlling for body esteem at
Time 1 and other control variables) and active
mediation and gender as moderators revealed that
princess engagement predicted higher body esteem
at Time 2 for boys at high levels of active mediation
(b=.251, t=2.44, p<.05, 95% CI [0.0471, 0.4547]).
Discussion
In the current study, we examined the relation
between Disney princess engagement and gender-
stereotypical behavior, body esteem, and prosocial
behavior during early childhood. Girls were much
more likely than boys to engage with Disney Prin-
cesses through viewing media, playing with toys,
and identifying with princesses. Disney Princess
engagement was also extremely stable across a 1-
year period, for both boys and girls. Our results
revealed that princess engagement was
Disney Princesses and Child Development 11
Figure 1. Longitudinal model of associations.
Note. Standardized values are shown. Findings for girls are presented first and boys second. For model simplicity, a number of items
are not included in the figure though they are in the original model, including error terms, correlations between exogenous variables
and control variables, covariances between error terms, and path weights between control and endogenous variables. If desired, please
contact the first author for complete model statistics. pt =parent report of toys; pa =parent report of activities; pc =parent report of
characteristics.
†
p<.10. *p<.05. **p<.01. ***p<.001.
12 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
concurrently related to higher levels of female
gender-stereotypical behavior for both boys and
girls. Furthermore, higher princess engagement was
associated with increased female gender-stereotypi-
cal behavior for both girls and boys (trend level)
1 year later, even after controlling for initial levels
of gender-stereotypical behavior. There were no
effects on male gender-stereotypical behavior, either
concurrently or longitudinally for boys or girls.
Based on previous research and theories of
media effects, the first hypothesis was that engage-
ment with Disney Princesses would be associated
with higher levels of female gender-stereotypical
behavior both concurrently and in the long term.
For girls, the results supported this hypothesis and
are consistent with previous research, theory, and
meta-analyses showing that increased exposure to
gender stereotyped media is related to more stereo-
typical behavior (e.g., Hust & Brown, 2008). It also
supports research showing that engagement with
the Disney Princess culture can influence gender
stereotypes and may contribute to a “girly girl”cul-
ture in which gendered behavior is common and
highly valued (Dinella, 2013; Orenstein, 2011; Wohl-
wend, 2009). Importantly, our model showed
evidence for a socializing effect between princess
engagement and adherence to gender stereotypes,
as higher levels of female gender-stereotypical
behavior was not related to later princess engage-
ment (after controlling for earlier princess engage-
ment) for girls or boys.
High levels of stereotypical female behavior have
potentially mixed implications for parents of young
girls. Although there is nothing inherently wrong
with expressing femininity or behaving in a gen-
dered manner, stereotypical female behavior may
potentially be problematic if girls believe that their
opportunities in life are limited because of precon-
ceived notions regarding gender or if they avoid
the types of exploration and activities that are
important to children learning about the world in
order to conform to stereotypical notions about
femininity (e.g., choosing not to explore or play cer-
tain games in order to avoid getting dirty; Huguet
&R
egner, 2007). Dinella (2013) also found that
grown women who self-identified as “princesses”
gave up more easily on a challenging task, were
less likely to want to work, and were more focused
on superficial qualities. These findings suggest that
the development of high levels of female gender-
stereotypical behavior during early childhood may
have implications for later development.
Notably, the current study only examined the
effect of engagement with the princess culture
across a 1-year time period. Future research should
examine the cumulative effect of exposure to
Disney Princesses from a young age on gender
development in both adolescence and adulthood.
Additionally, although preschool and kindergarten
are important developmental periods for examining
gender development, conceptions of gender begin
earlier than the preschool age. Accordingly, it is
possible that preference for gendered toys or activi-
ties may emerge prior to engagement with Disney
Princesses. This may be particularly true regarding
an understanding and preference for gender-stereo-
typical toys, which may emerge during infancy
(Zosuls et al., 2009). We did not examine very
young children in the current study and future lon-
gitudinal research may wish to examine these asso-
ciations in a younger group of participants.
Princess engagement was also associated with
higher levels of female gender-stereotypical behav-
ior among boys. Given concerns about hypermascu-
line messages in the media (e.g., Brown, Lamb, &
Tappan, 2009), especially in programs popular
among preschooler boys, such as superhero pro-
grams (e.g., Baker & Raney, 2007), this finding is of
particular interest. Princess media and engagement
may provide important models of femininity to
young boys, who are typically exposed to hyper-
masculine media. It may be that boys who engage
more with Disney Princesses, while simultaneously
being exposed to more androgynous Disney
princes, demonstrate more androgyny in early
childhood, a trait that has benefits for development
throughout the life span (e.g., Roos & Cohen, 1987).
