ORIGINAL RESEARCH

published: 28 October 2021

doi: 10.3389/fpsyg.2021.724710

Translation and Validation of a

State-Measure of Body Image
Satisfaction: The Body Image State
Scale
Luc Bardi 1,2* , Claire Arnaud 1 , Céline Bagès 1 , Frédéric Langlois 2 and Amelie Rousseau 1,3
1
Université de Lille, ULR 4072 – PSITEC – Psychologie: Interactions, Temps, Emotions, Cognition, Lille, France, 2 Laboratoire
sur l’anxiété et le Perfectionnisme, Department of Psychology, Université du Québec à Trois-Rivières, Trois-Rivières, QC,
Canada, 3 Centre d’Etudes et Recherches en Psychopathologie et Psychologie de la Santé (CERPPS), Université de
Toulouse (UT2J), Toulouse, France

The aim of the present study is to test the validity and reliability of the French Body Image
State Scale (F-BISS). The scale was translated using a back-translation technique,
with discrepancies being settled through consensus. Three hundred and twelve female
participants were recruited. Convergent validity was assessed using eating disorder
evaluation and social comparison. Exploratory and confirmatory factor analyses were
also conducted. The translated Body Image State Scale (BISS) demonstrated good
psychometric properties, with good internal consistency (α = 0.83), and adequate
Edited by: goodness-of-fit. The translated BISS presented a unifactorial structure, with one factor
Michail Mantzios,
Birmingham City University,
explaining 56% of the variance. The exploratory factor analysis led to the removal of
United Kingdom a single item due to insufficient factor loading (<0.45). Its convergent validity seems
Reviewed by: consistent with previous literature. Discriminant analyses showed a significant difference
Maya Libben, in F-BISS score between participants relative to eating disorder symptomatology
The University of British Columbia
Okanagan Campus, Canada (t = 11.65; p < 0.001). This translation could prove useful in both research and clinical
Lilac Lev-Ari, settings to assess state body satisfaction in French populations.
Ruppin Academic Center, Israel
Keywords: body satisfaction, state measure, validation, French, body image
*Correspondence:
Luc Bardi
luc.bardi@uqtr.ca
INTRODUCTION
Specialty section:
This article was submitted to A growing body of research warns of the public health issue represented by body dissatisfaction,
Eating Behavior, or negative body assessment (Bucchianeri and Neumark-Sztainer, 2014; Griffiths et al., 2017;
a section of the journal Bornioli et al., 2019). Developing reliable measurement tools or validating existing ones in
Frontiers in Psychology multiple languages in this context allows researchers to better understand the onset of eating
Received: 14 June 2021 disorder development. Indeed, body dissatisfaction is a risk factor for eating disorders (Stice, 2002;
Accepted: 05 October 2021 Stice et al., 2011). In western countries, women are pressured to achieve a thin body-ideal and
Published: 28 October 2021
western sociocultural influences have been proven to be a risk factor for general population body
Citation: dissatisfaction (Holmqvist and Frisén, 2010). In France, men’s body dissatisfaction differs from
Bardi L, Arnaud C, Bagès C, women’s, with men being more preoccupied about their muscularity and gaining weight (Pope
Langlois F and Rousseau A (2021)
et al., 2000) and women being more concerned about being thin and toned (Girard et al., 2018). In
Translation and Validation of a
State-Measure of Body Image
this study, we will focus on state-measurement tools, as they have raised criticism in the way some
Satisfaction: The Body Image State were developed and validated in the past (Cash et al., 2002; Bateson et al., 2007). More precisely, we
Scale. Front. Psychol. 12:724710. will focus on the validation of a positive body assessment, or body satisfaction, state scale: the Body
doi: 10.3389/fpsyg.2021.724710 Image State Scale (BISS) (Cash et al., 2002).

Frontiers in Psychology | www.frontiersin.org 1 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

