1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

The current study investigated the effects of brief exposure to appearance-related media on young girls' body image. One hundred and twenty-one girls aged 3–6 years old participated. Results indicated that exposure did not affect body dissatisfaction or engagement in appearance-related play behaviours. This is the first empirical study to provide support for previous findings that suggest media exposure does not affect body image in young girls. In contrast to older populations, it is possible that young children may adopt the persona of attractive characters with whom they identify rather than comparing themselves to the characters. Although nearly all girls liked the way they looked, self-report data indicated that nearly one-third of the participants would change something about their physical appearance and nearly half of the girls worried about being fat. Exposure to appearance-related media did not exacerbate concerns.

Recent literature suggests that girls as young as 6 years old experience body dissatisfaction, as evidenced by a preference for an ideal figure that is thinner than their perceived current body size (Ambrosi-Randic, 2000; Davison, Markey, & Birch, 2003; Dittmar, Halliwell, & Ive, 2006; Dohnt & Tiggemann, 2004, 2005, 2006a; Lowes & Tiggemann, 2003). Results, however, are generally mixed with regard to body dissatisfaction in children younger than 6 years old (e.g., Dohnt & Tiggemann, 2005; Hendy, Gustitus, & Leitzel-Schwalm, 2001; Lowes & Tiggemann, 2003). For example, 60% of 5-year-olds in one sample desired a heavier body than they currently perceived themselves to have (Dohnt & Tiggemann, 2006b). One explanation for the mixed findings can be attributed to the lack of available measures shown to possess sound psychometric properties with younger age groups. For instance, figural stimuli are employed frequently to assess the discrepancy between children's ideal and perceived self (Gardner, 2001; Offman & Bradley, 1992; Tiggemann & Wilson-Barrett, 1998). Some problems regarding the use of figural scales are the age appropriateness of the drawings and the use of stereotypical Caucasian features (Byrne & Hill, 1996; Collins, 1991; Gardner, 2001; Tiggemann & Wilson-Barrett, 1998; Truby & Paxton, 2002).

Researchers using alternative methods of assessment (i.e., interview) report more consistently that some children as young as 5 years old are concerned about their weight (Abramovitz & Birch, 2000; Davison, Earnest, & Birch, 2002; Davison, Markey, & Birch, 2000). Longitudinal data indicates that body dissatisfaction and weight concerns at the age of 5 are predictive of future reports at the ages of 7 and 9 (Davison et al., 2003). Dieting and, in some cases, problem eating behaviours also are more prevalent in 9-year-old girls who display signs of body dissatisfaction and weight concerns at younger ages than those who do not (Davison et al., 2003). Additionally, there is evidence to suggest that some young children are aware of the ‘thin ideal’ that exists in Western society and are able to identify dieting as a method used in order to attain the ideal, thin body (e.g., Davison et al., 2000; Dohnt & Tiggemann, 2004, 2005, 2006a; Lowes & Tiggemann, 2003). Due to studies like these that report awareness of the thin ideal and associations between body dissatisfaction and weight concerns in young children, it is not surprising that researchers focus on the underlying factors that may contribute to the onset of disturbance. One way researchers are focusing on underlying factors is by exploring sociocultural influences, such as family, peers, and the media – just like they have done with older populations.

It has been argued that the internalization of sociocultural appearance ideals is a contributing factor to the development of body image disturbance, weight concerns, and possibly even disturbed eating behaviour (Sands & Wardle, 2003; Striegel-Moore & Cachelin, 1999; Thompson, Heinberg, Altabe, & Tantleff-Dunn, 1999). In addition to parents and peers, the role of the media is often studied and cited as an important sociocultural influence on body image (see Levine & Smolak, 1996; Tiggemann, 2002). Recent cross-sectional studies have implicated media exposure via magazines and television programmes as a factor contributing to internalization of the thin ideal and body image concerns for girls as young as 6 years old (Dohnt & Tiggemann, 2006a, b; Harrison & Hefner, 2006). Increased media exposure also is related to children's preference for thinner adult figures that represent their future adult body (Harrison & Hefner, 2006). Similar to other sociocultural factors, however, studies fail to reveal a correlation between reported media exposure and body dissatisfaction in very young children (Harrison, 2000; Sands & Wardle, 2003). The age of 6 appears to be a particularly significant age for many children, especially girls, as it is the age during which sociocultural factors appear to begin forming stronger associations with measures of body dissatisfaction (Dohnt & Tiggemann, 2005, 2006a). However, a small yet surprising percentage of children younger than 6 years old express weight concerns and indicate a preference for thinness (Davison et al., 2000; Dohnt & Tiggemann, 2005).

