Ac

MK

a

ARRA

KFBBDE

swutwinbuiw

wMa(i(

B

(

h1

Body Image 11 (2014) 337–345

Contents lists available at ScienceDirect

Body Image

journa l h om epa ge: www.elsev ier .com/ locat e/bodyimage

naturalistic study of fat talk and its behavioral and affectiveonsequences

ichelle D. Jones ∗, Janis H. Crowther, Jeffrey A. Cieslaent State University, Kent, OH, USA

r t i c l e i n f o

rticle history:eceived 25 September 2013eceived in revised form 30 May 2014ccepted 30 May 2014

eywords:at talk

a b s t r a c t

Fat talk is a style of verbal expression among young women involving negative self-statements, com-plaints about physical appearance, and weight management. This research used ecological momentaryassessment to examine the impact of naturalistic fat talk experiences on body dissatisfaction, bodychecking, negative affect, and disordered eating behaviors. We examined trait self-objectification as amoderator. Sixty-five female college students completed a baseline questionnaire and responded to ques-tions when randomly prompted by palm pilot devices for five days. Results indicated fat talk is common

ody dissatisfactionody checkingisordered eatingMA

and associated with greater body dissatisfaction, body checking, negative affect, and disordered eatingbehaviors. Fat talk participation was associated with greater body checking than overhearing fat talk.Greater trait self-objectification was associated with greater body dissatisfaction and body checking fol-lowing fat talk. These results suggest that fat talk negatively impacts the cognitions, affect, and behaviorof young women and has increased negative effects for women higher in self-objectification.

Introduction

Over the past two decades, a distinct style of verbal expres-ion has become prevalent among adolescent and young adultomen. This form of expression, referred to as “fat talk”, can besed to describe a highly ritualized conversation involving nega-ive self-statements, complaints about physical appearance, andeight management (Nichter & Vuckovic, 1994). Fat talk often

nvolves self-statements about being fat or overweight, even whenot objectively true (Clarke, Murnen, & Smolak, 2010). It may alsoe accompanied by statements regarding behaviors women mightse to change their bodies, such as dieting or exercise. Examples

nclude: “I’m so fat,” “My legs are huge,” and “I’ve gained too mucheight. I should really start going to the gym.”

Research suggests that fat talk is much more common amongomen than men (Martz, Petroff, Curtin, & Bazzini, 2009; Payne,artz, Tompkins, Petroff, & Farrow, 2011) and more common

mong adolescent and young adult women than older women

Martz et al., 2009). The practice of engaging in fat talk often beginsn middle school, but persists into the high school and college yearsNichter, 2000). Women report more perceived pressure to engage

∗ Corresponding author at: Department of Psychology, Kent State University, P.O.ox 5190, Kent, OH 44242, USA. Tel.: +1 330 672 7777x22344; fax: +1 330 672 3786.

E-mail addresses: mthedfor@kent.edu (M.D. Jones), jcrowthe@kent.eduJ.H. Crowther), jciesla@kent.edu (J.A. Ciesla).

ttp://dx.doi.org/10.1016/j.bodyim.2014.05.007740-1445/© 2014 Elsevier Ltd. All rights reserved.

© 2014 Elsevier Ltd. All rights reserved.

in fat talk compared to men, regardless of their level of body dis-satisfaction, and it is well recognized as a norm for females inadolescent and young adult populations (Britton, Martz, Bazzini,Curtin, & LeaShomb, 2006). Women with eating disorders engagein fat talk more frequently than their non-eating-disordered coun-terparts (Ousley, Cordero, & White, 2008), suggesting a relationshipbetween fat talk, body image disturbance, and disordered eatingpractices.

Although Nichter (2000) suggested that fat talk may havepositive functions, such as building and maintaining social rela-tionships, creating and maintaining group affiliation, and providingpositive feedback and reassurance to others, laboratory researchindicates that exposure to fat talk primarily has a negative impact.In a study by Stice, Maxfield, and Wells (2003), participants engagedin a pre-scripted conversation with an attractive female confed-erate, during which the confederate either engaged in fat talkor discussed a neutral topic. Participants exposed to the fat talkconversation experienced significant increases in body dissatisfac-tion from pre- to post-conversation, whereas participants exposedto the neutral topic experienced no significant change in bodydissatisfaction. Neither group experienced significant change innegative affect, nor did trait thin-ideal internalization, body dis-satisfaction, or social support moderate the relationship between

fat talk and body dissatisfaction or negative affect (Stice et al.,2003).

In a later study, Tucker, Martz, Curtin, and Bazzini (2007) video-taped participants during an interview with a confederate research

3 y Imag

ameiadMeaeealrwct

tobtoctpdssabaffplp

iMeLhatatnaavsfwt

O

t&otr(t

38 M.D. Jones et al. / Bod

ssistant who provided either a negative, neutral, or positive com-ent about her own body. While participants with lower trait body

steem had significantly lower body satisfaction ratings, of majornterest to this research were findings that participants in the neg-tive body talk condition had the lowest body satisfaction ratingsuring the interview (Tucker et al., 2007). Finally, Salk and Engeln-addox (2012) assessed participant responses following fat talk

xposure from two confederates while viewing a picture of a thin,ttractive female model in a bikini. Results found overall negativeffects of fat talk, with participants exposed to two confederatesngaging in fat talk providing higher ratings of body dissatisfactionnd guilt compared to a condition in which a confederate chal-enged fat talk and a control condition characterized by neutralesponses. Participants were also more likely to engage in fat talkhen exposed to others participating in fat talk and women who

hose to engage in fat talk had higher trait body dissatisfaction thanhose who did not engage in fat talk.

Only one study to date has examined the effects of fat talk longi-udinally. Arroyo and Harwood (2012) conducted a study focusedn the communication mechanisms underlying fat talk to exploreoth the causes and outcomes of engaging in fat talk. Participants inhe first part of this two-part study completed frequency measuresf fat talk at six time points over a three week span in addition toompleting several psychological measures at the first and last ofhese time points. They found that fat talk frequency at time oneredicted decreased levels of body satisfaction and higher levels ofepression at time six, but found no evidence that fat talk impactself-esteem or sociocultural pressure to be thin. Participants in theecond part of this study completed the same fat talk frequencynd psychological measures as those in the first part of the study,ut completed all measures at two time points spaced two weekspart. Participants indicated how frequently they themselves madeat talk comments as well as how frequently they heard others useat talk comments. Increased frequency of using fat talk at time oneredicted higher levels of depression and pressure to be thin at the

ater time point and hearing fat talk at time one predicted highererceived pressure to be thin at time two.

To summarize, exposure to fat talk is associated with increasesn body dissatisfaction (Arroyo & Harwood, 2012; Salk & Engeln-

addox, 2012; Stice et al., 2003; Tucker et al., 2007) and negativemotional states (Arroyo & Harwood, 2012; Gapinski, Brownell, &aFrance, 2003; Salk & Engeln-Maddox, 2012). For those womenigher in risk for body dissatisfaction, fat talk likely has greater neg-tive consequences. However, these studies have not investigatedhe behavioral consequences of fat talk, including body checkingnd eating and weight control behaviors, and findings regardinghe consequences of exposure to fat talk in laboratory studies mayot generalize to the naturalistic environment. The limited researchvailable to date has not found support for trait body dissatisfactions a moderator of the relationship between fat talk and outcomeariables; however, given the centrality of appearance to women’self-worth in conjunction with the appearance-focused nature ofat talk, self-objectification theory may provide a useful frame-ork for understanding the function and effects of engaging in fat

alk.

bjectification Theory

Fat talk has been described as a “social extension of body objec-ification” (Martz et al., 2009). Objectification theory (Fredrickson

Roberts, 1997) suggests that women’s bodies are sexualized byur society and culture and by men as a whole. This sexual objec-

ification involves viewing a woman’s body or sexual functions asepresentative of her person. According to Fredrickson and Roberts1997), sexual objectification of women exists in the media, cul-ure, and everyday situations. They argue that the current culture

e 11 (2014) 337–345

socializes women to view and value themselves based on theirphysical appearance and to treat themselves, to some extent, asobjects. Self-objectification occurs when this practice becomesincorporated into a woman’s self-concept. The tendency for womento self-objectify often begins during or shortly after puberty andcontinues throughout the lifespan (Fredrickson & Roberts, 1997).Although the extent to which an individual experiences self-objectification varies, Fredrickson and Roberts (1997) conclude itis a common experience.