Future research should investigate to what degree
princess engagement is associated with androgyny
in young boys as well as the change in this influ-
ence over time.
One other aim of the study was to explore
whether parental active mediation would moderate
the effects of princess engagement on young chil-
dren. Specifically, the current study examined active
mediation, which includes discussing media content
with children. Results showed that princess engage-
ment predicted higher female stereotypical behav-
iors for both girls and boys at high levels of active
mediation. In other words, princess engagement
only longitudinally predicted adherence to female
gender stereotypes when parents talked to their chil-
dren about the media. Typically, we would expect
that discussing media content would lead to lower
levels of gender-stereotypical behavior (Nathanson
et al., 2002). However, we did not specifically ask
questions regarding the messages parents were dis-
cussing. Additionally, we do not have information
Disney Princesses and Child Development 13
on the extent to which parents were engaging with
their children in play with princess products, which
may serve as a type of mediation. Indeed, research
reveals that parents feel Disney Princesses are rela-
tively “safe”and many parents may actively
promote the stereotypical behavior their children
view in the films. Accordingly, socializing messages
may be strongest when parental and media mes-
sages support each other (Bussey & Bandura, 1999).
Again, we did not ask detailed questions regarding
the type of active mediation, so future research
should examine this in depth.
The second hypothesis was that engagement
with Disney Princesses would be related to poor
body esteem in girls, due to the perpetuation of the
thin ideal by the Disney princess characters. How-
ever, princess engagement was not associated with
concurrent body esteem for either boys or girls.
This finding is somewhat surprising, given research
on thin-ideal internalization that suggests that expo-
sure to thin role models in media and playing with
thin dolls would increase the likelihood of poor
body esteem in childhood (e.g., Anschutz & Engels,
2010; Grabe et al., 2008; Want, 2009). This result
may be a function of the age of the sample.
Although body esteem concerns can emerge during
the preschool years (Dohnt & Tiggemann, 2006),
most children generally have very positive body
esteem at this age (as was the case in our study).
Indeed, our findings do confirm research showing
there was no immediate effect on appearance-
related play in young children after viewing a short
clip from a Disney princess film (Hayes & Tantleff-
Dunn, 2010). When examining these associations
longitudinally, there was no effect on Disney Prin-
cess engagement on body esteem of girls 1 year
later (either positive or negative), perhaps because
body esteem declined overall. However, girls with
lower body image at Time 1 had higher levels of
princess engagement 1 year later, suggesting they
may be seeking out princesses as appearance-
related role models. Accordingly, associations
between princess engagement and body esteem
may be driven by preexisting body concerns. We
hope that research examines the long-term effect of
princess engagement on body esteem. It may be
that early princess engagement would predict poor
body esteem later in childhood or adolescence, as
Orenstein (2011) suggests. Prospective longitudinal
studies from preschool through adolescence would
be useful for investigating this possibility.
Additionally, for boys only, engagement with
Disney Princesses at Time 1 was associated with
better body esteem at Time 2, again only for those
with high levels of active mediation. This finding is
a difficult one to explain, but we offer one tentative
explanation. It may be that boys with higher Disney
Princess engagement are not only viewing prin-
cesses but also princes. Disney princes tend to por-
tray the muscular ideal, which is associated with
poor body esteem in boys and men (Barlett, Vow-
els, & Saucier, 2008). If parents are discussing that
the muscular ideal portrayed in the media is unre-
alistic and unattainable, this may positively influ-
ence boys’feelings about their own bodies. Our
explanation is pure speculation as we did not mea-
sure exposure to Disney princes in boys or how
parents specifically discuss the muscular ideal with
their young children; however, it raises some poten-
tial questions for future research to pursue.