With children watching television up to 20 h per week (Pardee, Norman, Lustig, Preud'homme, & Schwimmer, 2007), it is not surprising that they identify it as one of their favourite activities (Gilbert, 1998). In fact, young children often are seen imitating their favourite sports, movie, or television star, and they appear to exhibit heightened levels of confidence when they engage in this imitative play (Gilbert, 1998). Interestingly, some of Gilbert's (1998) 5- and 6-year-old participants identified animated cartoon characters as resembling them more frequently than their own family members. This may be of some concern given that many popular children's films contain appearance-related messages (Herbozo, Tantleff-Dunn, Gokee-LaRose, & Thompson, 2004).

Herbozo et al. (2004) recently conducted a content analysis of popular children's media, including 25 children's videos, and found that many popular films contain approximately 10 or more body image related messages. The researchers noted that good characters often are depicted as beautiful and thin, and attractiveness is associated with sociability, kindness, contentedness, and success. In contrast, ‘evil’ is linked more readily to obesity, cruelty, and general unattractiveness. Results of studies conducted by Klein and Shiffman (2005, 2006) echo these findings, even in animated television shows lasting less than 30 min. Klein and Shiffman (2005, 2006), also noted that the number of thin characters depicted in children's animated shows has increased steadily since the 1950s, whereas the number of overweight characters has decreased. Overall, media aimed specifically at children clearly depict an unrealistic thin ideal. Researchers suggest that the internalization of these images may promote body image disturbance and problem eating behaviours – just like non-animated media (Herbozo et al., 2004; Klein & Shiffman, 2006).

To examine the effects of media, many studies using adolescent and adult participants have utilized brief exposure methods (Durkin & Paxton, 2002; Hargreaves & Tiggemann, 2003; Heinberg & Thompson, 1995). A review of extant literature reveals that only one experimental study has examined the effect of brief exposure to appearance-related stimuli on young girls' body image. Dittmar et al. (2006) exposed 5.5- to 8.5-year-old girls to a storybook containing illustrations of Barbie dolls (unrealistic figure), Emme dolls (average-sized figure), or no dolls (control condition). Children exposed to the Barbie book exhibited greater body dissatisfaction compared to the other groups. However, this effect was observed only for the younger age groups (5.5–7.5 years). The researchers suggested that dolls like Barbie may serve as agents of social comparison for younger girls, just as peers and adolescents do for older children. In light of their findings, Dittmar and colleagues urged researchers to further explore the development of body image development in young children. Herbozo et al. (2004) suggested that an empirically based exposure study would contribute significantly to further understanding the media's contribution to the development of body dissatisfaction, body ideals, and biases in young children. To date, no study has examined the effects exposure to animated, appearance-related children's media might have on young girls' body image.

The aim of the present study was to examine the effects of exposure to popular animated children's media on young girls' body image and appearance-related behaviours. It was hypothesized that exposure to appearance-related media would result in more appearance-related play behaviour (e.g., playing dress-up more, playing with a vanity) than demonstrated by girls in the control group. A second hypothesis was that girls exposed to appearance-related media would exhibit greater body dissatisfaction as evidenced by a discrepancy between their ideal and perceived current body size using figural stimuli. Qualitative information obtained from child interviews is presented, and several exploratory analyses were conducted to examine young girls' appearance-related concerns and perceptions of princesses.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References


Participants were recruited from local preschools and elementary schools, e-mail solicitations via a weekly newsletter sent to employees of a large metropolitan university, and flyers distributed at numerous locations. Informed consent was obtained from a parent of each participant. As compensation for participation, all parents and legal guardians were offered the opportunity to receive a brief assessment of their child's general intellectual abilities.

Participants were 121 mother/daughter dyads from a metropolitan area in the Southeast United States. However, data from 11 mothers were not available (parent did not return completed survey packet [N=10] or parent was a non-native reader of English [N=1]). Girls ranged in age from 3 to 6 years old (M=4.44, SD=1.06) and were predominantly Caucasian (62.0%; 13.2% Biracial; 6.6% Hispanic; 5.0% African-American; 4.1% Asian; and 9.1% did not provide a response). The majority of children were of a healthy weight based on maternal report of height and weight (52.2% with a BMI in the 5th to 85th percentile based upon age; 22.2% obese [BMI≥95th percentile]; 12.2% overweight [85th percentile < BMI > 95th percentile]; and 13.4% underweight [BMI < 5th percentile]. Mothers, who served as informants, ranged in age from 23 to 69 years old (M=36.80, SD=6.50). Parental employment information indicated that the vast majority held professional jobs (60.4%; e.g., executives, professors, and lawyers), whereas 17% worked in sales, child care, real estate, or another position that did not require an advanced degree. Fifteen per cent were stay-at-home mothers and 7.5% full-time students.


Ten children participated in a pilot study to assess and refine all included measures. Pilot study results helped to inform the methodology of the current study, and information about pilot data is provided within each relevant section.