Self-objectification has many consequences that affect women’smental health and psychological well-being (Fredrickson & Roberts,1997). When individuals do not measure up to an internal-ized or cultural ideal or standard, they may experience shame(Muehlenkamp & Saris-Baglama, 2002). Many women also expe-rience appearance-related anxiety regarding the evaluation oftheir bodies by others, particularly when they are uncertain ofhow and when their bodies will be evaluated (Muehlenkamp& Saris-Baglama, 2002). When a woman becomes self-consciousas a result of objectification, whether self- or otherwise, herconcentration may be derailed, interfering with her ability tofully engage in productive and rewarding activity (Fredrickson &Roberts, 1997). Fitting in with the standards of the self and cul-ture often requires ignoring internal cues, such as disregardinghunger signals during periods of strict dieting, which may resultin a lack of awareness of internal bodily states (Fredrickson &Roberts, 1997; Myers & Crowther, 2008). Self-objectification is alsobelieved to contribute to depression, sexual dysfunction, and eat-ing disorders (e.g., Miner-Rubino, Twenge, & Fredrickson, 2002;Muehlenkamp & Saris-Baglama, 2002; Myers & Crowther, 2007;Noll & Fredrickson, 1998; Slater & Tiggemann, 2002; Tiggemann& Kuring, 2004; Tylka & Hill, 2004). Given that women whoexperience self-objectification have a greater focus on physicalappearance (Fredrickson & Roberts, 1997), it seems likely that fattalk would have more negative effects among women who engagein self-objectification.

Gapinski et al. (2003) examined the impact of fat talk expo-sure on women in objectifying situations by having participantsoverhear a confederate engaging in fat talk while trying on aswimsuit. These effects were compared to the experience of indi-viduals in a control condition, who were instructed to try on asweater rather than a swimsuit (Gapinski et al., 2003). As hypothe-sized, participants in the objectifying condition experienced higherlevels of negative emotional states compared to those in the non-objectifying condition. Somewhat surprisingly, for women in theobjectifying condition, hearing fat talk resulted in less negativeemotion than when the confederate made neutral comments. Incontrast, women in the control condition experienced greater neg-ative emotion when the confederate used fat talk compared tooverhearing neutral comments. Although fat talk appears to have anegative impact on women’s affective state at times, Gapinski et al.(2003) suggest that it may serve as a protective factor when womenare in objectifying situations (e.g., trying on a swimsuit in a fittingroom) because fat talk seems normative and appropriate and mayfeel reassuring in that context.

Present Study

Despite the prevalence of fat talk (Salk & Engeln-Maddox, 2011),the negative implications that fat talk has for women, particularlyadolescent girls and college students, and the context within whichit seems to occur, very little empirical research has examined thenature of this type of conversation and its emotional and behav-

ioral consequences. While the laboratory research conducted inthis area maximizes internal validity, thus enabling researchersto draw conclusions regarding causality, Stice et al. (2003) notedthat their experimental condition might have been too subtle in

y Imag

ipatifoueaatf2

inRcditctamncotaimoremattafwo

sdioteermitb(tsbtnawo

which a participant indicated no fat talk experiences, a “not appli-cable” option was available. These and other questions whose data

M.D. Jones et al. / Bod

ts fat talk presentation to elicit some of the hypothesized effects,articularly in relation to effects of fat talk exposure on negativeffect. Additionally, laboratory research lacks ecological validity,hus highlighting the importance of examining how fat talk andts consequences unfold in daily life. Only one study has assessedat talk longitudinally (Arroyo & Harwood, 2012). However, theynly assessed fat talk outcomes as baseline and post-study meas-res and did not utilize repeated measures of these variables toxamine more nuanced outcomes of fat talk exposure. Moreover,lthough research to date has focused on body dissatisfaction andffect, it is important to measure behavioral consequences so aso better inform our understanding of the relationships betweenat talk and disordered eating (Clarke et al., 2010; Ousley et al.,008).

To our knowledge, there are no studies of the frequency andmpact of fat talk in the naturalistic environment. In their exami-ation of body talk and body co-rumination among close friends,udiger and Winstead (2013) found that body talk and bodyo-rumination were both associated with body image cognitiveistortions and disordered eating. These results, which examined

nteractions that would not typically occur in a laboratory, pointo the need for evaluation of this highly ritualized communi-ation in a naturalistic setting. Stice et al. (2003) noted that inheir study, post-conversation measures were completed withinn hour of the exposure, and significant changes in negative affectay take a longer period of time to surface. This highlights the

eed for multiple daily assessments of affective and psychologicalonsequences of fat talk, including behavioral outcomes. Previ-us research has focused largely on experimental manipulationshat passively exposed women to “fat talk,” often by confeder-tes. Yet, there may be different consequences to experiencesn which women actively participate in fat talk as opposed to

erely hearing it, as implicit in active participation is commentaryn one’s own physical appearance which may trigger weight-elated concerns and behaviors. This suggests the importance ofxamining the impact of active participation in fat talk versuserely overhearing it. Finally, despite Gapinski et al.’s (2003)

ssertion that self-objectification may serve as a protective fac-or in some situations involving fat talk, it seems more likelyhat greater trait self-objectification would intensify the immedi-te negative outcomes associated with fat talk exposure becauseat talk draws attention to perceived flaws in physical appearance,hich may be central to the self-worth of individuals high on self-

bjectification.The purpose of this research was to better describe and under-

tand the nature and consequences of fat talk as it occurs on aay-to-day basis among college women. Specifically, the present

nvestigation sought to determine the frequency with which fat talkccurs in the naturalistic environment and to examine the affec-ive and behavioral consequences of fat talk. This study employedcological momentary assessment (EMA) to assess fat talk experi-nces as they occur (Stone & Shiffman, 1994), thus reducing biaselated to retrospective recall. Moreover, the use of multi-levelodeling allows analysis of non-aggregated data, thus maximiz-

ng variability. We hypothesized that (1) recent exposure to fatalk would be associated with greater state body dissatisfaction,ody checking, negative affect, and disordered eating behaviors,2) episodes of participating in fat talk (i.e., initiating or respondingo fat talk statements) would be associated with greater body dis-atisfaction, body checking, negative affect, and disordered eatingehaviors than episodes of overhearing fat talk, and (3) the rela-ionships between fat talk and body dissatisfaction, body checking,egative affect, and disordered eating behaviors would be moder-ted by trait self-objectification, such that the impact of fat talk

ould be more negative among individuals higher in trait self-

bjectification.

e 11 (2014) 337–345 339

Method

Participants

Participants were 67 female college students (MAge = 21.1;SDAge = 7.61) who volunteered via an online departmental researchwebsite to participate in a study about thoughts and feelings relatedto conversations regarding “fat talk” experiences. To be eligible,women had to be 18 years of age or older. The majority of the samplewas Caucasian (82.4%), and the remaining participants identifiedthemselves as African-American (9.8%) or Asian (7.8%). On average,the sample had a body mass index (BMI) of 24.21 (SD = 4.28).

