View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

Running head: Brief Report: Testing a Dissonance Body Image Intervention

Brief Report: Testing a Dissonance Body Image Intervention among Young Girls

Emma Halliwell & Phillippa C. Diedrichs

Centre for Appearance Research, University of the West of England, UK

Accepted Pre-publication Version

Full citation:

Halliwell, E., & Diedrichs, P.C. (2014). Brief report: Testing a dissonance body image

intervention among young girls. Health Psychology, 33, 201-204. doi: 10.1037/a0032585

Author Note

Emma Halliwell, Centre for Appearance Research, University of the West of England, UK;

Phillipa C. Diedrichs, Centre for Appearance Research, University of the West of England,

UK.

Address correspondence: Emma Halliwell, Centre for Appearance Research, Department of

Psychology, University of the West of England, Frenchay, Coldharbour Lane, Bristol, BS16

9QY. E-mail: [email protected]

The authors wish to acknowledge the assistance of Chantelle Bailey and Rebecca Bellow in

facilitating the intervention sessions.

1
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Abstract

Objective: Body image and eating disorder interventions based on cognitive dissonance have

been shown to be effective among girls and women aged 14 and above. This paper reports a

preliminary examination of whether a dissonance intervention is also effective when

delivered in a school setting to 12 and 13 year old girls in the UK. Methods: Girls (N = 106,

mean age= 12.07 years, SD = .27) were allocated to the intervention condition or a waitlist

control. Results: In contrast to the control group, girls in the intervention condition reported

significant reductions in body dissatisfaction and internalization of a thin body ideal post-

intervention. There was no significant change in self-reported dietary restraint for either

condition. In addition, compared to the control group, girls in the intervention condition

showed increased resilience to negative media effects one-month post intervention.

Conclusions: Results suggests that dissonance based programs can reduce body

dissatisfaction, internalization and negative media effects among a younger group of girls

than previously examined and in a UK school setting.

Keywords: Cognitive-dissonance, interventions, body dissatisfaction, adolescent, girls

2
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Brief Report: Testing a Dissonance Body Image Intervention among Young Girls

Body dissatisfaction is an important health issue; it is reported by 50-70% of adolescent girls

(Wertheim & Paxton, 2011) and has numerous negative consequences including increased

negative affect (Stice & Bearman, 2001) and reduced levels of physical activity (Neumark-

Sztainer, Paxton, Hannan, Haines, & Story, 2006). Body dissatisfaction is also a risk factor

for the onset of eating disorders (Stice, Marti, & Durant, 2011).

Cognitive dissonance-based (CD) interventions are emerging as efficacious and

effective eating disorder and body dissatisfaction prevention programs. Meta-analytic review

reveals that they are the most effective targeted interventions to date for girls aged 14 and

above (Stice, Shaw, & Marti, 2007). For girls with pre-existing body image issues, CD

interventions lead to reductions in thin-ideal internalization, body dissatisfaction, negative

affect, psychosocial impairment and risk for onset of eating disorders (Stice, et al., 2007) with

effects maintained up to three years post-intervention (e.g. Stice, Rohde, Shaw & Gau, 2011).

In addition, there is evidence that CD interventions are also effective as universal prevention

programs, although they have smaller effects in universal than in targeted trials (e.g. Becker,

Smith & Ciao, 2006).

CD interventions are multifaceted, incorporating dissonance, psycho educational,

media literacy and behavioral components. However, a critical component of the intervention

is engaging participants in counter-attitudinal activities that require them to speak out against

the thin ideal of beauty (Stice & Presnell, 2007). This creates dissonance because individuals

who internalize the thin ideal are acting in a manner inconsistent with their beliefs. As

dissonance is uncomfortable, individuals change their beliefs to correspond with their actions.

Therefore, facilitators encourage all participants to make counter-attitudinal statements

during the sessions. Comparison of CD activities and the full intervention indicates that both

are effective at reducing eating pathology (Roehrig, Thompson, Brannick, & van den Berg,

3
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

2006) and supports the central role of dissonance. The most widely used CD programs

involve four hours of intervention. However, two hours of intervention has been shown to be

effective at reducing body dissatisfaction, thin-ideal internalization and eating disturbance

among college women (Matusek et al, 2004).