Video stimuli

Children in the experimental group were shown a video containing appearance-related clips from 10 animated children's movies (e.g., Cinderella, Barbie in the Nutcracker, Little Mermaid, Anastasia). These clips included only animated characters. To ensure that the chosen clips contained appearance-related messages, 8 of the 10 films were selected based on information obtained from the content analysis conducted by Herbozo et al. (2004). Selected scenes featured main characters making appearance-related comments (e.g., in Beauty and the Beast, Gaston comments that Belle is ‘the most beautiful girl in town and that makes her the best’), or these clips depicted characters engaging in appearance-related behaviours (e.g., changing clothes to enhance one's beauty in Cinderella, Sleeping Beauty, and Anastasia). One to three clips were selected from each animated film, and each clip ranged from 1 to 2 min for a total of approximately 14 min.

The control group viewed a neutral montage of clips from seven animated films that did not contain any appearance-related messages (e.g., Dora the Explorer, Clifford, Dragon Tales, and Lilo & Stitch). Each clip contained animation of a human figure; however, unlike the humans depicted in the experimental clips, these humans were often secondary rather than primary characters. Two clips were selected from each film for a total of approximately 14 min. These clips were viewed and rated by children in a pilot study to ensure that they were comparable to the experimental video clips in interest and familiarity.

Behavioural rating scale

A behavioural rating scale was created and used to identify children's appearance-related behaviours after being exposed to the video clips. This method was employed to supplement additional assessments that have been reported to be less reliable for use with young children (e.g., Figural scales). Two undergraduate research assistants were trained by the researcher to code the observations. The number of activities in which a child could engage was finite and each activity was clearly defined a priori as appearance related (e.g., princess costumes, vanity) or non-appearance related (e.g., legos, kitchen). Raters noted the start and stop times of each activity in which a child engaged, and a 15-s interval was used to gain a frequency count of appearance-related and non-appearance-related behaviours. To assess inter-rater reliability, methods suggested by Cohen (1960) were employed to correct for chance agreement between the raters. Cohen's kappa ranged from .70 to 1.0, with an average of .91.

Child interview

A semi-structured interview was conducted before and after video exposure to assess any pre-test group differences and between-group effects. A trained interviewer asked the children questions about their appearance satisfaction (e.g., ‘Do you like the way you look?’). A visual scale with three markers to illustrate response options (never or almost never, sometimes, and nearly all the time) was used to help children answer some questions. Other appearance-related questions also were asked (e.g., ‘Could you be a princess?’, ‘What would you have to change to become a princess?’). Response options were coded numerically to allow for quantitative analyses. Verbal reinforcement was provided throughout the interview for answering questions regardless of the content of responses. Additionally, children were shown images of three girls and three women dressed as princesses. The images varied by weight but not height or any other characteristic. Children were asked to select the child or woman they believed to be the ‘real’ princess.

Body dissatisfaction

Child figures were created using a computer software program. Digital photographs of the participants' faces were added to the bodies of the computer-generated figures. Face and body colour were matched for each participant. Body dissatisfaction was measured using personalized figures in an effort to compensate for the limitations of generic figure rating scales that present a developmental challenge to younger children because they require the ability to think of the figure as a representation of self. Additionally, generic figure rating scales often fail to account for racial differences (Gardner, 2001). During pilot testing, children manipulated the size of a single figure using a computer mouse. The majority of pilot participants extended the body to an extreme size but reported that they did not think the figure actually looked like them. Given this trend, an alternative method of assessment was utilized during data collection. Each participant was presented two figures at a time on a 15.4 in. computer screen and asked to choose the figure that looked most like her. Children were presented with the figure they chose and another figure until the same figure was chosen twice. The process was then repeated, but the interviewer asked, ‘Which one do you want to look like the most?’ The order of initial figure presentation was consistent throughout the study. Presented figures became thinner or heavier depending upon the participant's initial choice. Body image was calculated by subtracting the figure chosen as the perceived current size from the figure chosen as the ideal size. A negative score indicates a desire for a thinner size, a positive score indicates a desire for a heavier size, and a score of zero indicates contentment with one's body size.

Demographic questionnaire

Mothers completed a demographic questionnaire. A subgroup of mothers provided information about their daughter's television viewing habits and number of Disney DVDs owned. Their report (N=79) indicated that girls watched an average of more than 10 h of television per week (M=11.37, SD=7.54; range: 1–47.5 h). With regard to Disney DVDs, girls (N=71) owned, on average, 18 DVDs (SD=16.47; range: 0–75). The mean number of hours per week spent watching television was comparable to other studies (Liebert & Sprafkin, 1988), and a wide range of responses was reported (3–47.5 h per week); therefore, the respondents likely were representative of the entire sample. BMI for each child was calculated based upon heights and weights provided by mothers. For children, weight percentiles were calculated and used to classify each child's weight category according to standards set forth by the Center for Disease Control.


Children were randomly assigned to the experimental or control condition. A digital headshot was taken of each child participant upon arrival. While their children participated in the experiment, parents completed a brief survey packet.