Apparatus

Palm Centro Personal Data Assistant (PDA). All participantswere given a Palm Centro PDA programmed to sound an alarmwhenever they were to complete a series of questionnaires. Sim-ilar apparatus have been employed in other studies utilizingEMA (Colautti, Fuller-Tyszkiewicz, Skouteris, McCabe, Blackburn,& Wyett, 2011; Heron & Smyth, 2013; Myers, Ridolfi, Crowther, &Ciesla, 2012; Ridolfi, Myers, Crowther, & Ciesla, 2011). The meas-ures were programmed into the PDA using the Purdue MomentaryAssessment Tool (PMAT; Weiss, Beal, Lucy, & MacDermid, 2004)software so that participants could complete the questionnairesdirectly on the PDA device.

Measures

Baseline questionnaire

Trait self-objectification. The Self-Objectification Question-naire (SOQ; Noll & Fredrickson, 1998) measures tendency towardself-objectification. Participants rank a list of 10 body attributesbased on impact on physical self-concept from greatest (rank = 1)to least (rank = 10). To score this measure, each item is assigneda value based on the participant’s rank (i.e., an item ranked firstreceived a value of 9, an item ranked second received a value of8, etc.), with each item receiving a value between 0 and 9. Twosubscales are calculated by separately summing five appearance-focused items and five competence-focused items. Total SOQ scoresare then calculated by subtracting the competence-focused sub-scale from the appearance-focused subscale. Possible scores rangefrom −25 to 25, with higher scores representing greater self-objectification. Consistent with previous research, we are notreporting internal consistency for the SOQ (e.g., Lindner, Tantleff-Dunn, & Jentsch, 2012; Slater & Tiggemann, 2012; Van Diest &Perez, 2013; Vandenbosch & Eggermont, 2013). Scores on the SOQpositively correlate with measures of body-size dissatisfaction andphysical appearance anxiety (Noll & Fredrickson, 1998).

Palm pilot measures

Fat talk questions. Participants were initially asked, “Since thelast alarm, have you overheard fat talk?” and, if so, “How manytimes have you heard fat talk since the last alarm?” They were alsoasked, “Since the last alarm, have you engaged in fat talk?” and, if so,“How many times have you engaged in fat talk since the last alarm?”Participants were then instructed to think about the most salientinstance of fat talk since the last alarm and asked to indicate whomade the initial comment from the following options: self, familymember, friend, acquaintance, and stranger. For those prompts in

are not reported here were chosen based on pilot data collectedthrough structured interviews with undergraduate women.

3 y Image 11 (2014) 337–345

twauco

(tsAtatmss1

RmepermPaegTtqu

–maeLTt

a

ticinatdtt

bdye

ipat

ing

and

over

hea

rin

g

in

fat

talk

, an

d

mod

erat

ing

effe

cts

of

trai

t

self

-obj

ecti

fica

tion

on

bod

y

dis

sati

sfac

tion

, bod

y

chec

kin

g,

neg

ativ

e

affe

ct, a

nd

dis

ord

ered

eati

ng

beh

avio

rs.

tisf

acti

on

Bod

y

chec

kin

g

Neg

ativ

e

affe

ct

Wei

ght

con

trol

and

dis

ord

ered

eati

ng

beh

avio

rs

SD

=

4.50

)

M

=

18.7

6

(SD

=

7.69

)

M

=

13.3

2

(SD

=

3.61

)

M

=

0.33

(SD

=

0.35

)

t(df

)

p

ˇ

t(df

)

p ˇ

t(df

)

p

ˇ

t(df

)

p

28.7

71

(63)

<.00

1

18.0

03

18.9

17

(63)

<.00

1

12.7

58

31.6

78

(63)

<.00

1

0.26

8

6.20

5

(63)

<.00

11.

996

(991

)

.046

1.60

8

5.61

7

(991

) <.

001

0.81

1

2.99

5

(991

)

.003

0.13

3

4.11

2

(991

)

<.00

1

28.9

24

(63)

<.00

1

18.4

38

19.2

02

(63)

<.00

1

12.9

23

33.3

17

(63)

<.00

1

0.29

2

6.87

4

(63)

<.00

11.

380

(990

)

.168

−0.1

26

−0.3

28

(990

)

.743

0.01

8

0.05

1

(990

)

.959

0.06

0

1.36

4

(990

)

.173

0.59

4

(990

)

.553

1.15

6

2.88

1

(990

)

.004

0.53

9

1.42

3

(990

)

.155

0.06

5

1.40

6

(990

)

.160

29.1

49

(62)

<.00

1

18.0

43

19.3

83

(62)

<.00

1

12.7

62

34.1

85

(62)

<.00

1

0.24

9

6.16

1

(62)

<.00

11.

787

(62)

.079

1.55

4 4.

717

(62)

<.00

1

0.77

4

2.81

3

(62)

.007

0.14

7

3.12

0

(62)

.003

1.46

2

(62)

.149

0.20

1 2.

872

(62)

.006

0.05

6

1.99

0

(62)

.051

0.00

3

0.86

3

(62)

.392

2.03

5

(62)

.046

0.05

4

2.10

2

(62)

.040

0.03

8

1.75

6

(62)

.084

0.00

5

1.26

2

(62)

.212

ten

cies

for

the

dep

end

ent

vari

able

s,

we

sele

cted

the

firs

t

pro

mp

t

for

wh

ich

a

par

tici

pan

t

com

ple

ted

the

self

-rep

ort

mea

sure

s.

Cro

nba

ch’s

alp

has

wer

e

as

foll

ows:

SSES

=

.87,

BC

=

.85,

and

NA

=

.89.

the

very

brie

f tim

e

per

iod

cove

red

by

any

one

asse

ssm

ent,

inte

rnal

con

sist

ency

was

not

calc

ula

ted

for

the

wei

ght

con

trol

and

dis

ord

ered

eati

ng

vari

able

.

40 M.D. Jones et al. / Bod

Weight control and disordered eating behaviors (EB). Par-icipants were asked to indicate if they engaged in the followingeight control and disordered eating behaviors since the last

larm: exercise, diet restriction, self-induced vomiting, laxativese, diuretic use, and binging. These six items were summed toreate a disordered eating behaviors variable, which reflected theccurrence of behaviors that occurred since the last alarm.1

State body dissatisfaction (BD). The State Self-Esteem ScaleSSES; Heatherton & Polivy, 1991) is a self-report measurehat assesses momentary self-esteem. The SSES contains threeubscales: Performance, Social, and Appearance. The 6-itemppearance subscale was chosen to assess state body dissatisfac-

ion in this study. Participants indicate how true an item is for themt this time using a 5-point Likert scale ranging from 1 = not at allo 5 = extremely. This subscale has been shown to correlate with

easures of dietary restraint, satisfaction with current figure, bodyize estimation, depression, and global self-esteem, and the overallcale shows good internal consistency ( ̨ = .92; Heatherton & Polivy,991).

Body checking (BC). The Body Checking Questionnaire (BCQ;eas, Whisenhunt, Netemeyer, & Williamson, 2002) is a self-reporteasure in which participants indicate how frequently they have

ngaged in checking the size, shape, and appearance of various bodyarts. The original 23-item scale shows good psychometric prop-rties, including internal consistency ( ̨ = .83–.92) and test–retesteliability (r = .94). A brief version consisting of the 10 most com-only endorsed items ( ̨ = .89) has been successfully adapted for

alm Pilot use in previous studies (see Ridolfi et al., 2011 for description of the adaptation and enumeration of the itemsmployed). For example, items assess checking one’s reflection inlass doors or mirrors or pinching body parts to measure fatness.he brief version was selected for use during the PDA portion ofhis study. Participants responded to each item, indicating how fre-uently they engaged in the behavior since the last alarm soundedsing a 5-point Likert scale ranging from 1 = never to 5 = very often.