To date, CD interventions have only been examined among girls aged 14 or older, and

there is currently no intervention with a similar evidence-base for early adolescent girls. This

is an important gap in the literature. Body dissatisfaction is often established before 14 and

internalization is evident among 7-11 year old girls (Evans, Tové, Boothroyd & Drewett, in

press). The need for early prevention efforts has been increasingly recognized by researchers

and clinicians (Holt & Ricciardelli, 2008). Given the efficacy of CD interventions with older

samples, it is useful to examine their suitability for girls in early adolescence.

Exposure to media portraying the thin ideal of beauty is a variable risk factor for the

development of body dissatisfaction and disordered eating among girls (Levine & Murnen,

2009). Meta-analysis reveals that exposure to thin-ideal media has an immediate negative

effect on adolescent girls’ body image (e.g. Grabe, et al., 2008). To date, the ability of CD

interventions to increase resilience to media exposure effects has not been investigated.

This study extends previous research by examining the effectiveness of a relatively

brief, CD intervention among 12 and 13 year old girls in the UK. It examines whether a CD

intervention has the capacity to increase girls’ resilience to thin ideal media, in addition to

reducing self-reported dietary restraint, thin-ideal internalization and body dissatisfaction.

Method

Participants

Participants were 104 girls aged 12 and 13 years (Mage=12.07, SDage = .27; MBMI=

19.92, SDBMI = 4.56), from a single-sex secondary school. Seventy percent of the girls were

White, 7% were Black, 12% were Asian, and 11% were Mixed Race. School staff assigned

4
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

two classes (n = 52) to the intervention condition and two classes (n = 52) to the control

condition. Ninety-six girls completed a follow-up media exposure study and were randomly

allocated to a model exposure (n = 47) or control condition (n = 49).

Materials

The intervention sessions

Six activities from the Body Project (Stice & Presnell, 2007) were selected for this

intervention. Participants also completed two additional psycho educational and media

literacy activities which focused on critiquing the thin-ideal. The content of each session is

reported in table 1.

Media exposure images

Advertisements were selected from magazines aimed at adolescent girls. In the model

exposure condition three featured thin models and one featured the product only. In the

control condition the four advertisements featured hair or perfume products and no models.

Measures

Internalization of the thin ideal. The Thin Ideal Internalization Scale (Stice & Agras,

1998) was administered. Pre-testing revealed that 12 year old girls did not understand the

term ‘shapely’ and one item was deleted for this study. Items were rated on a 5-point Likert-

scale (1= strongly disagree and 5= strongly agree). The Cronbach’s alpha was 0.77 at

baseline and 0.76 post-intervention.

Body dissatisfaction. Participants rated their satisfaction on four items (weight,

stomach, buttocks, and thighs) from the Body Parts Scale (Berscheid, Walster, & Bohrnstedt,

1973) on a 5-point Likert scale (1= very satisfied and 5= very dissatisfied). The Cronbach’s

alpha was 0.82 at baseline and 0.83 post-intervention.

5
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Dietary restraint. The Dutch Restrained Eating Scale (DRES; Van Strien, Frijiters,

Bergers, & Defares, 1986) was used with items rated on a 5-point Likert scale (1 = never and

5 = always). The Cronbach’s alpha was 0.87 at baseline and 0.88 post-intervention.

These trait measures are valid for adolescent girls (Stice et al., 2006).

State body satisfaction. Participants rated how satisfied they felt right now with their

‘body size and shape’ and their ‘weight’ on a 7-point scale (1 = extremely dissatisfied and 7 =

extremely satisfied). The items were highly correlated, r(94) = .85, p<.001.

Procedure

The study was approved by the university ethics committee. Passive parental consent

was obtained for 98% of girls. A week before the intervention girls were informed about the

study and invited to participate. All eligible girls consented and completed the baseline

questionnaire supervised by the first author. Girls in the control condition attended their

regular lessons which did not include content related to body image. Girls in the intervention

condition attended weekly 20 minute sessions for four weeks. They worked in groups of 6 to

8 facilitated by one of the authors or a doctoral student. The week after the last session all

participants completed the follow-up questionnaire supervised by a researcher.