The third hypothesis was that Disney Princess
engagement would be associated with higher levels
of prosocial behavior, both concurrently and longi-
tudinally. Results indicated that princess engage-
ment was not associated with higher levels of
prosocial behavior for girls. Disney movies contain
extremely high levels of prosocial behavior (Padilla-
Walker et al., 2013), and the princesses show many
prosocial traits, including being self-sacrificing,
kind, and defending those in trouble. However,
girls do not appear to be picking up on these
themes, at least in the context of other factors mea-
sured in the study. Conversely, princess engage-
ment predicted prosocial behaviors for boys at
Time 2, but only for boys at high levels of active
mediation. In other words, high amounts of active
mediation encourage boys, but not girls, to adopt
the prosocial behaviors portrayed by Disney Prin-
cesses. These results appear to coincide with prior
mediation research suggesting that active mediation
can encourage the learning of positive media mes-
sages. Perhaps young girls are already so involved
and immersed in the Disney Princess culture that
no amount of active mediation can further encour-
age their mimicry of prosocial behaviors. For boys,
it is possible that they are less involved in the Dis-
ney Princess culture so that high amounts of active
mediation can encourage them to process princess
content in a way that they learn the portrayed
prosocial behaviors.
Although the current study had many strengths,
including the longitudinal design, multiple out-
comes, multiple data collection sites, and multiple
reporters, there are a few limitations we would like
to note. First, the sample is somewhat limited as
many of the children came from White, middle-
class families; accordingly, future research should
examine these associations with diverse
14 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
populations. Second, at Time 2, all child outcome
variables were parent report. The reason for this is
because most of the children went to multiple dif-
ferent schools all over the county after Time 1.
Accordingly, it was not feasible to work with the
amount of schools required to obtain teacher data
at Time 2. Furthermore, the toy preference task
(parent report) may measure availability as well as
toy preference, as children would need to own the
toys to play with them. Additionally, future
research could examine both the context and con-
tent of children’s play with Disney Princess toys,
which may consist of many themes including
romance, beauty, prosocial behavior, and so on. We
also did not measure length of time enrolled at the
school. It may be that some teachers were more
reliable reporters on behavior for students poten-
tially enrolled at the school in earlier years. Addi-
tionally, although this study showed long-term
effects of princess engagement, children were only
tested 1-year apart. Future longitudinal research
should examine effects of princess engagement
across childhood and adolescence, as well as
attempt to disentangle the effects of media viewing
from product engagement, such as play with
media-related toys. Finally, the parental mediation
measure was general and did not specifically ask
about mediation regarding Disney Princesses or our
specific outcomes. Accordingly, we do not know
whether parents discuss Disney Princesses at all
and in what context.
Despite these limitations, this is the first study, to
our knowledge, to show a long-term effect of Disney
Princess engagement during early childhood. This
study shows that engagement with Disney Prin-
cesses can be limiting, as young girls especially are
more likely to embrace traditional female stereo-
types both concurrently and longitudinally. How-
ever, there were also some potential positive
benefits for boys, including better body esteem and
higher levels of prosocial behavior when parents
discussed the media with their children. Most child
developmentalists hope that children end up living
their own “happily ever after”; and greater under-
standing of the “princess culture”in early childhood
may, in a small part, help them on their way.
References
Anschutz, D., & Engels, R. (2010). The effects of playing
with thin dolls on body image and food intake in
young girls. Sex Roles,63, 621–630. doi:10.1007/s11199-
010-9871-6
Baker, K., & Raney, A. A. (2007). Equally super?: Gender-
role stereotyping of superheroes in children’s animated
programs. Mass Communication & Society,10,25–41.
doi:10.1080/15205430709337003
Bandura, A. (2001). Social cognitive theory: An agentic
perspective. Annual Review of Psychology,52,1–6.
doi:10.1146/annurev.psych.52.1.1
Barlett, C. P., Vowels, C. L., & Saucier, D. A. (2008).
Meta-analyses of the effects of media images on men’s
body-image concerns. Journal of Social and Clinical Psy-
chology,27, 279–310. doi:10.1521/jscp.2008.27.3.279
Best, J., & Lowney, K. S. (2009). The disadvantage of a
good reputation: Disney as a target for social problems
claims. The Sociological Quarterly,50, 431–449.
doi:10.1111/j.1533-8525.2009.01147.x
Brown, L. M., Lamb, S., & Tappan, M. (2009). Packaging
boyhood: Saving our sons from superheroes, slackers, and
other media stereotypes. New York, NY: St. Martin’s Press.
Buckley, K. E., & Anderson, C. A. (2006). A theoretical
model of the effects and consequences of playing video
games. In P. Vorderer & J. Bryant (Eds.), Playing video
games—Motives, responses, and consequences (pp. 363–
378). Mahwah, NJ: Erlbaum.
Burnett, F. H. (1905). A little princess. New York, NY:
Charles Scribner's Sons.