Negative affect (NA). The Positive and Negative Affective Scale Extended Form (PANAS-X; Watson & Clark, 1994) is a self-reporteasure with good psychometric properties used to assess state

ffect. Participants indicate how strongly they are currently experi-ncing a feeling or emotion that is presented to them using a 5-pointikert scale ranging from 1 = very slightly or not at all to 5 = extremely.he Negative Affect subscale, which totals 10 items, was chosen forhe purposes of this study.

See Table 1 for internal consistencies for the SSES, BC, and Neg-tive Affect subscale of the PANAS-X.

Procedure

Participants registered for two separate appointments usinghe university’s online psychological research website. During thenitial appointment, participants provided informed consent andompleted the SOQ. Participants were provided with the follow-ng definition of fat talk, “Fat talk refers to conversations involvingegative-self statements, complaints about physical appearance,nd weight management.” Participants also were informed thathe purpose was to understand how “fat talk” occurs in every-

ay life and its effects on thoughts and behaviors. After completinghe SOQ, participants were provided with a Palm Centro PDA andrained on its use. They then completed daily assessments on the

1 One question that arises is whether the weight control and disordered eatingehaviors variable should include exercise. We conducted all analyses with thisependent variable with and without the exercise item included. Because the anal-ses yielded identical findings, we chose to report the analyses that include thexercise item in the weight control and disordered eating behaviors variable. Ta

ble

1Ef

fect

s

of

fat

talk

exp

osu

re, p

arti

c

Bod

y

dis

sa

M

=

16.3

0

(

ˇ

Effe

cts

of

FTIn

terc

ept

16.2

03

FT

0.39

9

Part

icip

atio

nIn

terc

ept

16.2

36

Ove

rhea

rd

0.36

1

Part

icip

ated

0.16

3

Mod

erat

ion

Inte

rcep

t

16.1

75

Fat

talk

0.41

5

SO

0.06

1

FT

×

SO

0.03

7

Not

e:

To

calc

ula

te

inte

rnal

con

sis

Giv

en

the

nat

ure

of

the

item

s

and

y Imag

Pfitstspstataunfndpatdscta

riaprrsttd(

s[sBsbeeabwLiwpctfutL1nt

participating in the fat talk 80.0% of the time and overhearing fattalk by others 74.2% of the time. Participants’ responses to ques-tions assessing the nature of the most salient experiences of fat

M.D. Jones et al. / Bod

DA in response to 25 random signals over the next five days, withve random signals occurring per day. Signals occurred betweenhe hours of 8:30 am and 11:30 pm, and there was a minimumpacing of two hours between random signals. Participants had upo 10 minutes to respond to the initial signal when each randomignal sounded before the signal was considered missed and partici-ants were no longer able to respond to questionnaires. When theseignals occurred, participants responded to several questions per-aining to their experiences of fat talk since the last alarm, questionsssessing binge eating and compensatory weight control strategies,he Appearance subscale of the SSES, the brief version of the BCQ,nd the Negative Affect subscale of the PANAS-X. Participants couldse a do not disturb function, which allowed them to delay any sig-als that were scheduled to occur for up to an hour at a time. This

unction was provided to avoid having participants respond to sig-als when it would be disruptive or unsafe to do so (e.g., whenriving, in class, working). Following the completion of the PDAortion of the study, participants attended their second scheduledppointment and responded to four open-ended follow-up ques-ions assessing problems that may have occurred with the PDAevice, changes in thoughts and behaviors during the course of thetudy, and suggestions for changes to the study. Participants wereontacted via e-mail following their second appointment to notifyhem about their compliance during the PDA portion of the studynd their resulting compensation.

Participants received research credit toward their psychologyesearch requirements for taking part in this study. In order toncrease compliance rates, participants also received entries into

raffle for an iPad based on their level of compliance. Partici-ants who responded to more than 75% of signals received fouraffle entries, participants who responded to 51–75% of signalseceived two raffle entries, participants who responded to 25–50%ignals received one raffle entry, and participants who respondedo less than 25% of signals received no raffle entries in addition toheir research points. Similar methods of compensation have beenescribed to increase compliance in previous naturalistic studiesMyers et al., 2012; Ridolfi et al., 2011).

Statistical Analyses

We used multilevel modeling (MLM) to analyze the relation-hip between fat talk and its affective and behavioral consequencesi.e., body dissatisfaction (BD) as measured by the Appearance sub-cale of the SSES, body checking (BC) as measured by the briefCQ, negative affect (NA) as measured by the Negative Affect sub-cale of the PANAS-X, and weight control and disordered eatingehaviors (EB)] and to examine whether this relationship was mod-rated by trait self-objectification. In MLM, intra-individual Level-1quations are used to model within-subject relationships. For thesenalyses, Level-1 equations were used to examine the associationsetween fat talk and its consequences as well as whether thereere differences between participating in and overhearing fat talk.

evel-2 equations use between-subject predictors to model inter-ndividual differences in the Level-1 relationships. In this analysis

e examined trait self-objectification as a Level-2 variable, whichermits testing self-objectification as a moderator of the fat talk-onsequences relationships. MLM is appropriate for use with longi-udinal data and allows for analysis of multiple time points of dataor each participant (Bryk & Raudenbush, 1992), making it ideal forse with EMA data. The data were examined for missing data prioro running analyses. MLM analyses can handle missing data in the

evel-1 dataset, but not the Level-2 dataset (Bryk & Raudenbush,992). As a result of participants not responding to random sig-als, missing data were present in the Level-1 dataset; however,hese data were left as missing for analysis. When participants

e 11 (2014) 337–345 341

responded to the initial random prompt, only 0.53% of items usedin analyses were missing. In the Level-2 dataset, one participant didnot respond to any items on the SOQ. This participant’s data wereremoved from the Level-2 dataset that included SOQ. Because ofthe removal of this participant’s data, her data at Level-1 were notincluded in any analyses that included SOQ as a Level-2 variable.This participant’s data were not excluded from any other analyses.

Results

Quality Analyses

To determine participant cooperation with the study require-ments, we computed compliance rates. The maximum number ofcompleted entries possible was 25, as participants could be sig-naled up to five times per day by the PDA over the course of the5-day study. Participants responded to an average of 16.28 sig-nals (SD = 4.20), with number of responses ranging from 3 to 25for each participant. Two participants responded to less than 25%of random signals (12% and 16% respectively). Based on previousresearch (Ridolfi et al., 2011), this response rate was determinedto be inadequate for analysis. Thus, we excluded these two par-ticipants from further analyses, leaving 65 participants who wereincluded in all statistical analyses. Of the 1610 random signals, 1058were completed entries, 441 were skipped entries, and 111 weredo not disturb entries. We considered entries missed due to partic-ipants utilizing the do not disturb feature on the PDA device to bein compliance with the study protocol because participants werenot ignoring or missing the prompt. Thus, we calculated compli-ance rates by dividing the sum of the completed entries and thedo not disturb entries by the total number of entries, resulting ina compliance rate of 72.6% for the 65 participants included in theanalyses.