Body dissatisfaction or satisfaction are indeed variably structure, with every item loading on the main factor at a
conceptualized in two ways, “state,” and “trait,” meaning it can be >0.75 coefficient (Mebarak Chams et al., 2019). Convergent
viewed as an immediate state of being, or a stable personality trait validity across studies also appears coherent with literature:
(Cash et al., 2002). Multiple trait body-dissatisfaction scales have the BISS had a negative correlation with body-dissatisfaction
been translated in French, such as the Body Shape Questionnaire (r = −0.52, p < 0.001; Cash et al., 2002; r = −0.58, p < 0.001;
(BSQ; Cooper et al., 1987; Rousseau et al., 2005), the body- Mebarak Chams et al., 2019), eating disorder symptomatology
dissatisfaction subscale of the EDI (Garner et al., 1983; Archinard (r = −0.79, p < 0.01; Alleva et al., 2014) and general
et al., 1996), the Eating Disorder Examination Questionnaire’s psychopathology (r = −0.33, p < 0.001; Mebarak Chams et al.,
(EDE-Q’s) shape and weight concern subscales (Fairburn and 2019). Discriminant analyses also showed a significant difference
Beglin, 1994; Fairburn, 2008; Carrard et al., 2015), or the Male between Body Mass Index (BMI) groups with participants
Body Dissatisfaction Scale (Ochner et al., 2009; Rousseau et al., having a higher BMI being less satisfied than those with a
2014). The use of trait-scales lie in evaluating body-dissatisfaction lower BMI (Mebarak Chams et al., 2019). However, no study
in a punctual fashion, for research or clinical purposes such as has yet replicated a long-term test–retest procedure like in
determining if a patient is suitable for therapy. Moreover, trait- the original study. Indeed, the only test–retest data available
measures are often retrospective, and do not seem much suited comes from Carraro et al. (2010), who assessed test–retest
for assessing change over a shorter period of time. Therefore, fidelity only an hour after an experimental task (r = 0.87,
life-changes that may harm body image such as receiving a p < 0.01). The BISS appears to have been neither translated nor
surgery (Sarwer et al., 2010) could only be assessed after a validated into French.
period of time when using a trait-measure. Since negative Validation of the BISS in French would be of use for a variety
body image is linked to pathology, having to wait to correctly of professions. State scales as said are indeed very useful in
assess it could hinder patients’ treatment. Cash et al. (2002) research and clinical settings. In a research setting, they may be
emphasized the fact that most measures of body dissatisfaction used to assess the immediate impact of an exposure task such
or satisfaction are trait measures, such as the BSQ (Cooper as the ones used in the study of social media’s impact on body
et al., 1987). To the research team’s knowledge this statement image (Tiggemann and Zaccardo, 2015; Cohen et al., 2019). In
holds true today in France with only one figure or contour- a clinical setting, they could be used by the therapist as a self-
drawing based state scale being validated in French (Moussally evaluation tool for patients to better understand what causes
et al., 2017) and another Contour-Drawing Rating Scale (CDRS; body image-related distress, and assess differences over time. The
Thompson and Gray, 1995) being used in some research with BISS is even more relevant for patients’ use, as it is a short and
French-speaking samples (Duchesne et al., 2017; Rivière et al., easy scale to use.
2018) with seemingly no prior validation. However, this type To validate the BISS, it seemed important to take notice of
of scale has been criticized for not being representative enough the differences in body dissatisfaction expression between men
of female body diversity (Bateson et al., 2007). When searching and women’s body image. Indeed, research has shown men tend
for the keywords “body-dissatisfaction” and “state scale” and to express dissatisfaction related to their muscle mass (Karazsia
“French,” no relevant results are found on the PsycNet or et al., 2017). As mentioned earlier, this finding seems to hold
ScienceDirect databases. true in a French sample, with men wanting to gain weight and
Cash et al. (2002) also underlined the issue that tools used muscle mass (Pope et al., 2000). French women, on the other
to measure state body satisfaction in previous research have hand, seem to be pressured to reach a more thin and toned body-
often been developed from trait-scales, or constructed without shape (Girard et al., 2018). Moreover, 70% of French women
prior validation. As said earlier, these scales may lack sensitivity. want to lose weight (Valls et al., 2013) and 34.6% state that
For this purpose, Cash et al. (2002) developed the Body-Image their self-opinion depends on their weight (Lachaud et al., 2004).
State Scale. This 6-item state-measure of body satisfaction is While those results underline the importance of validating a
non-specific, exclusively text-based and originally available in state-measure in French to adequately assess immediate body-
English. Its 9-point scale with specific body-related statements satisfaction, this would also mean we would need different
to choose from should allow participants to give a precise measures of convergent validity for each gender. Therefore, we
answer. It is less prone to the unrealistic body representation wished to focus at first on a female population.
associated with contour-drawing based scales, as participants Secondly, elements in research made us consider validating the
are asked about their own body, without being shown one. BISS in a younger population. Indeed, research showed that cut-
The BISS has been widely used in previous literature (i.e., off scores to the EDE-Q (Fairburn and Beglin, 1994; Fairburn,
Etu and Gray, 2010; Walker et al., 2012; Boersma and Jarry, 2008) varied between age groups, being higher in a younger
2013). Currently, Spanish, Italian, and Dutch versions have been population (Rø et al., 2015). Moreover, older women tend to
developed, with only the Spanish version being the subject of display a lesser desire, or drive, to be thinner (Pruis and Janowsky,
a validation study (Carraro et al., 2010; Alleva et al., 2014; 2010); this would be a bias since tools available for this study
Mebarak Chams et al., 2019). Convergent validity has been were mainly validated in a younger population and addressed
tested using the BSQ (Cooper et al., 1987), and the EDE- thinness-related preoccupations.
Q (Fairburn and Beglin, 1994). Validation studies have shown The main objective of this study is therefore to translate and
very good psychometric qualities (Cash et al., 2002; Mebarak validate the BISS (Cash et al., 2002) in an 18–25-year-old female
Chams et al., 2019). The BISS appears to have a single-factor French population.

Frontiers in Psychology | www.frontiersin.org 2 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