Each child was accompanied to a room with a couch, television, and mirror by one of five trained female research assistants who acted as the child's playmate. The research assistant ‘playmates’ all had previous experience working with children (e.g., extensive child care background, early education internship) and were trained carefully regarding their role in interacting with the participants. Each playmate was Caucasian, slender, and in her early 20s. All playmates wore casual, appropriate clothing typical of what one might wear to baby-sit (i.e., solid coloured shirts or sweaters and jeans).

After several minutes of chatting to build rapport, each participant was asked to answer questions related to physical appearance. Given the developmental limitations related to interviewing young children, the girls were asked to look at their reflection in the mirror and think of themselves when they answered. Participants then were shown two pictures of themselves on a computer screen with two different body sizes, and they were asked to pick the picture that looked most like them. A forced choice was imposed until the same figure was chosen twice or the most extreme (thinnest or heaviest) figure was chosen. This process was repeated to obtain an ideal figure rating. It is important to note that pre-test ratings were not obtained for the first 34 participants. After the first 34 participants, in order to improve the flow and timing of activities, the participants' pictures were taken 10 min earlier in the procedure which allowed for assessment of pre-exposure body image. Analyses to compare the first 34 participants who did not complete the body image rating scale pre-exposure to the next 34 participants who did complete this task indicated no statistically significant differences between the two groups on any dependent variables when collapsed across control and experimental groups and no significant differences within those groups either. Based on these analyses and the fact that this was not the only body image related activity prior to exposure, it is unlikely that the procedural change had an impact on overall results.

Participants were then shown a series of three child and three adult figures of varying weights dressed as princesses. The adult and child princesses were presented as Caucasian with blond hair and brown eyes, which is consistent with the presentation of princesses in popular children's media. Child princesses wore a pink ballerina outfit and adult princesses wore a sleeveless pink gown. Computer software allowed for the manipulation of proportional weight changes while keeping height stable. Participants were told that, although each figure looked the same, only one was a ‘real princess’ and that they were to identify the ‘real’ one.

After completion of all pre-test interview questions and measures, girls in the experimental condition viewed clips featuring animated characters from films containing appearance-related messages (e.g., Beauty and the Beast), whereas girls in the control condition viewed clips featuring animated human characters from television shows and movies that were not appearance focused (e.g., Dora the Explorer). The playmate encouraged the child participants to pay attention to the ‘movie’ and reinforced attentive behaviour.

After viewing the video clips, each participant was taken to a playroom next door in which children's music was played softly to create a more comfortable and inviting environment. The playroom contained different play stations including a dress-up rack consisting of costumes similar to those worn by characters featured in the experimental video clips as well as costumes that were unrelated to the clips (e.g., fireman, doctor). The playroom also contained a vanity (including brush, ‘play’ make-up, hair accessories, etc.), blocks and legos, dinosaurs, a dollhouse, and a kitchen set. Free play was observed for 8–15 min. Variance in playtime was largely due to the individual child's interest and participation. Given attentional limitations in young children, play time was shortened when it became evident that a child was becoming disinterested or unfocused. Although the playmate accompanied the child to the playroom, she did not initiate any particular play activities. She joined the child in an activity only if she was invited, and she refrained from directing the child or reinforcing any particular play activity. After playing, each participant was accompanied back to the other room where the pre-test interview and measures were administered again at post-test.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

All variables were screened for skewness, kurtosis, outliers, and homogeneity of variance using Levene's test. There were only two variables for which both non-normality of distribution and unequal variance were a problem, suggesting that the majority of variables did not violate statistical assumptions underlying the parametric statistics used (Keppel & Zedeck, 1989). Violations of normality were observed for analyses on body dissatisfaction at pre- and post-test. Attempts to correct these violations were unsuccessful using square-root and logarithmic transformations. Therefore, these variables remained untransformed.

Based on independent t tests and chi-squared analyses, there were no significant differences between the experimental and control groups in age, race, weight classification category, weekly television viewing hours, or number of Disney DVDs owned.

Participants with missing data were excluded from analyses on a case-by-case basis. All analyses examined the effect of age on dependent variables. Two age groups were formed [3- and 4-year-olds (32 in the control condition and 32 in the experimental condition) vs. 5- and 6-year-olds (27 in the control condition and 30 in the experimental condition)]. As recommended by Cohen (1992), the current study utilized a power of .80 and an alpha of .05. In light of the study design and main analyses, G*Power3 (Faul, Erdfelder, Lang, & Buchner, 2007) recommended a total sample size of 88.

Effect of appearance-related media on play behaviour

A 2 × 2 univariate analysis of variance (ANOVA; condition × age) was conducted to determine if girls exposed to appearance-related children's media would be more likely to engage in appearance-related play activities than control group participants. Although no a priori hypotheses were made about age, it was included as an independent variable to explore any potential developmental differences. Results of the ANOVA suggested that the manipulation failed to produce significant between-group differences for condition, F (1,111)=0.004, p=.95; η2=.00, or age, F (1,111)=3.04, p=.08; η2=.03. The interaction between condition and age also was non-significant, F (1,111)=0.39, p=.53; η2=.004. Overall, the percentage of time girls in the control group engaged in appearance-related play (3- and 4-year-olds: M=35.89, SD=30.28; 5- and 6-year-olds: M=29.40, SD=34.27) was comparable to that of girls in the experimental group (3- and 4-year-olds: M=39.18, SD=33.78; 5- and 6-year-olds: M=25.41, SD=25.12).