We assessed reactivity over the 5-day course of the palm pilotportion of the protocol by using the day of the study to predict thefour outcome variables in a series of MLM analyses. We used thefollowing equation for these analyses2:

Yti = ˇ00 + ˇ10(Day)ti + r0i + eti

Results yielded a statistically significant negative coefficient forthe slope of day on BC, indicating a significant decrease in bodychecking (ˇ10 = −0.367, SE = 0.072, t(1496) = −5.074, p < .001) overthe course of the study. The results for body dissatisfaction, neg-ative affect, and weight control and disordered eating behaviorswere nonsignificant [BD: ˇ10 = 0.006, SE = 0.051, t(1496) = 0.121,p = .903; NA: ˇ10 = −0.126, SE = 0.068, t(1496) = −1.837, p = .066; EB:ˇ10 = −0.004, SE = 0.009, t(1496) = 1.380, p = .671].

Aggregate and Descriptive Analyses

Descriptive statistics indicated that 96.9% of participantsreported experiencing at least one fat talk episode during the courseof the study, with the remaining 3.1% reporting no fat talk experi-ences during the study. Total number of fat talk episodes reportedby participants over the course of five days ranged from zero to27. On average, participants experienced 9.92 episodes of fat talk(SD = 6.89) during the five days of the study. When they reportedexperiencing at least one fat talk episode, participants indicated

2 The terms for this equation are defined as follows: Yti = the dependent out-come variable (BD, BC, NA, or EB), ˇ00 = the intercept, ˇ10 = the slope of day on thedependent variable, r0i = the level-2 residual, and eti = the Level-1 residual.

3 y Imag

tte[obotortip

ECD

ssv(pwsw

mattl1ale

Y

tsttp(e

E

wfdfasal1o

vw

tionship between fat talk exposure and BD (ˇ11 = 0.037, SE = 0.018,t(62) = 2.035, p = .046) and BC (ˇ11 = 0.054, SE = 0.026, t(62) = 2.102,p = .040) such that for every unit increase in SO, fat talk was asso-ciated with greater BD and BC. SO did not interact with fat talk

4

42 M.D. Jones et al. / Bod

alk during the study revealed that participants initiated 34.5% ofhese experiences. The remaining 65.5% of the most salient fat talkxperiences were initiated by someone other than the participanti.e., a family member (13.1%), friend (46.1%), acquaintance (2.7%),r stranger (3.6%)]. To examine if participants’ body dissatisfaction,ody checking, negative affect, and weight control and eating dis-rdered behaviors differed as a function of the initiator of the fatalk experience, we aggregated the data by initiator (self versusther) across the course of the study. For those participants whoeported both self- and other-initiated fat talk experiences, paired-tests yielded no significant differences between self- and other-nitiated episodes on BD, t(34) = −1.303, p = .201, BC, t(35) = 0.194,

= .848, NA, t(35) = 0.497, p = .622, and EB, t(35) = 0.080, p = .936.

ffect of Fat Talk Experience on Body Dissatisfaction, Bodyhecking, Negative Affect, and Weight Control andisordered Eating Behaviors

Prior to conducting analyses to test study hypotheses, foureparate unconditional models were conducted, which yieldedtatistically significant variance components for all four outcomeariables, indicating that it was appropriate to add Level-1 variablesBryk & Raudenbush, 1992). Because study day was a significantredictor of frequency of body checking, we conducted analysesith and without day as a Level-1 predictor. Given that the inclu-

ion of day did not change any findings, analyses are reportedithout the inclusion of day in the equation.

To examine Hypothesis 1, we conducted four separate MLModels, one for each outcome variable. For these analyses, we

dded a dummy-coded fat talk variable to the Level-1 equation. Fatalk exposure was operationally defined as an experience in whichhe participant participated in fat talk or overheard fat talk since theast alarm. This fat talk variable (FT) was coded such that a value of

indicated experiencing one or more fat talk episodes since the lastlarm and a value of 0 indicated not experiencing fat talk since theast alarm. The reference group for these analyses was no fat talkxperience. We used the following equation for these analyses3:

ti = ˇ00 + ˇ10(FT)ti + r0i + eti

Statistically significant positive coefficients for the slope of fatalk for BD, BC, NA, and EB indicated a main effect for expo-ure to fat talk. Results indicate that participants exposed to fatalk experienced significantly greater BD (ˇ10 = 0.399, SE = 0.200,(991) = 1.996, p = .046), BC (ˇ10 = 1.608, SE = 0.286, t(991) = 5.617,

< .001), NA (ˇ10 = 0.811, SE = 0.271, t(991) = 2.995, p = .003), and EBˇ10 = 0.133, SE = 0.032, t(991) = 4.112, p < .001) as compared to noxposure to fat talk.

ffects of Participating in Versus Overhearing Fat Talk

To examine Hypothesis 2, we used MLM analyses to determinehether participating in fat talk and overhearing fat talk had dif-

erential effects on BD, BC, NA, and EB. For these analyses, twoummy-coded variables, one of which represented participating inat talk and the other which represented overhearing fat talk, weredded to the Level-1 equation. The overheard variable was codeduch that a value of 1 indicated overhearing fat talk since the lastlarm and a value of 0 indicated not overhearing fat talk since the

ast alarm. The participation variable was coded such that a value of

indicated participating in fat talk since the last alarm and a valuef 0 indicated not participating in fat talk since the last alarm. The

3 The terms for this equation are defined as follows: Yti = the dependent outcomeariable (BD, BC, NA, or EB), ˇ00 = the intercept, ˇ10 = the slope for participants whoere exposed to fat talk, r0i = the Level-2 residual, and eti = the Level-1 residual.

e 11 (2014) 337–345

reference group for these analyses was no fat talk experience. Weused the following equation for these analyses4:

Yti = ˇ00 + ˇ10(Overheard)ti + ˇ20(Participated)ti + r0i + eti

A statistically significant positive coefficient for the slopeof the participated variable for BC indicated that participa-tion in fat talk resulted in significantly greater BC (ˇ20 = 1.156,SE = 0.401, t(990) = 2.881, p = .004). The coefficient for the slope ofthe overheard was nonsignificant for BC (ˇ10 = −0.126, SE = 0.383,t(990) = −0.328, p = .743). Additional analyses examined whetherbody checking differed as a function of participating in versusoverhearing fat talk. Results indicated that participation in fat talkwas associated with significantly greater body checking than over-hearing fat talk (�2 = 4.730, p = .028). The remaining findings werenonsignificant (see Table 1).

Evaluation of Moderation by Trait Self-Objectification

To examine Hypothesis 3, we used MLM analyses to deter-mine whether trait self-objectification moderated the relationshipbetween exposure to fat talk and BD, BC, NA, and EB. Fat talk wasentered as a Level-1 variable. Trait self-objectification (SO), a con-tinuous between-subjects variable, was grand-mean centered andentered as a Level-2 variable. We examined moderation as a cross-level interaction using the Level-1 fat talk variable and the Level-2SO variable. We conducted separate MLM models for each of thefour outcome variables and used the following equation for theseanalyses5:

Yti = ˇ00 + ˇ01(SO)ti + ˇ10(FT)ti + ˇ11(SO × FT)ti

+ r0i + r1i(FT)ti + eti

Statistically significant positive coefficients for the slope offat talk exposure for BC, NA, and EB indicated exposure to fattalk at the average level of SO was associated with signifi-cantly greater BC (ˇ10 = 1.554, SE = 0.329, t(62) = 4.717, p < .001), NA(ˇ10 = 0.774, SE = 0.275, t(62) = 2.813, p = .007), and EB (ˇ10 = 0.147,SE = 0.047, t(62) = 3.120, p = .003). The coefficient for the slope offat talk exposure for BD was nonsignificant (ˇ10 = 0.415, SE = 0.232,t(62) = 1.787, p = .079). Statistically significant positive coefficientsfor the slope of SO for BC indicated a main effect for SO. Resultsindicated that participants higher in SO experienced significantlyhigher levels of BC (ˇ01 = 0.201, SE = 0.070, t(62) = 2.872, p = 0.006).The coefficients for the slope of SO were nonsignificant for BD(ˇ01 = 0.061, SE = 0.042, t(62) = 1.462, p = .149), NA (ˇ01 = 0.056,SE = 0.028, t(62) = 1.990, p = .051), and EB (ˇ01 = 0.003, SE = 0.003,t(62) = 0.863, p = .392).