Bussey, K., & Bandura, A. (1999). Social cognitive theory
of gender development and differentiation. Psychological
Review,106, 676–713. doi:10.1037/0033-295X.106.4.676
Collins, R. L. (2011). Content analysis of gender roles in
media: Where are we now and where should we go?
Sex Roles,64, 290–298. doi:10.1007/s11199-010-9929-5
Corder-Bolz, C. (1980). Mediation: The role of significant
others. Journal of Communication,30, 106–118.
Coyne, S. M., Linder, J., Rasmussen, E. E., Nelson, D. A.,
& Collier, K. M. (2014). It’s a bird!It’s a plane!It’sa
gender stereotype!: Longitudinal associations between
superhero viewing and gender stereotyped play. Sex
Roles,70, 416–430. doi:10.1007/s11199-014-0374-8
Crick, N. R., Casas, J. F., & Ku, H. (1999). Relational and
physical forms of peer victimization in preschool.
Developmental Psychology,35, 376–385. doi:10.1037/
0012-1649.35.2.376
Crick, N. R., Casas, J. F., & Mosher, M. (1997). Relational
and overt aggression in preschool. Developmental Psy-
chology,33, 579–588. doi:10.1037/0012-1649.33.4.579
Dinella, L. (2013, April). Internalization of princess culture: A
cross-sectional study of early childhood and emerging adult-
hood. Paper presented at the Society for Research on
Child Development biannual conference, Seattle, WA.
Disney Consumer Products. (2011). Disney princess.
Retrieved from https://www.Disneyconsumerprod-
ucts.com/Home/display.jsp?contentId=dcp_home_our
franchises_disney_princess_us&forPrint=false&languag
e=en&preview=false&imageShow=0&pressRoom=US&
translationOf=nul®ion=0&first=0&last=
Do Rozar
ıa, R. A. C. (2004). The princess and the magic
kingdom: Beyond nostalgia, the function of the Disney
princess. Women’s Studies in Communication,27,1.
Disney Princesses and Child Development 15
Dohnt, H. K., & Tiggemann, M. (2006). Body esteem con-
cerns in young girls: The role of peers and media prior
to adolescence. Journal of Youth and Adolescence,35, 141–
151. doi:10.1007/s10964-005-9020-7
Ehrenreich, B. (2007). Bonfire of the Disney Princesses. The
Nation. Retrieved from http://www.thenation.com/
article/bonfire-disney-princesses
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
Evans, C. A. (2009). Making sense of their world: Sensory
reactivity and sensory awareness as predictors of social
interaction in early childhood. Dissertation Abstracts
International,69, 6450.
Fagot, B. I. (1977). Consequences of moderate cross-gen-
der behaviour in preschool children. Child Development,
48, 902–907.
Fagot, B. I., Leinbach, M. D., & O’Boyle, C. (1992). Gen-
der labeling, gender stereotyping, and parenting behav-
iors. Developmental Psychology,28, 225–230. doi:
10.1037/0012-1649.28.2.225
Gangestad, S. W., & Scheyd, G. J. (2005). The evolution of
human physical attractiveness. Annual Review of Anthro-
pology,34, 523–548. doi:10.1146/annurev.anthro.33.0702
03.143733
Golombok, S., & Rust, J. (1993). The Pre-School Activities
Inventory: A standardized assessment of gender role in
children. Psychological Assessment,5, 131–136. doi:
10.1037/1040-3590.5.2.131
Gomez, J. (2014). Why “Frozen”became the biggest animated
movie of all time. Retrieved from http://www.businessin-
sider.com/why-frozen-is-a-huge-success-2014-4
Goudreau, J. (2012). Disney princess tops list of the 20 best-
selling entertainment products. Retrieved from http://
www.forbes.com/sites/jennagoudreau/2012/09/17/di
sney-princess-tops-list-of-the-20-best-selling-entertainme
nt-products/
Grabe, S., Ward, L. M., & Hyde, J. S. (2008). The role of
the media in body esteem concerns among women: A
meta-analysis of experimental and correlational studies.