Four weeks later participants took part in a media exposure study. This was presented

as an unrelated study investigating attitudes towards advertising in teenage magazines and

was administered by a different researcher. Participants were asked to complete a

questionnaire evaluating the effectiveness of four advertisements either featuring models or

control images. They were told that current mood can influence our preferences and were

asked to rate how they were feeling right now about various aspects of their lives including

two items assessing state body satisfaction. Participants were debriefed. After the study was

completed the control group of girls we also invited to take part in the intervention.

Results

6
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Fourteen participants did not complete the follow up questionnaire; they did not differ

significantly from the participants who completed follow up on any variables. Maximum-

likelihood estimation was used to impute missing data for these participants as it is preferable

to case deletion (Schafer & Graham, 2002). At baseline there were no significant differences

between the girls assigned to each condition on BMI, internalization, or restraint. However,

girls in the intervention condition reported higher body dissatisfaction (M = 3.09, SD = .93)

than girls in the control condition (M = 2.70, SD = .86), t(102) = -2.32, p = .02, d = .44.

The impact of the intervention sessions

A mixed-design MANOVA with time as a within-subject factor (Time 1, Time 2) and

condition as a between-subject factor (intervention, no intervention) revealed a significant

interaction effect, Lambda = .88, F(3, 100) = 4.49, p = .01, partial η2 = .12. Significant

univariate interaction effects between time and condition were evident for internalization,

F(1, 102) = 7.10, p = .009, partial η2 = .07 and body dissatisfaction, F(1, 102) = 10.02, p =

.002, partial η2 = .09, but not for restraint, F(1, 102) = .77, p = .38, partial η2 = .01.

In the control group there were no statistically significant differences across time for

internalization, t(51) = -1.13, p = .26, d = -.12 (T1 M = 3.16, SD = .67, T2 M = 3.24, SD =

.69) or body dissatisfaction, t(51) = -1.20, p = .24, d = -.10 (T1 M = 2.68, SD = .86, T2 M =

2.77, SD = .89). However, the intervention group reported significantly reduced

internalization, t(51) = 2.44, p = .02, d = .36, (T1 M = 3.20, SD = .83 T2 M = 2.91, SD = .80)

and body dissatisfaction, t(51) = 3.18, p = .003, d = .29 (T1 M = 3.09, SD = .93 T2 M = 2.82,

SD = .96).

Vulnerability to media exposure

A 2 (intervention vs. control) by 2 (model condition vs. no model condition) ANOVA

revealed non-significant main effects for intervention F(1, 80) = 2.16, ns, partial η2 = .03, and

7
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

media exposure, F(1, 80) = 1.32, ns, partial η2 = .02. However, there was a significant

interaction between the intervention and exposure, F(1, 73) = 6.61, p < .05, partial η2 = .08.

The control participants reported less body satisfaction after exposure to thin models

(M = 3.63, SD = 1.75) than after control images (M = 4.78, SD = 1.63), F(1,37) = 4.46, p <

.05, partial η2 = .11. In the intervention group there was no significant difference between

state body satisfaction reported after viewing thin models (M = 4.88, SD = 1.71) or control

images (M = 4.55, SD = 1.27), F(1,43) = .57, ns, partial η2 = .01.

Discussion

Girls aged 12 and 13 reported significantly lower levels of trait body dissatisfaction

and internalization after participating in a relatively short CD intervention. These findings are

promising as they suggest that CD interventions may be effective among a younger group of

girls than has been previously studied and that the intervention is effective in the UK, as well

as the USA. Another novel finding of this study is that the girls who took part in the

intervention were protected from negative media exposure effects five weeks post-

intervention. This suggests that CD interventions may also provide resilience to thin ideal

media imagery, a variable risk factor for the development of eating disorders (Levine &

Murnen, 2009). The media exposure results also have the advantage of being free from

demand characteristics typically associated with repeated measurement in intervention trials.