Effect of appearance-related media on body image

A 2 × 2×2 (time × condition × age) mixed model ANOVA was conducted to evaluate the effects of exposure to appearance-related media on young girls' body image. Body image ratings at pre- and post-test (N=86) were examined as a function of condition and age. With regard to body dissatisfaction, results failed to reveal any main effects or interactions.

Additional mixed model ANOVAs examined any changes from pre- to post-test for ideal and perceived current body sizes. With regard to ideal size ratings, results failed to reveal any main effect of time or interactions between time, condition, and age. However, there was a main effect of age, F (1,82)=4.64, p=.03; η2=.05 indicating that, across condition and time, older girls (M=3.65, SD=1.12) chose a significantly thinner ideal shape than younger girls (M=4.17, SD=1.12).

For perceived body sizes, there was a significant main effect of time, F (1,82)=5.93, p=.02; η2=.07, indicating that across both conditions and age groups, girls identified a significantly heavier figure as her perceived size at pre-test (M=4.40, SD=1.31) than at post-test (M=3.98, SD=1.59). There also was a main effect of age, F (1,82)=5.49, p=.02, η2=.06, indicating that across both time and condition, older girls (M=4.48, SD=1.21) selected a significantly thinner figure as her perceived size than younger girls (M=3.87, SD=1.21). Results failed to reveal any significant interactions between time, condition, and age.

Exploratory analyses

Effect of appearance-related media on girls' concerns about weight

At pre-test, 45.5% (N=55) of girls reported that they never worry about being fat, 18.2% (N=22) reported that they sometimes worry, 31.4% (N=38) reported that they almost always worry, and 5% (N=6) were non-responsive or refused to answer. Comparable responses were observed in all girls regardless of age (χ2(2)=3.14, p=.21) and weight classification category (χ2(6)=7.49, p=.28). Mixed model ANOVAs were conducted to evaluate the effects of exposure to appearance-related media on young girls' (N=111) weight concerns. Results revealed an interaction of time, condition, and age, F (1,106)=5.81, p=.02; η2=.05. Follow-up protected t tests suggested that older girls in the control group decreased significantly from pre- to post-test (M=1.89, SD=0.93 to M=1.37, SD=0.69), t (23)=3.25, p=.004, whereas older girls in the experimental group did not significantly change from pre-test to post-test (M=1.54, SD=0.79 to M=1.68, SD=0.86), t (25)=−0.70, p=.49. The mean increase observed in older experimental participants and decrease observed in older control participants from pre-test to post-test is responsible for the significant interaction of time, condition, and age. Although slight mean decreases were observed, younger girls weight concerns did not significantly change in the control group (M=1.93, SD=0.90 to M=1.82, SD=0.82), t (27)=0.593, p=.558, or in the experimental group (M=2.11, SD=0.89 to M=1.78, SD=0.85), t (26)=0.803, p=.083.

Results also revealed a main effect of age such that, across time and condition, younger girls (M=1.91, SD=0.70) reported more frequent fat worry than older girls (M=1.62, SD=0.70), F (1,106)=4.67, p=.03; η2=.044. Results did not reveal any main effect of condition or interaction between condition and age. To further explore children's thoughts about fatness, several participants were asked to express their thoughts about being fat, and responses ranged from ‘being fat is bad’ to ‘my mommy thinks she's fat’. Additional example responses included, ‘I don't want to be fat because I don't like it’, and ‘mom don't want to be fat’.

Effect of appearance-related media on girls' perceptions of princesses

At pre-test, 50% of girls identified the thinnest child figure as the ‘real’ princess (28% identified the average-sized figure and 23% identified the heaviest figure). A mixed model ANOVA was conducted to assess the effects of appearance-related media exposure on girls' choice of the ‘real’ child princess. There were no significant main effects or interactions.

With regard to adult princess selection, 39% of girls identified the thinnest woman as the ‘real’ princess at pre-test (28% identified the average-sized figure and 34% identified the heaviest figure). Results of a mixed model ANOVA revealed a significant main effect of age, F (1, 109)=7.35, p=.008; η2=.063, indicating that, across time and condition, older girls (M=1.70, SD=0.67) identified a significantly thinner adult princess than younger girls (M=2.07, SD=0.71). There were no other significant main effects or interactions.