It was expected here that a French version of the BISS were again prompted to input their code. Finally, a written debrief
would negatively correlate with measures of body dissatisfaction, was sent via email.
social comparison, and eating symptomatology, and would
present a single-factor structure, as demonstrated in relevant Study Material
research (Mebarak Chams et al., 2019). Moreover, we wished to Body Image State Scale
replicate differences between BMI groups and eating disorder This 6-item self-reported scale (Cash et al., 2002) is rated on a 9-
symptomatology groups found in the Mebarak Chams (2019) point Likert scale. Each item begins with the sentence “Right now,
study; it is expected that higher BMI categories will have a lesser I feel” [for instance: “Right now, I feel (Extremely dissatisfied to
French Body Image State Scale (F-BISS) score than lower BMI Extremely satisfied) with my physical appearance”]. Every rating’s
groups. It is also expected that participants with a clinically phrasing is different. For instance, item 1 ranges from Extremely
significant score to an eating disorder symptomatology scale dissatisfied (1) to Extremely satisfied (9), while item 4 ranges
will have a lower score than participants with a clinically non- from Extremely physically attractive (9) to Extremely physically
significant score (Mebarak Chams et al., 2019). unattractive (1). Score is a mean of every item. Higher scores
denote higher body satisfaction. Half items have reverse rating (2,
4, 6). Internal consistency calculated by Cronbach’s alpha in the
MATERIALS AND METHODS original study is 0.77 in women, and 0.83 in this study’s sample.
Participants Body Shape Questionnaire 8-Item
The study included 312 participants. Seventy-eight participants This self-reported questionnaire was developed by Cooper et al.
completed the test two times at a 2-week interval (N Time 1 = 312; (1987), and abbreviated by Evans and Dolan (1993). This
N Time 2 = 78). Participants were 18–25-year-old female students abridged version has been validated in French by Lentillon-
(M = 21.07, SD = 1.82). Self-reported values of height and weight Kaestner et al. (2014). This 8-item questionnaire measures body
were used to derive BMI (M = 22.88, SD = 4.48). Participants dissatisfaction over the past 4 weeks [item example: “Has thinking
did not receive compensation. For sample size, Everitt’s (1975) about your shape interfered with your ability to concentrate (e.g.,
recommendations of a subject to item ratio of 10 were used. while watching television, reading, listening to conversations?”)].
A total score of trait body dissatisfaction is obtained by adding
Procedure the score for each Likert-scale item (1–6; maximum score of 48).
This study received approval by the University’s Board of A higher score denotes higher body dissatisfaction. In this study’s
Ethics. Recommendations for translation from Cha et al. (2007) sample, internal consistency measured by Cronbach’s alpha is
were followed. The scale was translated into French by several 0.93, and scores range from 8 to 48, with a mean score of
members of the research team, one of whom was fluent in 25.35 (SD = 10.47).
English (C1 level; Council of Europe, n.d.). Due to a lack
of independent translators, the translation was then back- Eating Disorder Examination Questionnaire
translated by another English-fluent team member and reviewed This self-reported questionnaire (Fairburn and Beglin, 1994;
by the research team. Minor discrepancies were settled through Fairburn, 2008), validated in French by Carrard et al. (2015),
consensus. An advertisement for a study on body image was assesses eating disorder symptomatology. Items include
posted on various French-speaking student Facebook groups, questions such as “Have you gone for long periods of time (8
with a message stating the research team’s intent to recruit waking hours or more) without eating anything at all in order to
18–25-year-old female participants. Willing participants had to influence your shape or weight?” Answers are on a 7-point Likert
fill out the Body Shape Questionnaire 8-item (BSQ-8C), the scale. While the questionnaire was originally designed with four
EDE-Q, the Physical Appearance Comparison Scale-4 (PACS- subscales in mind, it is advised to only use the overall score, as
4), the BISS, and a sociodemographic data questionnaire, a single-factor structure is thought to be more robust (Friborg
in that order. When ending the questionnaire, participants et al., 2013). The four subscales are restraint (items 1, 2, 3, 4, 5;
were asked to create a code (enabling anonymous test–retest α = 0.84), eating concerns (items 7, 9, 19, 21, 20; α = 0.80), shape
analyses) consisting of the two last digits of participants’ phone concerns (items 6, 8, 23, 10, 26, 27, 28, 11; α = 0.91), and weight
number, the first two letters of their first name and the last concerns (items 22, 24, 8, 25, 12; α = 0.89). Cut-off scores from
two digits of their birth year. Participants were then asked Rø et al. (2015) study were used. While Rø et al. (2015) evaluated
if they wished to take part in the study’s second phase. If Norwegian adults, and warned about eventual cross-cultural
they did, they had to input their participant code again and differences, a lack of similar data in the French validation article
provide their email address. Willing participants received an (Carrard et al., 2015) forced us to use a near cultural equivalent
email 2 weeks later with a link to another questionnaire. To for the same scale. For an underweight BMI, cut-off was 1.62. For
replicate the original study, we chose to respect the same 2- a normal BMI, cut-off was 2.51. For an overweight BMI, cut-off
week delay between questionnaire answering (Cash et al., 2002). was 3.15. Finally, for an obese BMI, cut-off was 3.26. A score
A 2-week interval also seems to be considered the highest superior to these values was considered clinically significant. In
appropriate interval for retest of a state-scale (Polit, 2014). The this study’s sample, internal consistency measured by Cronbach’s
questionnaire contained the French BISS, as well as another 4- alpha is 0.91, and scores range from 0 to 5.57, with a mean score