As expected, the effect sizes in this study (d = .36 for internalization and d = .29 for

body dissatisfaction) are lower than effect sizes from pre to immediately post treatment in

targeted trials (internalization, d = 1.09 and body dissatisfaction, d = .74, Stice et al, 2006).

However, our effect sizes are also smaller than in other universal trials (internalization, d =

.53 and body dissatisfaction, d = .33, Becker et al., 2008). The shorter intervention may have

lower efficacy or dissonance may be less powerful for 12 and 13 year old girls than for older

adolescents and adult women. These issues warrant further investigation.

8
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Contrary to predictions the intervention did not significantly impact on self-reported

restraint. Dietary restraint is a contentious construct in the literature. Measures of dietary

restraint do not correlate with actual calorie restriction (Stice et al., 2007) and moderate levels

of dieting can be related to positive health behaviors, such as increased consumption of fruit

and vegetables and decreased fat consumption (Daee et al., 2002). However, an eight year

prospective study with adolescent girls indicates that self-reported dietary restraint assessed

via the DRES is a risk factor for the onset of eating disorders (Stice et al., 2011). Although

this was a non-targeted sample, the initial levels of restraint (intervention M = 2.19, SD = .84,

control M = 2.37, SD = .86) were similar to those in a targeted intervention with girls aged14

to 19 (Stice et al, 2009; intervention M = 2.35, SD = .86, control M = 2.31, SD = .86). It may

be that the restraint items are interpreted differently by early and middle/late adolescent girls.

The intervention is multifaceted and it is not possible to tease apart the influence of

dissonance and media literacy components in this study. It would be informative to examine

the impact of this intervention on a more comprehensive measure of body satisfaction and on

a measure of restraint that taps into actual reductions in food intake, like the Eating Inventory

(Stunkard & Messer, 1988). Another limitation of this study is that allocation to condition

was not randomized. Constraints on condition allocation are typical of applied research. In

this case, concerns about lack of randomization are reduced by the known efficacy of the

intervention and because the analysis accounts for baseline differences in the study variables.

The choice of a non-targeted, shortened intervention was driven by the need to have

tools that can be easily integrated into the curriculum. The results of this initial investigation

of a CD intervention with a younger age group are promising. They findings clearly need

replication and future research should evaluate the longer term impact of the CD intervention.

However, these results add to the evidence that relatively short CD interventions are effective

for adolescent girls.

9
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

References

Becker, C. B., Smith, L. M., & Ciao, A. C. (2006). Peer-facilitated eating disorder prevention: A

randomized effectiveness trial of cognitive dissonance and media advocacy. Journal of

Counseling Psychology, 53, 550-555.

Berschied, E., Walster, E., & Bohrnstedt, G. (1973, November). The happy American body: A

survey report. Psychology Today, 7(6), 119-131.

Daee, A., Robinson, P., Lawson, M., Turpin, J.S., Gregory, B., & Tobias, J.D. (2002). Psychologic

and physiologic effects of dieting in adolescents. Southern Medical Journal, 95, 1032-1041.

Grabe, S., Ward, L.M., & Hyde, J.S. (2008). The role of the media in body image concerns among

women: A meta-analysis of experimental and correlational studies. Psychological Bulletin,

134, 460-476.

Holt, K. E., & Ricciardelli, L. A. (2008). Weight concerns among elementary school children: A

review of prevention programs. Body Image, 5, 233-243.

Levine, M.P., & Murnen, S.K. (2009). “Everybody knows that mass media are/are not [pick one] a

cause of eating disorders”. A critical review of evidence for a causal link between media,

negative body image, and disordered eating in females. Journal of Social and Clinical

Psychology, 28, 9-42.

Matusek, J.A., Wendt, S.J., & Wiseman, C.V. (2004). Dissonance thin-ideal and didactic healthy

behavior eating disorder prevention programs: Results from a controlled trial. International

Journal of Eating Disorders, 36, 376-388.

Neumark-Sztainer, D., Paxton, S.J., Hannan, P.J., Haines, J. & Story, M. (2006). Does body

dissatisfaction matter? Journal of Adolescent Health, 39, 244-251.