Psychological Bulletin,134, 460. doi:10.1037/0033-
2909.134.3.460
Guiney, K. M., & Furlong, N. E. (1999). Correlates of
body satisfaction and self-concept in third- and sixth-
graders. Current Psychology: A Journal for Diverse Per-
spectives on Diverse Psychological Issues,18, 353–367.
doi:10.1007/s12144-999-1009-z
Hargreaves, D. A., & Tiggemann, M. (2009). Muscular
ideal media images and men’s body image: Social com-
parison processing and individual vulnerability. Psy-
chology of Men & Masculinity,10, 109–119. doi:10.1037/
a0014691
Harrison, K. (2000). Television stereotyping, fat stereotyp-
ing, body shape standards, and eating disorder symp-
tomatology in grade school children. Communication
Research,27, 617–640.
Hayes, A. F. (2013). Introduction to mediation, moderation,
and conditional process analysis. New York, NY: Guilford.
Hayes, S., & Tantleff-Dunn, S. (2010). Am I too fat to be a
princess? Examining the effects of popular children’s
media on young girls’body esteem. British Journal of
Developmental Psychology,28, 413–426. doi:10.1348/
026151009X424240
Herrett-Skjellum, J., & Allen, M. (1996). Television pro-
gramming and sex stereotyping: A meta-analysis. In B.
R. Burleson & B. R. Burleson (Eds.), Communication
yearbook 19 (pp. 157–185). Thousand Oaks, CA: Sage.
Hogan, M. (2001). Parents and other adults: Models and
monitors of healthy media habits. In D. G. Singer & J.
L. Singer (Eds.), Handbook of children and the media (pp.
663–680). Thousand Oaks, CA: Sage.
Hohlstein, L. A., Smith, G. T., & Atlas, J. G. (1998). An
application of expectancy theory to eating disorders:
Development and validation of measures of eating and
dieting expectancies. Psychological Assessment,10,49–58.
doi:10.1037/1040-3590.10.1.49
Huguet, P., & R
egner, I. (2007). Stereotype threat among
schoolgirls in quasi-ordinary classroom circumstances.
Journal of Educational Psychology,99, 545. doi:10.1037/
0022-0663.99.3.545
Hust, S. T., & Brown, J. D. (2008). Gender, media use,
and effects. In S. L. Calvert & B. J. Wilson (Eds.), The
handbook of children, media, and development (pp. 98–120).
Oxford, UK; Malden, MA: Blackwell.
Lacroix, C. (2004). Images of animated others: The orien-
talization of Disney’s cartoon heroines from The
Little Mermaid to The Hunchback of Notre Dame. Popular
Communication,2, 213–229. doi:10.1207/s15405710pc
0204_2
Mares, M., & Woodard, E. (2005). Positive effects of tele-
vision on children’s social interactions: A meta-analysis.
Media Psychology,7, 301–322. doi:10.1207/S1532785XME
P0703_4
Mendelson, B. K., Mendelson, M. J., & White, D. R.
(2001). Body-esteem scale for adolescents and adults.
Journal of Personality Assessment,76,90–106.
doi:10.1207/S15327752JPA7601_6
Moriarty, C., & Harrison, K. (2008). Television exposure
and disordered eating among children: A longitudinal
panel study. Journal of Communication,58, 361–381.
Nathanson, A. I., & Botta, R. A. (2003). Shaping the
effects of television on adolescents’body image distur-
bance. Communication Research,30, 304–331.
Nathanson, A. I., Wilson, B. J., McGee, J., & Sebastian, M.
(2002). Counteracting the effects of female stereotypes
on television via active mediation. Journal of Communi-
cation,52, 922–937.
Orenstein, P. (2011). Cinderella ate my daughter. New York,
NY: HarperCollins.
Orth, U. (2013). How large are actor and partner effects
of personality on relationship satisfaction? The impor-
tance of controlling for shared method variance. Person-
ality and Social Psychology Bulletin,39, 1359–1372.
doi:10.1177/0146167213492429
Padilla-Walker, L. M., & Coyne, S. M. (2011). “Turn that
thing off!”Parent and adolescent predictors of
16 Coyne, Linder, Rasmussen, Nelson, and Birkbeck
proactive media monitoring. Journal of Adolescence,34,
705–715. doi:10.1016/j.adolescence.2010.09.002
Padilla-Walker, L. M., Coyne, S. M., Fraser, A. M., &
Stockdale, L. A. (2013). Is Disney the nicest place on
earth? A content analysis of prosocial behavior in ani-
mated Disney films. Journal of Communication,63, 393–
412. doi:10.1111/jcom.12022
Pike, J. J., & Jennings, N. (2005). The effects of commer-
cials on children’s perceptions of gender appropriate
toy use. Sex Roles,52,83–91.