item questionnaire to be validated in another study. Participants of 1.95 (SD = 1.40).

Frontiers in Psychology | www.frontiersin.org 3 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

The Physical Appearance Comparison Scale-4 at times 1 and 2, with a two-way mixed model. Convergent
This self-reported 4-item questionnaire (Thompson et al., 1991; validity was determined using Pearson’s correlations between
Dany and Urdapilleta, 2012) measures the general tendency of BSQ-8C, EDE-Q, PACS-4 scores and BISS scores. Concurrent
individuals to compare themselves with others in social situations validity was assessed using ANOVAs between BMI groups
(item example: “At parties or other social events, I compare and t-tests between eating disorder symptomatology groups.
my physical appearance to the physical appearance of others.”). Univariate normality was assessed for t-tests. A cut-off interval
Participants are presented a Likert scale ranging from 1 (never) of [−2; 2] was used as per George and Mallery’s (2010)
to 5 (always). Overall score is calculated by adding each item’s recommendations. For clinically significant participants, kurtosis
individual score. The higher the score (maximum score of 20), was 0.33 and skewness was −0.43. For non-clinically significant
the more likely the individual is to use social comparison. In participants, kurtosis was −0.61 and skewness was 0.36.
this study’s sample, internal consistency measured by Cronbach’s
alpha is 0.84, and scores range from 4 to 20, with a mean score of
11.92 (SD = 3.85). RESULTS
Sociodemographic Data Questionnaire
Item-Analysis and Construct Validity
This questionnaire was used to report on participants’ age,
The exploratory factor analysis was computed using a maximum
education level, and area of study. As some parts of this
likeliness extraction method. KMO index was satisfactory
questionnaire were explicitly meant for female participants
(KMO = 0.82), and Bartlett’s test of sphericity produced a
(EDEQ items on menstruation and contraceptive pill usage),
significant result (p < 0.001). Only one component reached
and since the recruitment campaign was explicitly directed
eigenvalue >1.00 (eigenvalue = 3.35) (see Figure 1). A parallel
at cisgendered women 18–25 years old, no questions about
analysis confirmed a one-factor solution (eigenvalue of second
gender were added.
factor: 0.92; parallel analysis eigenvalue for a two-factor solution:
1.14). According to the chosen cut-off, item 2 does not
Statistical Analyses adequately load on the main factor (see Table 1). When assessing
Construct validity was assessed using exploratory and communalities, item 2 was the only one under our chosen cut-off
confirmatory factor analyses, as well as inter-item correlations. point (see Table 2). Therefore, item 2 was removed from the rest
The exploratory factor analysis was conducted on SPSS 25, while of the analyses.
the confirmatory factor analysis was carried out on RStudio The confirmatory analysis also returned a single factor
using the lavaan package (Rosseel, 2012). Both used a maximum solution (see Figure 2). The MI function on RStudio indicated
likelihood estimation method. For the factor analyses, a “fair” higher MI values when allowing correlation between items 4 and
item loading cut-off (0.45) was set (Tabachnick and Fidell, 2007). 6 (MI = 37.86). When allowing correlation between items 4 and
A 0.2 cut-off was set for communalities (Child, 2006). The 6, goodness-offit indices were adequate for the second model (see
Kaiser criterion was used, meaning valid factors should have an Table 3). All items had high item-total correlation coefficients at
eigenvalue greater than one (Costello and Osborne, 2005). For time 1, ranging from r = 0.65 (p < 0.001) to r = 0.85 (p < 0.001;
the exploratory factor analysis, examination of the scree-plot and see Table 4).
a parallel analysis were also used to corroborate factor solutions.
Data from one random split-half of the sample (n = 156) was
selected for the exploratory factor analysis. Data from the other Reliability
random split-half of the sample (n = 156) was selected for the Cronbach’s alpha and McDonald’s omega were used to define
confirmatory factor analysis. While no clear recommendation the F-BISS’s internal consistency. At time 1 and time 2, internal
exists on what constitutes a satisfactory percentage of explained consistency without item 2 was satisfactory, with α = 0.83 at both
variance, values between 50 and 60% were chosen to be retained time 1 and time 2. Using McDonald’s omega, internal consistency
(Peterson, 2000). Goodness-of-fit indices were established prior remained satisfactory, with ω = 0.85 at time 1, and ω = 0.86.
to testing (Hooper et al., 2008). RMSEA values close to 0.06 The test–retest reliability of the F-BISS was calculated without
were considered indicators of good fit. SRMR values under item 2 over a 2-week period. The coefficient of correlation was
0.05 were retained, as well-fitting models tend to obtain similar r = 0.86 (p < 0.001) for single measures.
values. Finally, as it is recommended that a CFI index should
be > 0.95, it was decided to keep that value for this testing. Convergent Validity
Model improvement was assessed through modification index Pearson’s correlations were used to establish links between the
values (MI), implemented in R by the lavaan package. Higher MI F-BISS and BMI, trait measures of body image, measures of
values indicate a better fit of the corresponding model, using the social comparison (PACS-4) and measures of eating disorder
LaGrange multiplier. symptomatology (EDE-Q). F-BISS scores without item 2 strongly
For inter-item correlations, moderate correlations (0.3–0.7) and negatively correlated with body dissatisfaction (BSQ;
were considered the lowest acceptable degree of correlation. r = −0.74, p < 0.001), eating disorder symptomatology (EDE-
Internal consistency was evaluated using Cronbach’s alpha and Q; r = −0.72, p < 0.001), and moderately correlated with BMI
McDonald’s omega. Test–retest fidelity was controlled using (r = −0.36, p < 0.001), and comparison to others (PACS-4;
absolute agreement intra-class correlations between BISS scores r = −0.43, p < 0.001).