Roehrig, M., Thompson, J.K., Brannick, M., & van den Berg, P. (2006). Dissonance-based eating

disorder prevention program: A preliminary dismantling investigation. International Journal

of Eating Disorders, 39, 1-10.

10
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Schafer, J.L., & Graham, J.W. (2002). Missing data: Our view of the state of the art. Psychological

Methods, 7, 147-177.

Evans, E.H., Tovée, M.J., Boothroyd, L. G., & Drewett, R.F. (in press). Body dissatisfaction and

disordered eating attitudes in 7- to 11-year-old girls: Testing a sociocultural model. Body

Image (2012), http:dx.doi.org/10.1016/j/bodyim.2012.10.001.

Stice, E., & Bearman, K. (2001). Body-image and eating disturbances prospectively predict increases

in depressive symptoms in adolescent girls. Developmental Psychology, 37, 507-607.

Stice, E., Shaw, H., Burton, E., & Wade, E. (2006). Dissonance and healthy weight eating disorder

prevention programs. Journal of Consulting and Clinical Psychology, 74, 263-275.

Stice, E., & Presnell, K. (2007). The body project: Promoting body acceptance and preventing

eating disorders (Facilitator Manual). New York: Oxford University Press.

Stice, E., Cooper, J, Al, Schoeller, D. A., Tappe, K., & Lowe, M.R. (2007). Are dietary restraint

scales valid measures of moderate- to long-term dietary restriction? Objective biological and

behavioural data suggest not. Psychological Assessment, 19, 4, 449-458.

Stice, E., Rohde, P., Shaw, H., & Gau, J. (2011). An effectiveness trial of a selected dissonance-

based eating disorder prevention program for female high school students: Long-term effects.

Journal of Consulting and Clinical Psychology, 79, 500-508.

Stice, E., Shaw, H., & Marti, C. N. (2007). A meta-analytic review of eating disorder prevention

programs: Encouraging findings. Annual Review of Clinical Psychology, 3, 207-231.

Stunkard, A.J., & Messick, S. (1988). The Eating Inventory. San Antonio: Psychological Corp.

Van Strien, T., Frijiters, J.E., Bergers, G.P.A., & Defares, P.B. (1996). The Dutch Eating Behavior

Questionnaire for assessment of restrained, emotional, and external eating behavior.

International Journal of Eating Disorders, 19, 1, 83-92.

Wertheim, E & Paxton, S.J. ( 2011). Body image development in adolescent girls. p. 76-84, In Cash,

T., & Smolak, L. (eds.). Body Image: A handbook of science, practices and prevention.

11
 Running head: Brief Report: Testing a Dissonance Body Image Intervention

Table 1: The content of the intervention sessions

1 1. Introduction to the program and voluntary commitment

2. Defining and exploring the origins of the thin-ideal of female beauty

3. Homework: Mirror exercise.

2 1. Review of the homework

2. Cost of pursuing the thin ideal. Due to the age of the participants this discussion

was limited to the costs of the thin ideal to themselves and other girls, and the

instructions were rephrased to ask ‘what negative things might happen to

someone trying to look like the thin ideal?’

3. Homework: Participants were asked to write a letter to a teenage girl who is

struggling with her body image to persuade the girl not to pursue the thin-ideal.

3 1. Review of the homework.

2. The session focused on critiquing beauty ideals and beauty myths. Participants

watched a short (1 minute) film called Evolution, produced by Dove to

demonstrate airbrushing.*

3. Appearance facts or fiction exercise. Participants review whether 6 statements

were generally regarded as fact or fiction. Five of the statements reflected

dominant beauty myths, (e.g., ‘Most models and actresses are a healthy weight’,

‘Attractive people are happier than less attractive people’), and one focused on the

media (‘Looking at fashion magazines can make girls and women feel bad about

themselves’). They were then given evidence challenging each beauty myth and

supporting the media effect and asked to reassess their original answers.*

4 1. Role-play to resist pressure to pursue the thin ideal. The facilitator took the role of

a severe dieter and participants attempted to dissuade her from dieting

*Additional activities

12