With respect to the interaction term between fat talk and SO,statistically significant coefficients for the slope of SO on the slopeof fat talk for BD and BC indicated that SO moderates the rela-

The terms for this equation are defined as follows: Yti = the dependent out-come variable (BD, BC, NA, or EB), ˇ00 = the intercept, ˇ10 = the slope for participantswho overheard fat talk, ˇ20 = the slope for participants who participated in fat talk,r0i = the level-2 residual, and eti = the Level-1 residual.

5 The terms for this equation are defined as follows: Yti = the dependent outcomevariable (BD, BC, NA, or EB), ˇ00 = the intercept, ˇ01 = the slope of self-objectificationon the dependent variable, ˇ10 = the slope for participants who were exposed to fattalk, ˇ11 = the slope of self-objectification on the slope of fat talk exposure on thedependent variable, r0i = the Level-2 residual, r1i = the Level-2 residual, and eti = theLevel-1 residual.

y Imag

t(

otfawfitbh

rmwaiiubabiwewfehncd

ipesitiaitle(fs2eit

ssoftns

M.D. Jones et al. / Bod

o predict NA (ˇ11 = 0.038, SE = 0.022, t(62) = 1.756, p = .084) or EBˇ11 = 0.005, SE = 0.004, t(62) = 1.262, p = .212).

Discussion

Results of this study suggest that fat talk has a detrimental effectn college aged women’s thoughts, feelings, and behaviors relatedo body image and eating habits. Specifically, recent exposure toat talk is associated with significantly greater body dissatisfactionnd negative affect, and greater frequencies of body checking andeight control and disordered eating behaviors. Participation in

at talk appears to have a greater effect on body checking behav-ors as compared to merely overhearing others participating in fatalk. Additionally, fat talk exposure has a greater negative effect onody dissatisfaction and frequency of body checking for individualsigher in trait self-objectification.

One overarching conclusion that can be drawn from theseesults is that participation in fat talk in the naturalistic environ-ent has an impact on young women. Furthermore, the impact onomen’s cognitions, affect, and behavior appears to be largely neg-

tive. This finding is consistent with previous laboratory researchndicating that exposure to fat talk is associated with increasesn body dissatisfaction (Stice et al., 2003). This study furthers thenderstanding of the link between fat talk and body dissatisfactiony demonstrating that fat talk not only affects cognitions associ-ted with body dissatisfaction, but it also negatively impacts theehavioral manifestation of body dissatisfaction (i.e., body check-

ng). Additionally, the finding that fat talk exposure is associatedith an increase in frequency of weight control and disordered

ating behaviors complements existing knowledge that individualsith clinically significant disordered eating engage in fat talk more

requently than those without eating disorder symptoms (Ousleyt al., 2008). These results suggest that participating in or over-earing fat talk may trigger thoughts of body dissatisfaction andegative affect, which may in turn lead to body checking, weightontrol behaviors (e.g., exercise and compensatory behaviors), andisordered eating (i.e., binging).

Although there are no significant differences between self-nitiated episodes and other-initiated episodes of fat talk,articipating in and overhearing fat talk appear to have differ-nt effects on those who are exposed. Fat talk participation has aignificantly greater negative impact on frequency of body check-ng compared to merely overhearing others participating in fatalk. There are several potential explanations for why engagingn fat talk may be associated with more frequent body checkings compared to overhearing fat talk conversations. Participatingn fat talk may be a more salient experience than overhearing fatalk, because women process the fat talk conversation at a deeperevel. Essentially, fat talk participation in which women consider orxpress thoughts about their body may trigger underlying schemasVitousek & Hollon, 1990) that result in women engaging in morerequent body checking. Additionally, overhearing fat talk may beuch a normative experience for most young women (Martz et al.,009; Nichter, 2000) that they have become desensitized to theffects of merely overhearing fat talk in a naturalistic setting. If thiss the case, participation in fat talk may increase sensitivity enougho elicit significant negative effects.

The relationships between fat talk exposure and body dis-atisfaction and body checking were each moderated by traitelf-objectification, such that individuals higher in trait self-bjectification experienced greater body dissatisfaction and more

requent body checking following fat talk exposure as comparedo individuals lower in trait self-objectification. These results areot consistent with previous research suggesting that fat talk mayerve as a protective factor in objectifying situations (Gapinski

e 11 (2014) 337–345 343

et al., 2003). While we have no data available regarding the con-text in which the fat talk conversations occurred, these findingsindicate that fat talk, which frequently takes place in objectify-ing situations or may elicit objectifying thoughts, has a negativeimpact across many domains. Consistent with the theoreticalfocus (e.g., Fredrickson & Roberts, 1997), it would appear that inthe naturalistic environment, individuals who are higher on traitself-objectification are especially susceptible to the negative body-related consequences associated with fat talk. Self-objectificationdid not emerge as a moderator of the relationship between fattalk and negative affect or fat talk and weight control and disor-dered eating behaviors. Future research should consider examiningother potential moderators of these relationships. For example,individuals high in perfectionism or trait body dissatisfaction mayexperience an exacerbation of negative outcomes following fat talkexposure.

As fat talk appears to trigger negative affect and cognitionsrelated to body dissatisfaction and more frequent body check-ing and weight control and disordered eating behaviors over thecourse of a few minutes to a few hours, it may be that it has evengreater negative cumulative effects after many exposures over along period of time. Individuals at risk for disordered eating maybe more vulnerable to the effects of fat talk and engaging in fattalk may contribute to the development or exacerbation of exist-ing eating disorder symptoms. As such, making efforts toward fattalk prevention seems important to the larger goal that many haveof decreasing body image disturbance in young women today. Fattalk rates among young women are alarmingly high (93%, Salk &Engeln-Maddox, 2011), and nearly 97% of our sample reporting fattalk exposure during the 5-daystudy. One method that may ulti-mately reduce frequency of fat talk is to increase fat talk awareness.Many participants in the study anecdotally reported being sur-prised at how frequently they overheard and participated in fat talkand some even reported believing that this heightened awarenessmight ultimately have positive effects on their participation leveland reaction to fat talk in the future. Given that similar preven-tion efforts have been successful in decreasing body dissatisfactionand disordered eating behaviors in college populations (Becker,Ciao, & Smith, 2008; Becker, McDaniel, Bull, Powell, & McIntyre,2012; Becker, Smith, & Ciao, 2006; Becker, Wilson, Williams, Kelly,McDaniel, & Elmquist, 2010; Perez, Becker, & Ramirez, 2010), effortsto increase awareness of fat talk and the negative effects that it hasmay have a similar positive impact on young women in a largercontext. Future fat talk interventions should also encourage youngwomen to reduce or eliminate fat talk due to its effect on their ownthoughts and emotions, its effects on the thoughts and emotionsof others, and its reinforcing effects on society’s objectification ofwomen. Particular focus should be given to the greater negativeeffects associated with participating in fat talk compared to merelyoverhearing fat talk.