Frontiers in Psychology | www.frontiersin.org 4 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

FIGURE 1 | Scree-plot for the F-BISS scale. Dashed line represents the Kaiser criterion.

FIGURE 2 | Confirmatory factor analysis of the BISS (model 2). Sts, main factor. One-way arrows represent factor loading. Two-way arrows represent correlations
between items.

Differences between BMI categories were significant underweight and normal categories (p = 0.86), no
(F = 13.85; p < 0.001), with Bonferroni post hoc significant differences between overweight and obese
testing revealing no significant differences between categories (p = 1.00), but significant differences between

Frontiers in Psychology | www.frontiersin.org 5 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

underweight/normal and overweight/obese categories The F-BISS was negatively correlated to overall EDE-
(p < 0.001). Q score. Moreover, participants with a clinically significant
Finally, differences between women with clinically significant symptomatology were less satisfied than other participants.
and non-clinically significant symptomatology for eating This is consistent with eating disorder literature, as a lower
disorders were significant (t = 11.65; p < 0.001). The clinically F-BISS score indicated lower body satisfaction, and thus greater
non-significant group had significantly higher satisfaction. dissatisfaction, a risk factor in eating disorder development (Stice
et al., 2011). As with the Spanish validation of the scale (Mebarak
Chams et al., 2019) the F-BISS was negatively correlated to a
trait measure of body dissatisfaction (BSQ). Finally, the F-BISS
DISCUSSION was negatively correlated with social comparison. Again, this is
consistent with research (Rodgers et al., 2011) as the original
The results yielded showed the French version of the BISS
BISS, and thus the F-BISS, measure body satisfaction (Cash
(F-BISS) to have good psychometric qualities. Items 1, 3, 4,
et al., 2002). BMI’s correlation to the F-BISS was somewhat
5, and 6 were adequately loaded on a main factor (>0.45),
consistent with the Spanish validation (r = −0.28; Mebarak
moderately to strongly intercorrelated, and internal consistency
Chams et al., 2019) but not with the original validation
was good after removing item 2. A one-factor solution also
(r = −0.53; Cash et al., 2002). Furthermore, an ANOVA
showed adequate fit after removal of item 2, and allowing for
showed that significant differences in F-BISS scores between
correlation between items 4 and 6.
BMI groups lie between “clusters” formed by the underweight
and normal groups, and the overweight and obese groups.
These findings could indicate either that BMI underestimates
TABLE 1 | Exploratory factor analysis of the F-BISS.
obesity prevalence by categorizing obese people as overweight
Factor 1 (Shah and Braverman, 2012), or that BMIs in the original
1. Right now, I feel (Extremely dissatisfied to Extremely satisfied) with 0.90
my physical appearance
En ce moment je me sens (Extrêmement insatisfaite à Extrêmement TABLE 3 | Confirmatory factorial analysis, and goodness-of-fit indices of
satisfaite) de mon apparence physique the F-BISS.
Right now, I feel (Extremely satisfied to Extremely dissatisfied) with my 0.26
Factor 1 (model 2)
body size and shape
En ce moment je me sens (Extrêmement satisfaite à Extrêmement Item 1 0.93
insatisfaite) de la taille et la forme de mon corps Item 3 0.77
Right now, I feel (Extremely dissatisfied to Extremely satisfied) with my 0.73 Item 4 0.51
weight
Item 5 0.48
En ce moment je me sens (Extrêmement insatisfaite à Extrêmement
Item 6 0.51
satisfaite) de mon poids
Right now, I feel (Extremely physically attractive to Extremely 0.74 Model 1 (without item 2) Model 2 (without
physically unattractive) item 2; item 4–item 6)
En ce moment je me sens (Extrêmement physiquement attirante à
Extrêmement physiquement repoussante) χ2 (df) 47.14 (5) 7.92 (4)
Right now, I feel (A great deal worse to A great deal better) about my 0.65 CFI 0.85 0.99
looks than I usually feel
RMSEA 0.23 0.08
En ce moment je me sens (Beaucoup moins bien à Beaucoup mieux)
RMSEA CI [0.18; 0.30] [0.00; 0.16]
à propos de mon apparence que d’habitude
SRMR 0.09 0.03
Right now, I feel (A great deal better to A great deal worse) than the 0.72
average person looks –: correlation between items allowed by modification index analysis.
En ce moment je me sens (Beaucoup mieux à Vraiment moins bien) CI, confidence Interval.
que la moyenne des gens n’en a l’air