Self-report of appearance-related concerns: The child interview

Pre-test interviews revealed that all but three participants liked the way they looked; however, when asked if there was anything they disliked, close to half the sample spontaneously provided a response [24.8% (N=30) of all participants indicated that they disliked something about their physical appearance (e.g., hair, skin colour, and body part), 16.5% (N=20) revealed that they were dissatisfied with their clothing, and 2.5% (N=3) disliked something about herself that was non-appearance related]. In response to being asked ‘If you could change anything about the way you look, what would it be?’, 30.6% (N=37) of girls noted that they would change something about their physical appearance. Of those responders, 59.5% (N=22) would change their hair, 27% would change something about their body [i.e., skin colour (N=8), make their legs skinnier (N=2), or alter something else (N=5) such as clothing or accessories, change into a princess character, or look more like a female friend or family member].

The majority of girls (N=99) believed that they could be a princess regardless of their weight (χ2(3)=2.40, p=.49) and age (χ2(3)=1.96, p=.58). The majority of girls indicated that something non-appearance related (e.g., have a queen for a mom; 32.3%, N=32) or their clothing or accessories (e.g., have a pretty dress or a crown; 53.3%, N=53) would make them a princess. Approximately, 8% (N=8) endorsed needing to change their hair or skin colour to become a princess. Example responses included ‘my hair would have to grow long’, ‘I'd need yellow hair’, ‘I'd paint myself white’, and ‘I would change from brown skin to white skin’. Overall, interview responses were largely comparable at post-test regardless of assigned condition, age, or weight category.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References

Recent literature suggests animated children's media contains a surprising number of appearance-related messages. The aim of the current study was to investigate the impact exposure to this type of media may have on very young girls' body image. The current experimental study was the first to test the effects of exposure to appearance-related animated media (e.g., Beauty and the Beast) with girls as young as 3 years old. Although it was hypothesized that exposure to appearance-related media would result in more appearance-related play activity, results failed to reveal any differences between the exposure conditions. In fact, a wide range of play behaviours were observed within each group, with some children engaging exclusively in a single type of activity. Similarly, results also failed to reveal any direct negative effect on girls' body dissatisfaction. These results are in contrast to what was hypothesized; however, they are consistent with several cross-sectional studies that also have reported that media exposure does not affect body dissatisfaction in girls younger than 6 years old (Dohnt & Tiggemann, 2004, 2005). In general, 5- and 6-year-olds selected significantly thinner figures as their perceived and ideals sizes compared with 3- and 4-year-olds. This finding may illustrate the first developmental shift that may contribute to the development of body dissatisfaction. It is possible that as some girls continue to age their selection of ideal size continues to decrease while their perceived size remains stable or increases. Although not statistically significant, it may be of clinical interest to note that approximately one-third of all participants desired a thinner ideal figure compared to their current perceived size. This trend may be of particular importance in light of the work by Davison et al. (2003) that suggested weight concerns at younger ages are predictive of disturbed eating behaviour at older ages.

Data obtained through interviews suggested that girls were generally satisfied with their appearance and that the majority of participants believed that they could be princesses. However, close to a third of girls indicated that they would, if possible, change something about their physical appearance (i.e., hair, skin colour, or weight). Additionally, approximately half of girls reported worrying about being fat sometimes or almost always. Surprisingly, an interaction of time, condition, and age was observed with regard to girls' self-report of weight concerns. This result reflected the slight increase of fat worry reported by older girls in the experimental group and significant decrease in fat worry reported by older girls in the control group. Younger girls in the experimental group observed a decrease in reported worry that approached significance, whereas younger girls in the control group experienced a slight non-significant decrease. The reason for this finding is unclear. However, it is possible that girls may have reported their frequency of fat worry in the moment rather than in general. In particular, exposure to a mirror and a series of questions focused on physical appearance may have increased baseline concern and worry for older girls. The decrease over time for older control participants' weight concerns may be reflective of their increased comfort in the study environment, a suggestion that has been made by other researchers who have observed similar control group changes (Agliata & Tantleff-Dunn, 2004). For older experimental participants, exposure to appearance-related messages and stimuli may have been potent enough to maintain baseline ratings for the majority of participants and increase fat worry for some.

With regard to perceptions of princesses, 5- and 6-year-olds, across condition and time, identified significantly thinner adult figures as the ‘real’ princess compared with 3- and 4-year-olds. Older girls likely have had more exposure to idealized images of princesses, which might have contributed to this result. Alternatively, this finding may reflect a preference for thinness that become more apparent in early elementary school-age children. Although significant age differences were not observed for the child princess, the vast majority of participants believed the ‘real’ child and adult princess was the thinnest or average-sized figure. This may reflect routine exposure to idealized images of thinness and the lack of positive portrayals of females with larger figures.