Although this study produced significant and relevant findingsregarding the effects of fat talk in a naturalistic environment, itwas not without limitations. First, all measures used in the study,including the SOQ and the measures administered on the PDA, wereself-report. Despite the inherent problems that exist with the useof self-report measures, self-report remains the most efficient andcost-effective method of measuring participants’ internal statesand behaviors, particularly when assessing such information usingEMA protocols. Additionally, there are no other known methods ofaccurately measuring such cognitive data, such as thoughts aboutfat talk and state body dissatisfaction. A second limitation of thisresearch is the level of participant compliance to the PDA protocol.

Some participants did not respond to several of the random signalsduring the course of the study. However, this study’s compliancerate of 72.6% is comparable to the rates reported in other EMA stud-ies utilizing electronic diaries (Heron & Smyth, 2013; Myers et al.,

3 y Imag

2H

mrsttpotrHetppiibosooebqfi

tcbbtpcwss

hfgTcrtrttpisiftfrto

A

44 M.D. Jones et al. / Bod

012; Ridolfi et al., 2011) and meets the minimum standards set byox (2010) in order to maintain statistical power in MLM.

A third limitation of this research is the use of retroactive assess-ent of fat talk experiences. Although this limitation is somewhat

educed by having participants respond to fat talk-related itemseveral times throughout each day of the PDA portion of the study,he fact remains that participants were at times reporting on eventshat occurred up to several hours prior to the assessment timeoint. Future research examining the effects of fat talk exposurer participation could utilize event sampling, which requires par-icipants to self-initiate PDA questionnaires and would involveesponding to items immediately after each fat talk experience.owever, given that participants reported experiencing up to 27pisodes of fat talk over the course of the 5-day PDA protocol,he use of event sampling may raise concerns regarding potentialarticipant burden. In addition, one might question whether com-leting the PDA questionnaires multiple times over several days

nfluenced our findings. Interestingly, frequency of body check-ng decreased across days, suggesting that individuals may haveecome sensitized to this behavior. However, this did not impactur findings. Finally, participants were volunteers and may haveelf-selected to participate in this study due to an interest or previ-us knowledge of fat talk. Given the range in the reported frequencyf fat talk as well as the data indicating that only 34% of fat talkpisodes were initiated by our participants, we have no reason toelieve that our participants were predisposed to report greater fre-uencies or experience greater negative outcomes. However, thesendings are in need of replication.

Despite these limitations, this study expanded upon existing fatalk literature and furthered our understanding of the nature andonsequences of fat talk experiences by examining the associationetween daily episodes of fat talk and cognitive, emotional, andehavioral outcomes in the naturalistic environment. Importantly,his study identified that not only is fat associated with negativesychological outcomes, but it also impacts the frequency of bodyhecking and disordered eating and weight control behaviors. Itill be important to explore these questions in other populations,

uch as older and younger women, racially and culturally diverseamples, individuals with eating disorders, and men.

One of the interesting facets of this study is the relativelyigh frequencies of fat talk reported by participants. Although we

ocused on negative outcomes of fat talk, the high frequency sug-ests that engaging in fat talk has some value or reinforcing effects.herefore, future research might focus on potential positive out-omes of fat talk (e.g., perceived social support, social inclusion,eassurance, positive affect). Additionally, this study focused onhe outcomes and consequences of fat talk participation. Futureesearch should explore potential antecedents of fat talk participa-ion as well as the potential reciprocal relationships between fatalk and its outcomes. For example, to the extent that fat talk is aublic expression of body checking, it is possible that body check-

ng drives fat talk for the purposes of obtaining reassurance andupport. To our knowledge, this is the first study examining fat talkn a naturalistic setting, which has helped to illuminate facets ofat talk and its effects that were previously unable to be identifiedhrough laboratory studies (e.g., Stice et al., 2003). Understandingat talk as it occurs in the natural environment is critical not only inespect to assessing the impact of fat talk on participants, but alsoo inform prevention and intervention efforts aiming to decreaser eliminate fat talk among target groups.

References

rroyo, A., & Harwood, J. (2012). Exploring the causes and consequences ofengaging in fat talk. Journal of Applied Communication Research, 40, 167–187.http://dx.doi.org/10.1080/00909882.2012.654500

e 11 (2014) 337–345

Becker, C. B., Ciao, A. C., & Smith, L. M. (2008). Moving from efficacy to effectivenessin eating disorders prevention: The Sorority Body Image Program. Cognitive andBehavioral Practice, 15, 18–27. http://dx.doi.org/10.1016/j.cbpra.2006.07.06

Becker, C. B., McDaniel, L., Bull, S., Powell, M., & McIntyre, K. (2012). Can we reduceeating disorder risk factors in female college athletes? A randomized exploratoryinvestigation of two peer-led interventions. Body Image: An International Journalof Research, 9, 31–42. http://dx.doi.org/10.1016/j.bodyim.2011.09.005

Becker, C. B., Smith, L. M., & Ciao, A. C. (2006). Peer-facilitated eatingdisorder prevention: A randomized effectiveness trial of cognitive disso-nance and media advocacy. Journal of Counseling Psychology, 53, 550–555.http://dx.doi.org/10.1037/0022-0167.53.4.550

Becker, C. B., Wilson, C., Williams, A., Kelly, M., McDaniel, L., & Elmquist, J. (2010).Peer-facilitated cognitive dissonance versus healthy weight eating disordersprevention: A randomized comparison. Body Image: An International Journal ofResearch, 7, 280–288. http://dx.doi.org/10.1016/j.bodyim.2010.06.004

Britton, L. E., Martz, D. M., Bazzini, D. G., Curtin, L. A., & LeaShomb, A. (2006). Fattalk and self-presentation of body image: Is there a social norm for womento self-degrade? Body Image: An International Journal of Research, 3, 247–254.http://dx.doi.org/10.1016/j.bodyim.2006.05.006

Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications anddata analysis methods. Newbury Park, CA: Sage Publications.

Clarke, P. M., Murnen, S. K., & Smolak, L. (2010). Development and psychometricevaluation of a quantitative measure of “fat talk”. Body Image: An InternationalJournal of Research, 7, 1–7. http://dx.doi.org/10.1016/j.bodyim.2009.09.006

Colautti, L. A., Fuller-Tyszkiewicz, M., Skouteris, H., McCabe, M., Blackburn,S., & Wyett, E. (2011). Accounting for fluctuations in body dissatis-faction. Body Image: An International Journal of Research, 8, 315–321.http://dx.doi.org/10.1016/j.bodyim.2011.07.001

Fredrickson, B. L., & Roberts, T. (1997). Objectification theory: Toward understand-ing women’s lived experiences and mental health risks. Psychology of WomenQuarterly, 21, 173–206. http://dx.doi.org/10.1111/j.1471-6402.1997.tb00108.x

Gapinski, K. D., Brownell, K. D., & LaFrance, M. (2003). Body objectification and “fattalk”: Effects on emotion, motivation, and cognitive performance. Sex Roles, 48,377–388. http://dx.doi.org/10.1023/A:1023516209973

Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for mea-suring state self-esteem. Journal of Personality and Social Psychology, 60, 895–910.http://dx.doi.org/10.1037/0022-3514.60.6.895

Heron, K. E., & Smyth, J. M. (2013). Is intensive measurement of body imagereactive? A two-study evaluation using Ecological Momentary Assessmentsuggests not. Body Image: An International Journal of Research, 10, 35–44.http://dx.doi.org/10.1016/j.bodyim.2012.08.006

Hox, J. J. (2010). Multilevel analysis: Techniques and applications. New York, NY: Rout-ledge.