Eigenvalue of factor 1 = 3.35. TABLE 4 | French Body Image State Scale items inter-correlations and item-total
Percentage of variance explained = 56%. correlations (ITC).

Item 1 Item 2 Item 3 Item 4 Item 5 Item 6 ITC

TABLE 2 | Communalities for the exploratory factor analysis of the F-BISS.
Item 1 –
Initial Extraction Item 2 0.26*** –
Item 3 0.72*** 0.21*** –
Item 1 0.70 0.82
Item 4 0.57*** 0.30*** 0.39*** –
Item 2 0.07 0.07
Item 5 0.50*** 0.15** 0.44*** 0.36*** –
Item 3 0.53 0.54
Item 6 0.55*** 0.24*** 0.41*** 0.65*** 0.42*** –
Item 4 0.56 0.55
ITC 0.85*** 0.53*** 0.76*** 0.74*** 0.65*** 0.74*** –
Item 5 0.38 0.42
Item 6 0.53 0.52 **p < 0.01, ***p < 0.001.

Frontiers in Psychology | www.frontiersin.org 6 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

study displayed stronger differences between groups, perhaps should have simply been translated as “silhouette” and not
explained by a possibly higher prevalence of female obesity “taille et forme du corps.” The latter translation was thought
in the United States than in France (Diouf et al., 2010; An, to be more literal, and closer to the original English phrasing.
2014). In that case, it would be beneficial to assess differences However, this item’s removal does not seem detrimental to
in overall BISS score between American and French samples; the overall validity of the scale, as borne out in section
even if western countries are supposed to display similar rates “Results,” making corrections probably unnecessary. Finally,
of body dissatisfaction (Holmqvist and Frisén, 2010), a higher EDE-Q cut-off scores used in this study were validated in a
obesity prevalence could be linked to higher body dissatisfaction Norwegian adult sample, not a French sample. While results
in one population over another (Weinberger et al., 2016). Finally, are coherent with literature, it would be needed to establish
results showing a significant difference between participants psychometric norms in a French population to provide more
with a clinically significant eating disorders symptomatology reliable cut-off scores. Moreover, a clinically significant EDE-
score are coherent with research establishing body dissatisfaction Q score cannot be considered an eating disorder diagnosis.
as a risk-factor of eating disorder development (Stice, 2002; Therefore, it would be needed in future research to assess
Stice et al., 2011). F-BISS scores in women with and without clinically established
Test–retest reliability was also slightly higher than the original eating disorders.
scale with a 2-week interval (r = 0.69; Cash et al., 2002), and Future studies should focus on replicating the original study’s
comparable to the first test–retest coefficient from the Italian protocol, and test the F-BISS in different situational contexts,
validation with an hour interval (r = 0.87; Carraro et al., 2010). such as a day on the beach, or reading a fashion magazine alone.
This may imply that this translation is more stable than expected. Indeed, the F-BISS was only tested in a neutral (questionnaire
In both our study and the original study, test–retest reliability was filling) situation without any physical intervention of the research
assessed in a neutral “questionnaire-filling” context. However, in team, which could explain its relative stability. When the
our study, participants had to fill the questionnaire at home using situational context is positive for body image, F-BISS scores
a computer, while they had to come to a laboratory in the original should be higher, indicating body satisfaction (Cash et al., 2002).
study. Perhaps the presence of the research team in the original When the situational context is negative or threatening for
study could have induced social comparison of appearance before body image, F-BISS scores should be lower. Other measures of
the experiment, causing a bias as social comparison of appearance convergent validity should also be used, such as indicators of
is linked to lesser body satisfaction (Myers and Crowther, 2009). thin-ideal internalization. The SATAQ-4R (Schaefer et al., 2017)
State-mood, a variable we have not controlled, has been shown to or the DKB-35 (Zohar et al., 2017; Lev-Ari et al., 2020), another
have an effect on state body dissatisfaction; when an individual trait-measure of body image, would be adequate choices. Another
feels better, their body dissatisfaction tends to be lower (Colautti measure of convergent validity that should be used is state-mood.
et al., 2011). With this in mind, it could be more stable in As said earlier, state-mood is positively linked to state body-
a familiar environment like the one in our study than in a dissatisfaction (Colautti et al., 2011). Conducting further analyses
laboratory context. Moreover, we have not verified if scores would in other ethnic groups, age groups, genders, or other French-
remain stable after subjecting participants to different contexts, speaking samples, such as Belgian or French-Canadian samples,
like in the original study (Cash et al., 2002). Finally, perhaps could prove worthwhile for generalization. Finally, validation in
testing the scale with shorter retest intervals could be appropriate, an eating disorder diagnosed population and comparison with a
to reduce eventual lability (Polit, 2014). general population would allow verification of our t-test results
The research conducted presents some limitations. Firstly, in a clinical setting.
it has not been tested in a male population, unlike every In a research setting, this scale could be used to identify,
other validation study. This poses a concern with regard to for instance, the effect of exposure to body image-threatening
generalization. As mentioned, the reason for men’s exclusion content in 18–25-year-old females. In a clinical setting, such as
is that a meta-analysis has shown that men present higher eating disorder treatment, its ability to measure state body image
dissatisfaction about muscularity (Karazsia et al., 2017). This satisfaction would allow for enhanced monitoring of the patient’s
study would have required other measures of convergent validity state, and a more comprehensive understanding of the daily life
for men, as the scales used were focused on thin appearance situations that increase or decrease body satisfaction.
and weight loss, and were validated in a female population
(Dany and Urdapilleta, 2012; Lentillon-Kaestner et al., 2014;
Carrard et al., 2015). Secondly, the research has never been DATA AVAILABILITY STATEMENT
tested within a broader age group. This should be corrected
in future studies by using appropriate scales. Indeed, older The raw data supporting the conclusions of this article will be
women are also affected by body-image issues (Marshall et al., made available by the authors, without undue reservation.
2012), despite being less thinness-driven than younger women
(Pruis and Janowsky, 2010). Thirdly, contrary to the Spanish
translation (Mebarak Chams et al., 2019), item 2 had to be ETHICS STATEMENT
deleted due to insufficient factor loading and low inter-item
correlations. This item made reference to body size and shape. The studies involving human participants were reviewed
It was surmised after the study that “body size and shape” and approved by the Université de Lille ethical