Rasmussen, E. E. (2013). Theoretical underpinnings of
reducing the media’s negative effect on children: Per-
son-centered, negative-evaluative mediation within a
persuasion framework. In E. L. Cohen (Ed.), Communi-
cation yearbook 37 (pp. 379–406). New York, NY: Rout-
ledge.
Roos, P. E., & Cohen, L. H. (1987). Sex roles and social
support as moderators of life stress adjustment. Journal
of Personality and Social Psychology,52, 576–585.
doi:10.1037/0022-3514.52.3.576
Ruble, D. N., & Martin, C. L. (1998). Gender develop-
ment. In W. Damon (Series Ed.) & N. Eisenberg (Vol.
Ed.), Handbook of child psychology, Vol. 3: Social, emo-
tional, and personality development (5th ed., pp. 933–
1016). New York, NY: Wiley.
Samakow, J. (2013). Merida from ‘Brave’gets an unnecessary
makeover, sparks change.org petition. Retrieved from
http://www.huffingtonpost.com/2013/05/08/merida-
brave-makeover_n_3238223.html
Schooler, D., Kim, J. L., & Sorsoli, L. (2006). Setting rules
or sitting down: Parental mediation of television con-
sumption and adolescent self-esteem, body image, and
sexuality. Sexuality Research & Social Policy,3,49–62.
Sherman, A. M., & Zurbriggen, E. L. (2014). “Boys can be
anything”: Effect of Barbie play on girls’career cogni-
tions. Sex Roles,70, 195–208. doi:10.1007/s11199-014-
0347-y
Signorielli, N. (2011). Television’s gender-role images and
contribution to stereotyping: Past, present, and future.
In D. G. Singer & J. L. Singer (Eds.), Handbook of chil-
dren and the media (2nd ed., pp. 321–339). Thousand
Oaks, CA: Sage.
Smith, S., Pieper, K., Granados, A., & Choueite, M.
(2010). Assessing gender-related portrayals in top-
grossing G-rated films. Sex Roles,62, 774–786.
Smolak, L., & Murnen, S. K. (2011). The sexualization of
girls and women as a primary antecedent of self-objec-
tification. In R. M. Calogero, S. Tantleff-Dunn, & J.
Thompson (Eds.), Self-objectification in women: Causes,
consequences, and counteractions (pp. 53–75). Washington,
DC: American Psychological Association. doi:10.1037/
12304-003
Spence, J. T., & Helmreich, R. L. (1978). Child test battery:
The Toy Preference Questionnaire. Unpublished manu-
script, University of Texas.
Thompson, J., & Stice, E. (2001). Thin-ideal internaliza-
tion: Mounting evidence for a new risk factor for body-
image disturbance and eating pathology. Current Direc-
tions in Psychological Science,10, 181–183. doi:10.1111/
1467-8721.00144
Tremblay, L., Lovsin, T., Zecevic, C., & Lariviere, M.
(2011). Perceptions of self in 3-5 year old children: A
preliminary investigation into the early emergence of
body dissatisfaction. Body Esteem,8, 287–292.
doi:10.1016/j.bodyim.2011.04.004
Vandewater, E. A., Rideout, V. J., Wartella, E. A., Huang,
X., Lee, J. H., & Shim, M. (2007). Digital childhood:
Electronic media and technology use among infants,
toddlers, and preschoolers. Pediatrics,119, e1006–e1015.
Want, S. C. (2009). Meta-analytic moderators of experi-
mental exposure to media portrayals of women on
female appearance satisfaction: Social comparisons as
automatic processes. Body Esteem,6, 257–269.
doi:10.1016/j.bodyim.2009.07.008
Wilson, B. J., & Cantor, J. (1987). Reducing children’s fear
reactions to mass media: Effects of visual exposure and
verbal explanation. In M. M. McLaughlin (Ed.), Commu-
nication yearbook (Vol. 10, pp. 553–573). Newbury Park,
CA: Sage.
Wohlwend, K. E. (2009). Damsels in discourse: Girls con-
suming and producing identity texts through Disney
princess play. Reading Research Quarterly,44,57–83.
doi:10.1598/RRQ.44.1.3
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 gender-typed play. Developmental Psychology,45,
688–701. doi:10.1037/a0014053
Supporting Information
Additional supporting information may be found in
the online version of this article at the publisher’s
website:
Table S1. Coefficient Paths for All Associations
for the Longitudinal Model
Disney Princesses and Child Development 17