Limitations and future directions

Limitations of the current study should be considered when evaluating the results. First, the relative brevity of exposure is a limitation of the current study (14 min of clips), particularly since children are saturated by media exposure on a regular basis. Rather than relying on longer exposure periods with young children who have shorter attention spans, future studies should further explore the impact of children's everyday exposure to appearance-related media. Alternatively, researchers may wish to utilize repeated exposure methods. A second limitation of the current study is that all participants were exposed to appearance-related stimuli (i.e., toys, princess images, and related questions) prior to post-test. In future studies, it will be important to avoid or limit control group exposure to appearance-related stimuli prior to post-test assessments to isolate exposure effects. The demographic make-up of the current sample is a third limitation. Although the current sample was more ethnically and racially diverse than typically reported in other studies (Dittmar et al., 2006; and Dohnt & Tiggemann, 2005; reported that more than 90% of their participants were Caucasian), the overall sample size did not allow for examination of differences between the various groups. Examination of race may be of particular interest given that the majority of princess characters are illustrated, almost exclusively, as Caucasian. The current study also only examined girls' perceptions of Caucasian princesses. Future studies may wish to determine if girls' choice of princess differs according to race as well as size. Particularly, in light of some of the children's comments regarding the need to change their skin colour to be a princess, a cross-cultural study on the impact of children's media is warranted. Another limitation of the current sample was that parents were, on average, highly educated, of at least middle-class status, and in their mid- to late 30s. Replication with a more heterogeneous sample would increase the generalizabiltiy of the current findings to a broader population.

Finally, the current measure of body dissatisfaction has not been standardized with a large, heterogeneous sample. Body dissatisfaction was assessed using a computerized figure rating scale that was personalized in consideration of developmental concerns and psychometric limitations (Gardner, 2001) associated with generic figure rating scales. The measure was personalized in hopes of increasing young girls' immersion in the task and their identification with the figure. Researchers should seek larger samples of children to determine if the personalization of the body dissatisfaction instrument yields adequate reliability estimates with young children, thereby providing a useful alternative to unreliable generic figure rating scales.


Collectively, results of this study and other children's media studies suggest strongly that developmental considerations must be made when defining and investigating body image in children. Despite objective evidence that indicates children's media contains many appearance-related messages that may affect body dissatisfaction, very young girls do not appear to be affected by these messages in ways comparable to their older counterparts. This may be because, at younger ages, children frequently engage in pretend play (thus, adopting the role of the character) and may not be capable of making subtle social comparisons. However, as children become older and more cognitively savvy, they engage less in pretend play and as a result may stop identifying themselves as the characters they idolize. Based upon extant literature, subtle cognitive shifts likely occur around the critical age of 6 years old. These changes might contribute to a movement from identification with favourite media figures to an inappropriate comparison with them. Longitudinal data are needed to determine the long-term effects of early exposure to media that illustrates the thin ideal and conveys beauty messages and ideals that are often associated with body dissatisfaction and disturbed eating behaviour in much older populations.