Frontiers in Psychology | www.frontiersin.org 7 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

review board (ref: 2019-345-S71). The patients/participants FUNDING

provided their written informed consent to participate
in this study. This study was supported by the PSITEC laboratory (which
typography and referral is Univ. Lille, ULR 4072 – PSITEC –
Psychologie: Interactions, Temps, Emotions, Cognition, F3 59000
Lille, France) and the Direction d’Appui à la Recherche of the
AUTHOR CONTRIBUTIONS University of Lille (file number: 4839).

LB: original draft, methodology, recruitment, conceptualization,

and reviewing and editing. CA: reviewing and editing, ACKNOWLEDGMENTS
recruitment, methodology, and conceptualization. CB and FL:
reviewing and editing and formal analysis. AR: data supervision We wish to thank everybody who volunteered to take part in
and curation, project administration, reviewing and editing, our study, as well as our undergraduate students for participating
and formal analysis. All authors contributed to the article and in data collection. We also wish to thank Brian Stacy for
approved the submitted version. proofreading this manuscript.

REFERENCES Costello, A. B., and Osborne, J. (2005). Best practices in exploratory factor analysis:
four recommendations for getting the most from your analysis. Pract. Assess.
Alleva, J. M., Martijn, C., Jansen, A., and Nederkoorn, C. (2014). Body language: Res. Eval. 10, 1–9. doi: 10.7275/jyj1-4868
affecting body satisfaction by describing the body in functionality terms. Council of Europe (n.d.). The CEFR Levels. Available Online at: https:
Psychol. Women Q. 38, 181–196. doi: 10.1177/0361684313507897 //www.coe.int/en/web/common-european-framework-reference-languages/
An, R. (2014). Prevalence and trends of adult obesity in the US, 1999–2012. ISRN level-descriptions (accessed September 5, 2020).
Obes. 2014:185132. doi: 10.1155/2014/185132 Dany, L., and Urdapilleta, I. (2012). Validation of a French measure of body
Archinard, M., Rouget, P., Painot, D., and Liengne, C. (1996). “Inventaire des comparison: the physical appearance comparison scale. Rev. Int. Psychol. Soc.
troubles du comportement alimentaire 2,” in Protocoles et Échelles D’évaluation 25, 97–112.
en Psychiatrie et en Psychologie, 1st Edn, eds M. Bouvard and J. Cottraux Diouf, I., Charles, M. A., Ducimetière, P., Basdevant, A., Eschwege, E., and Heude,
(Issy-les-Moulineaux: Masson), 275–276. B. (2010). Evolution of obesity prevalence in France: an age-period-cohort
Bateson, M., Cornelissen, P. L., and Tovée, M. J. (2007). “Methodological issues in analysis. Epidemiology 21:360. doi: 10.1097/EDE.0b013e3181d5bff5
studies of female attractiveness,” in The Body Beautiful, eds V. Swami and A. Duchesne, A. P., Dion, J., Lalande, D., Bégin, C., Émond, C., Lalande, G.,
Furnham (London: Palgrave Macmillan), 46–62. et al. (2017). Body dissatisfaction and psychological distress in adolescents:
Boersma, K. E., and Jarry, J. L. (2013). The paradoxical moderating effect of body is self-esteem a mediator? J. Health Psychol. 22, 1563–1569. doi: 10.1177/
image investment on the impact of weight-based derogatory media. Body Image 1359105316631196
10, 200–209. doi: 10.1016/j.bodyim.2012.11.002 Etu, S. F., and Gray, J. J. (2010). A preliminary investigation of the relationship
Bornioli, A., Lewis-Smith, H., Smith, A., Slater, A., and Bray, I. (2019). Adolescent between induced rumination and state body image dissatisfaction and anxiety.
body dissatisfaction and disordered eating: predictors of later risky health Body Image 7, 82–85. doi: 10.1016/j.bodyim.2009.09.004
behaviours. Soc. Sci. Med. 238:112458. doi: 10.1016/j.socscimed.2019.112458 Evans, C., and Dolan, B. (1993). Body shape questionnaire: derivation of shortened
Bucchianeri, M. M., and Neumark-Sztainer, D. (2014). Body dissatisfaction: an “alternate forms”. Int. J. Eat. Disord. 13, 315–321.
overlooked public health concern. J. Public Ment. Health 13, 64–69. doi: 10. Everitt, B. S. (1975). Multivariate analysis: the need for data, and other problems.
1108/JPMH-11-2013-0071 Br. J. Psychiatry 126, 237–240. doi: 10.1192/bjp.126.3.237
Carrard, I., Rebetez, M. M., Mobbs, O., and Van der Linden, M. (2015). Factor Fairburn, C. G. (2008). Cognitive Behavior Therapy and Eating Disorders.
structure of a French version of the eating disorder examination-questionnaire New York, NY: Guilford Press.
among women with and without binge eating disorder symptoms. Eat. Weight Fairburn, C. G., and Beglin, S. J. (1994). Assessment of eating disorders: interview
Disord. 20, 137–144. doi: 10.1007/s40519-014-0148-x or self−report questionnaire? Int. J. Eat. Disord. 16, 363–370.
Carraro, A., Nart, A., and Scarpa, S. (2010). Effects of a single session of physical Friborg, O., Reas, D. L., Rosenvinge, J. H., and Rø, Ø (2013). Core pathology of
exercise on body state image. Rev. Bras. Cienc. Esporte 32, 173–184. doi: 10. eating disorders as measured by the eating disorder examination questionnaire
1590/S0101-32892010000400012 (EDE-Q): the EATING DISORDERS 11 predictive role of a nested general (g)
Cash, T. F., Fleming, E. C., Alindogan, J., Steadman, L., and Whitehead, A. (2002). and primary factors. Int. J. Methods Psychiatr. Res. 22, 1–10. doi: 10.1002/mpr.
Beyond body image as a trait: the development and validation of the Body image 1389
States Scale. Eat. Disord. 10, 103–113. doi: 10.1080/10640260290081678 Garner, D. M., Olmstead, M. P., and Polivy, J. (1983). Development and validation
Cha, E. S., Kim, K. H., and Erlen, J. A. (2007). Translation of scales in of a multidimensional eating disorder inventory for anorexia nervosa and
cross−cultural research: issues and techniques. J. Adv. Nurs 58, 386–395. doi: bulimia. Int. J. Eat. Disord. 2, 15–34.
10.1111/j.1365-2648.2007.04242.x George, D., and Mallery, M. (2010). SPSS for Windows Step by Step: A Simple Guide
Child, D. (2006). The Essentials of Factor Analysis, 3rd Edn. New York, NY: and Reference, 17.0 update, 10th Edn. London: Pearson.
Continuum. Girard, M., Rodgers, R. F., and Chabrol, H. (2018). Prospective predictors of
Cohen, R., Fardouly, J., Newton-John, T., and Slater, A. (2019). # BoPo on body dissatisfaction, drive for thinness, and muscularity concerns among young
Instagram: an experimental investigation of the effects of viewing body positive women in France: a sociocultural model. Body Image 26, 103–110. doi: 10.1016/
content on young women’s mood and body image. New Med. Soc. 21, 1546– j.bodyim.2018.07.001
1564. doi: 10.1177/1461444819826530 Griffiths, S., Murray, S. B., Bentley, C., Gratwick-Sarll, K., Harrison, C., and Mond,
Colautti, L. A., Fuller-Tyszkiewicz, M., Skouteris, H., McCabe, M., Blackburn, S., J. M. (2017). Sex differences in quality of life impairment associated with
and Wyett, E. (2011). Accounting for fluctuations in body dissatisfaction. Body body dissatisfaction in adolescents. J. Adolesc. Health 61, 77–82. doi: 10.1016/
Image 8, 315–321. doi: 10.1016/j.bodyim.2011.07.001 j.jadohealth.2017.01.016
Cooper, P. J., Taylor, M. J., Cooper, Z., and Fairburn, C. G. (1987). The development Holmqvist, K., and Frisén, A. (2010). Body dissatisfaction across cultures: findings
and validation of the body shape questionnaire. Int. J. Eat. Disord. 6, 485–494. and research problems. Eur. Eat. Disord. Rev. 18, 133–146. doi: 10.1002/erv.965