Lindner, D., Tantleff-Dunn, S., & Jentsch, F. (2012). Social comparison andthe ‘circle of objectification’. Sex Roles, 67, 222–235. http://dx.doi.org/10.1007/s11199-012-0175-x

Martz, D. M., Petroff, A. B., Curtin, L., & Bazzini, D. G. (2009). Gender differences infat talk among American adults: Results from the psychology of size survey. SexRoles, 61, 34–41. http://dx.doi.org/10.1007/s11199-009-9587-7

Miner-Rubino, K., Twenge, J. M., & Fredrickson, B. L. (2002). Trait self-objectificationon women: Affective and personality correlates. Journal of Research in Personal-ity, 36, 147–172. http://dx.doi.org/10.1006/jrpe.2001.2343

Muehlenkamp, J. J., & Saris-Baglama, R. N. (2002). Self-objectification and its psy-chological outcomes for college women. Psychology of Women Quarterly, 26,371–379. http://dx.doi.org/10.1111/1471-6402.t01-1-00076

Myers, T. A., & Crowther, J. H. (2007). Sociocultural pressures, thin-ideal internal-ization, self-objectification, and body dissatisfaction: Could feminist beliefs be amoderating factor? Body Image: An International Journal of Research, 4, 296–308.http://dx.doi.org/10.1016/j.bodyim.2007.04.001

Myers, T. A., & Crowther, J. H. (2008). Is self-objectification related to inter-oceptive awareness? An examination of potential mediating pathways todisordered eating attitudes. Psychology of Women Quarterly, 32, 172–180.http://dx.doi.org/10.1111/j.1471-6402.2008.00421.x

Myers, T. A., Ridolfi, D. R., Crowther, J. H., & Ciesla, J. A. (2012). The impactof appearance-focused social comparisons on body image disturbance in thenaturalistic environment: The roles of thin-ideal internalization and fem-inist beliefs. Body Image: An International Journal of Research, 9, 342–351.http://dx.doi.org/10.1016/j.bodyim.2012.03.005

Nichter, M. (2000). Fat talk: What girls and their parents say about dieting. Cambridge,MA: Harvard University Press.

Nichter, M., & Vuckovic, N. (1994). Fat talk. In N. Sault (Ed.), Many mirrors: Body imageand social relations (pp. 109–131). New Brunswick, NJ: Rutgers University Press.

Noll, S. M., & Fredrickson, B. L. (1998). A mediational model linking self-objectification, body shame, and disordered eating. Psychology of WomenQuarterly, 22, 623–636. http://dx.doi.org/10.1111/j.1471-6402.1998.tb00181.x

Ousley, L., Cordero, E. D., & White, S. (2008). Fat talk among college stu-dents: How undergraduates communicate regarding food and body weight,shape, and appearance. Eating Disorders, 16, 73–84. http://dx.doi.org/10.1080/10640260701773546

Payne, L. O., Martz, D. M., Tompkins, K. B., Petroff, A. B., & Farrow, C. V. (2011). Gendercomparisons of fat talk in the United Kingdom and the United States. Sex Roles,

65, 557–565. http://dx.doi.org/10.1007/s11199-010-9881-4

Perez, M., Becker, C. B., & Ramirez, A. (2010). Transportability of an empirically sup-ported dissonance-based prevention program for eating disorders. Body Image:An International Journal of Research, 7, 179–186. http://dx.doi.org/10.1016/j.bodyim.2010.02.006

y Imag

R

R

R

S

S

S

S

S

S

M.D. Jones et al. / Bod

eas, D. L., Whisenhunt, B. L., Netemeyer, R., & Williamson, D. A. (2002). Devel-opment of the Body Checking Questionnaire: A self-report measure of bodychecking behaviors. International Journal of Eating Disorders, 31, 324–333.http://dx.doi.org/10.1002/eat.10012

idolfi, D. R., Myers, T. A., Crowther, J. H., & Ciesla, J. A. (2011). Do appearance focusedcognitive distortions moderate the relationship between social comparisons topeers and media images and body image disturbance? Sex Roles, 65, 491–505.http://dx.doi.org/10.1007/s11199-011-9961-0

udiger, J. A., & Winstead, B. A. (2013). Body talk and body-related co-rumination: Associations with body image, eating attitudes, and psychologicaladjustment. Body Image: An International Journal of Research, 10, 462–471.http://dx.doi.org/10.1016/j.bodyim.2013.07.010

alk, R. H., & Engeln-Maddox, R. (2011). “If you’re fat, then I’m humongous!”: Fre-quency, content, and impact of fat talk among college women. Psychology ofWomen Quarterly, 35, 18–28. http://dx.doi.org/10.1177/0361684310384107

alk, R. H., & Engeln-Maddox, R. (2012). Fat talk among college women is bothcontagious and harmful. Sex Roles, 66, 636–645. http://dx.doi.org/10.1007/s11199-011-0050-1

later, A., & Tiggemann, M. (2002). A test of objectification theory in girls. Sex Roles,46, 343–349. http://dx.doi.org/10.1023/A:1020232714705

later, A., & Tiggemann, M. (2012). Time since menarche and sport participation aspredictors of self-objectification: A longitudinal study of adolescent girls. SexRoles, 67, 571–581. http://dx.doi.org/10.1007/s11199-012-0200-0

tone, A. A., & Shiffman, S. (1994). Ecological momentary assessment(EMA) in behavioral medicine. Annals of Behavioral Medicine, 16,199–202.

tice, E., Maxfield, J., & Wells, T. (2003). Adverse effects of social pres-sure to be thin on young women: An experimental investigation of the

e 11 (2014) 337–345 345

effects of “fat talk”. International Journal of Eating Disorders, 34, 108–117.http://dx.doi.org/10.1002/eat.10171

Tiggemann, M., & Kuring, J. K. (2004). The role of body objectification in disorderedeating and depressed mood. British Journal of Clinical Psychology, 43, 299–311.http://dx.doi.org/10.1348/0144665031752925

Tucker, K. L., Martz, D. M., Curtin, L. A., & Bazzini, D. G. (2007). Examining “fattalk” experimentally in a female dyad: How are women influenced by anotherwoman’s body presentation style? Body Image: An International Journal ofResearch, 4, 157–164. http://dx.doi.org/10.1016/j.bodyim.2006.12.005

Tylka, T. L., & Hill, M. S. (2004). Objectification theory as it relates todisordered eating among college women. Sex Roles, 51, 719–729.http://dx.doi.org/10.1007/s111999-004-0721-2

Van Diest, A. M. K., & Perez, M. (2013). Exploring the integration of thin-ideal inter-nalization and self-objectification in the prevention of eating disorders. BodyImage, 10, 16–25. http://dx.doi.org/10.1016/j.bodyim.2012.10.004

Vandenbosch, L., & Eggermont, S. (2013). Sexualization of adolescent boys:Media exposure and boys’ internalization of appearance ideals, self-objectification, and body surveillance. Men and Masculinities, 16, 283–306.http://dx.doi.org/10.1177/109718X13477866

Vitousek, K. B., & Hollon, S. D. (1990). The investigation of schematic content andprocessing in eating disorders. Cognitive Therapy and Research, 14, 191–214.http://dx.doi.org/10.1007/BF01176209

Watson, D., & Clark, L. A. (1994). The PANAS-X: Manual for the Positive And Negative

Affect Schedule-Expanded Form. Iowa City: University of Iowa. Retrieved fromhttp://www.psychology.uiowa.edu/Faculty/Watson/Watson.html

Weiss, H. M., Beal, D. J., Lucy, S. L., & MacDermid, S. M. (2004). Constructing EMA stud-ies with PMAT: The Purdue Momentary Assessment Tool’s user manual. Retrievedfrom http://www.ruf.rice.edu/∼dbeal/pmatusermanual.pdf