  1. Top of page
  2. Abstract
  3. Method
  4. Results
  5. Discussion
  6. References
  • Abramovitz, B. A., Birch, L. L.. Five-year-old girls' ideas about dieting are predicted by mother's dieting. Journal of the American Dietetic Association, 100(10), 11571163, 2000.
  • Agliata, D., Tantleff-Dunn, S.. The impact of media exposure on males' body image. Journal of Social and Clinical Psychology, 23(1), 722, 2004.
  • Ambrosi-Randic, N.. Perception of current and ideal body size in preschool age children. Perceptual and Motor Skills, 90, 885889, 2000.
  • Byrne, N. M., Hill, A. P.. Should body-image scales designed for adults be used with adolescents?. Perceptual and Motor Skills, 82, 747753, 1996.
  • Cohen, J.. A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20(1), 3747, 1960.
  • Cohen, J.. A power primer. Psychological Bulletin, 112(1), 155159, 1992.
  • Collins, M. E.. Body figure perceptions and preferences among preadolescent children. International Journal of Eating Disorders, 10(2), 199208, 1991.
  • Davison, K. K., Earnest, M. B., Birch, L. L.. Participation in aesthetic sports and girls' weight concerns at ages 5 and 7 years. International Journal of Eating Disorders, 31(3), 312317, 2002.
  • Davison, K. K., Markey, C. N., Birch, L. L.. Etiology of body dissatisfaction and weight concerns among 5-year-old girls. Appetite, 35, 143151, 2000.
  • Davison, K. K., Markey, C. N., Birch, L. L.. A longitudinal examination of patterns in girls' weight concerns and body dissatisfaction from ages 5 to 9 years. International Journal of Eating Disorders, 33, 320332, 2003.
  • Dittmar, H., Halliwell, E., Ive, S.. Does Barbie make girls want to be thin? The effect of experimental exposure to images of dolls on the body image of 5- to 8-year-old girls. Developmental Psychology, 42(2), 283292, 2006.
  • Dohnt, H., Tiggemann, M.. Development of perceived body size and dieting awareness in young girls. Perceptual and Motor Skills, 99(3), 790792, 2004.
  • Dohnt, H., Tiggemann, M.. Peer influences on body dissatisfaction and dieting awareness in young girls. British Journal of Developmental Psychology, 23(1), 103116, 2005.
  • Dohnt, H., Tiggemann, M.. Body image concerns in young girls: The role of peers and media prior to adolescence. Journal of Youth and Adolescence, 35(2), 141151, 2006a.
  • Dohnt, H., Tiggemann, M.. The contribution of peer and media influences to the development of body satisfaction and self-esteem in young girls: A prospective study. Developmental Psychology, 42(5), 929936, 2006b.
  • Durkin, S. J., Paxton, S. J.. Predictors of vulnerability to reduce body image satisfaction and psychological wellbeing in response to exposure to idealized female media images in adolescent girls. Journal of Psychosomatic Research, 53, 9951005, 2002.
  • Faul, F., Erdfelder, E., Lang, A.-G., Buchner, A.. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175191, 2007.
  • Gardner, R. M.. Assessment of body image disturbance in children and adolescents. Thompson, J. K., Smolak, L.. Body image, eating disorders, and obesity in youth: Assessment, prevention, and treatment, 193213 Washington, DC, American Psychological Association, 2001.
  • Gilbert, K.. The body, young children, and popular culture. Yelland, N.. Gender in early childhood, 5571 Florence, KY, Taylor and Francis, 1998.
  • Hargreaves, D., Tiggemann, M.. The effect of thin ideal television commercials on body dissatisfaction ad schema activation during early adolescence. Journal of Youth and Adolescence, 32, 367373, 2003.
  • Harrison, K.. Television viewing, fat stereotyping, body shape standards, and eating disorder symptomatology in grade school children. Communication Research, 27(5), 617640, 2000.
  • Harrison, K., Hefner, V.. Media exposure, current and future body ideals, and disordered eating among preadolescent girls: A longitudinal panel study. Journal of Youth and Adolescence, 35(2), 153163, 2006.
  • Heinberg, L. J., Thomson, J. K.. Body image and televised images of thinness and attractiveness: A controlled laboratory investigation. Journal of Social and Clinical Psychology, 14(4), 325338, 1995.
  • Hendy, H. M., Gustitus, C., Leitzel-Schwalm, J.. Social cognitive predictors of body image in preschool children. Sex Roles, 44, 557569, 2001.
  • Herbozo, S., Tantleff-Dunn, S., Gokee-LaRose, J., Thompson, J. K.. Beauty and thinness messages in children's media: A content analysis. Eating Disorders, 12(1), 2134, 2004.
  • Keppel, G., Zedeck, S.. Data analysis for research designs: Analysis of variance and multiple regression/correlation approaches. New York, Freeman, 1989.
  • Klein, H., Shiffman, K. S.. Thin is in and stout is out: What animated cartoons tell viewers about body weight. Eating and Weight Disorders, 10(2), 107116, 2005.
  • Klein, H., Shiffman, K. S.. Messages about physical attractiveness in animated cartoons. Body Image, 3(4), 353363, 2006.
  • Levine, M. P., Smolak, L.. Media as a context for the development of disordered eating. Smolak, L., Levine, M. P.. The developmental psychopathology of eating disorders: Implications for research, prevention, and treatment, 235257 Hillsdale, NJ, Erlbaum, 1996.
  • Liebert, R. M., Sprafkin, J.. The early window: Effects of television on children and youth 3rd ed. New York, Pergamon, 1988.
  • Lowes, J., Tiggemann, M.. Body dissatisfaction, dieting awareness and the impact of parental influence in young children. British Journal of Health Psychology, 8(2), 135147, 2003.
  • Offman, H. J., Bradley, M. D.. Body image of children and adolescents and its measurement: An overview. Canadian Journal of Psychiatry, 37, 417422, 1992.
  • Pardee, P. E., Norman, G. J., Lustig, R. H., Preud'homme, D., Schwimmer, J. B.. Television viewing and hypertension in obese children. American Journal of Preventative Medicine, 33(6), 439443, 2007.
  • Sands, E. R., Wardle, J.. Internalization of ideal body shapes in 9–12-year old girls. International Journal of Eating Disorders, 33, 193204, 2003.
  • Striegel-Moore, R. H., Cachelin, F. M.. Body image concerns and disordered eating in adolescent girls: Risk and protective factors. Johnson, N. G., Roberts, M. C.. Beyond appearance: A new look at adolescent girls, 85108 Washington, DC, American Psychological Association, 1999.
  • Thompson, J. K., Heinberg, L. J., Altabe, M., Tantleff-Dunn, S.. Exacting beauty: Theory, assessment, and treatment of body image disturbance. Washington, DC, American Psychological Association, 1999.
  • Tiggemann, M.. Media influences on body image development. Cash, T. F., Pruzinsky, T.. Body image, 9198 New York, Guilford Press, 2002.
  • Tiggemann, M., Wilson-Barrett, E.. Children's figure ratings: Relationship to self-esteem and negative stereotyping. International Journal of Eating Disorders, 23, 8388, 1998.
  • Truby, H., Paxton, S. J.. Development of the children's body image scale. British Journal of Clinical Psychology, 41, 185203, 2002.