Frontiers in Psychology | www.frontiersin.org 8 October 2021 | Volume 12 | Article 724710

Bardi et al. Translation and Validation: BISS

Hooper, D., Coughlan, J., and Mullen, M. R. (2008). Structural equation modelling: Rousseau, A., Knotter, R. M., Barbe, R. M., Raich, R. M., and Chabrol, H. (2005).
guidelines for determining model fit. Electronic J. Bus. Res. Methods 6, 53–60. Étude de validation de la version française du body shape questionnaire.
Karazsia, B. T., Murnen, S. K., and Tylka, T. L. (2017). Is body dissatisfaction Encephale 31, 162–173. doi: 10.1016/S0013-7006(05)82383-8
changing across time? A cross-temporal meta-analysis. Psychol. Bull. 143, Sarwer, D. B., Wadden, T. A., Moore, R. H., Eisenberg, M. H., Raper, S. E., and
293–320. doi: 10.1037/bul0000081 Williams, N. N. (2010). Changes in quality of life and body image after gastric
Lachaud, M., Miquel, V., Rousseau, A., and Chabrol, H. (2004). Pratiques bypass surgery. Surg. Obes. Relat. Dis. 6, 608–614. doi: 10.1016/j.soard.2010.07.
alimentaires et préoccupations corporelles chez les jeunes femmes. 015
J. Ther. Comportementale Cogn. 14, 131–134. doi: 10.1016/S1155-1704(04) Schaefer, L. M., Harriger, J. A., Heinberg, L. J., Soderberg, T., and Kevin Thompson,
97460-5 J. (2017). Development and validation of the sociocultural attitudes towards
Lentillon-Kaestner, V., Berchtold, A., Rousseau, A., and Ferrand, C. (2014). Validity appearance questionnaire−4−revised (SATAQ−4R). Int. J. Eat. Disord. 50,
and reliability of the French versions of the body shape questionnaire. J. Person. 104–117. doi: 10.1002/eat.22590
Assess. 96, 471–477. doi: 10.1080/00223891.2013.843537 Shah, N. R., and Braverman, E. R. (2012). Measuring adiposity in patients: the
Lev-Ari, L., Zohar, A. H., and Bachner-Melman, R. (2020). Enjoying your body: utility of body mass index (BMI), percent body fat, and leptin. PLoS One
the psychometric properties of an English version of the Dresden Body Image 7:e33308. doi: 10.1371/journal.pone.0033308
Questionnaire. Aust. J. Psychol. 72, 267–275. Stice, E. (2002). Risk and maintenance factors for eating pathology: a meta-analytic
Marshall, C., Lengyel, C., and Utioh, A. (2012). Body dissatisfaction: among review. Psychol. Bull. 128, 825–848. doi: 10.1037/0033-2909.128.5.825
middle-aged and older women. Can. J. Diet. Pract. Res. 73, 241–247. doi: 10. Stice, E., Marti, C. N., and Durant, S. (2011). Risk factors for onset of eating
3148/73.2.2012.e241 disorders: evidence of multiple risk pathways from an 8-year prospective study.
Mebarak Chams, M. R., Tinoco, L., Mejia-Rodriguez, D., Martinez-Banfi, M. L., Behav. Res. Ther. 49, 622–627. doi: 10.1016/j.brat.2011.06.009
Preuss, H., Hammerle, F., et al. (2019). The spanish body image state scale: Tabachnick, B. G., and Fidell, L. S. (2007). Using Multivariate Statistics, 5th Edn.
factor structure, reliability and validity in a colombian population. Front. Boston, MA: Allyn & Bacon.
Psychol. 10:2553. doi: 10.3389/fpsyg.2019.02553 Thompson, J. K., Heinberg, L., and Tantleff-Dunn, S. (1991). The physical
Moussally, J. M., Grynberg, D., Goffinet, S., Simon, Y., and Van der Linden, M. appearance comparison scale. Behav. Therap. 14:174.
(2017). Novel assessment of own and ideal body perception among women: Thompson, M. A., and Gray, J. J. (1995). Development and validation of a new
validation of the computer-generated figure rating scale. Cogn. Ther. Res. 41, body-image assessment scale. J. Person. Assess. 64, 258–269. doi: 10.1207/
632–644. doi: 10.1007/s10608-016-9827-4 s15327752jpa6402_6
Myers, T. A., and Crowther, J. H. (2009). Social comparison as a predictor of Tiggemann, M., and Zaccardo, M. (2015). « Exercise to be fit, not skinny »: the
body dissatisfaction: a meta-analytic review. J. Abnorm. Psychol. 118, 683–698. effect of fitspiration imagery on women’s body image. Body Image 15, 61-67.
doi: 10.1037/a0016763 doi: 10.1016/j.bodyim.2015.06.003
Ochner, C. N., Gray, J. A., and Brickner, K. (2009). The development and initial Valls, M., Rousseau, A., and Chabrol, H. (2013). Influence des médias, insatisfaction
validation of a new measure of male body dissatisfaction. Eat. Behav. 10, envers le poids et l’apparence et troubles alimentaires selon le genre. Psychol. Fr.
197–201. doi: 10.1016/j.eatbeh.2009.06.002 58, 229–240. doi: 10.1016/j.psfr.2013.06.002
Peterson, R. A. (2000). A meta-analysis of variance accounted for and factor Walker, D. C., Murray, A. D., Lavender, J. M., and Anderson, D. A. (2012). The
loadings in exploratory factor analysis. Mark. Lett. 11, 261–275. doi: 10.1023/A: direct effects of manipulating body checking in men. Body Image 9, 462–468.
1008191211004 doi: 10.1016/j.bodyim.2012.06.001
Polit, D. F. (2014). Getting serious about test–retest reliability: a critique of retest Weinberger, N. A., Kersting, A., Riedel-Heller, S. G., and Luck-Sikorski, C. (2016).
research and some recommendations. Qual. Life Res. 23, 1713–1720. doi: 10. Body dissatisfaction in individuals with obesity compared to normal-weight
1007/s11136-014-0632-9 individuals: a systematic review and meta-analysis. Obes. Facts 9, 424–441.
Pope, H. G. Jr., Gruber, A. J., Mangweth, B., Bureau, B., Decol, C., Jouvent, R., et al. doi: 10.1159/000454837
(2000). Body image perception among men in three countries. Am. J. Psychiatry Zohar, A. H., Lev-Ari, L., Bachner-Melman, R., and Kreitler, S. (2017). The
157, 1297–1301. doi: 10.1176/appi.ajp.157.8.1297 well rounded body image: the dresdner körperbildfragebogen DKB-35. Isr. J.
Pruis, T. A., and Janowsky, J. S. (2010). Assessment of body image in younger Psychiatry Relat. Sci. 54, 22–27.
and older women. J. Gen. Psychol. 137, 225–238. doi: 10.1080/00221309.2010.
484446 Conflict of Interest: The authors declare that the research was conducted in the
Rivière, J., Rousseau, A., and Douilliez, C. (2018). Effects of induced rumination on absence of any commercial or financial relationships that could be construed as a
body dissatisfaction: is there any difference between men and women? J. Behav. potential conflict of interest.
Ther. Exp. Psychiatry 61, 1–6. doi: 10.1016/j.jbtep.2018.05.005
Rø, Ø, Reas, D. L., and Stedal, K. (2015). Eating disorder examination questionnaire Publisher’s Note: All claims expressed in this article are solely those of the authors
(EDE−Q) in Norwegian adults: discrimination between female controls and and do not necessarily represent those of their affiliated organizations, or those of
eating disorder patients. Eur. Eat. Disord. Rev. 23, 408–412. doi: 10.1002/erv. the publisher, the editors and the reviewers. Any product that may be evaluated in
2372 this article, or claim that may be made by its manufacturer, is not guaranteed or
Rodgers, R., Chabrol, H., and Paxton, S. J. (2011). An exploration of the tripartite endorsed by the publisher.
influence model of body dissatisfaction and disordered eating among Australian
and French college women. Body Image 8, 208–215. doi: 10.1016/j.bodyim.2011. Copyright © 2021 Bardi, Arnaud, Bagès, Langlois and Rousseau. This is an open-
04.009 access article distributed under the terms of the Creative Commons Attribution
Rosseel, Y. (2012). Lavaan: an R package for structural equation modeling and License (CC BY). The use, distribution or reproduction in other forums is permitted,
more. Version 0.5–12 (BETA). J. Stat. Softw. 48, 1–36. doi: 10.18637/jss.v048.i02 provided the original author(s) and the copyright owner(s) are credited and that the
Rousseau, A., Denieul, M., Lentillon, V., and Valls, M. (2014). French validation original publication in this journal is cited, in accordance with accepted academic
of the male body dissatisfaction scale in a sample of young men. J. Ther. practice. No use, distribution or reproduction is permitted which does not comply
Comportementale Cogn. 24, 122–129. doi: 10.1016/j.jtcc.2014.07.001 with these terms.

Frontiers in Psychology | www.frontiersin.org 9 October 2021 | Volume 12 | Article 724710