Content uploaded by Albert J. Ksinan
Author content
All content in this area was uploaded by Albert J. Ksinan on Jan 17, 2017
Content may be subject to copyright.
Running head: WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES
The Effect of Belief in Weight Controllability on Anti-fat Attitudes: An Experimental
Manipulation
Albert J Ksinan
University of Kentucky
Carlos A. Almenara, Martin Vaculík
Masaryk University
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 2
The Effect of Belief in Weight Controllability on Anti-fat Attitudes: An Experimental
Manipulation
In 2008, the World Health Organization (WHO) declared the existence of an obesity
epidemic (James, 2008). Obesity rates in the Western World have been rising sharply over the
past twenty years, and at present time, a historically large number of Western people are
overweight (Ng et al., 2014).
People who are “fat” (a preferred term) are commonly target of prejudice and
discrimination and these negative experiences have been identified as a potential source related
to adverse health outcomes (Puhl & Suh, 2015). A term commonly used in the literature to
describe these negative attitudes toward fat people is “anti-fat attitudes.” Some authors have
suggested that these negative attitudes are the basis of “one of the last socially acceptable forms
of discrimination” (Fairburn & Brownell, 2002, p. 108), because this type of derision can be
readily found in mainstream media, such as television (Himes & Thompson, 2007) or print
media (Geier, Schwartz, & Brownell, 2003). The aim of the current research was to examine
anti-fat attitudes among adults in the Czech Republic and to evaluate the effect of a novel
experimental manipulation regarding the levels of anti-fat attitudes.
Anti-fat Attitudes
There are three main constructs underlying weight bias: prejudices, stereotypes, and
discrimination against fat people, which correspond to the attitudes, beliefs, and behavioral
components of weight bias, respectively (Lee, Ata, & Brannick, 2014). In this sense, anti-fat
attitudes consist of negative evaluations of fat people (prejudices), commonly based on
convictions about the causes of obesity (stereotypes). One important source of anti-fat attitudes is
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 3
the weight controllability belief (i.e., the belief that body weight is fully under volitional control
of an individual) and it is suggested that this attribution to personal responsibility serves as a
justification for prejudice (Crandall, 1994).
Studies have found several stereotypes typically connected with fat people, usually
describing them as “mean, stupid, ugly, unhappy, less competent, sloppy, lazy” (Puhl &
Brownell, 2006, p. 1802), and have also observed that weight discrimination occurs almost as
often as racial discrimination and sometimes even more frequently than age or gender
discrimination (Puhl, Andreyeva, & Brownell, 2008). The prejudice against fat people seems to
develop in early childhood (Cramer & Steinwert, 1998; Richardson, Goodman, Hastorf, &
Dornbusch, 1961), and is widespread within society. It is found in the workplace, among health
care professionals, and even in close interpersonal relationships (Puhl & Heuer, 2009). The
consequences of weight stigma include negative psychosocial and physiological outcomes (e.g.,
diminished motivation, high physiological reactivity to stress) that can interfere with the well-
being of obese individuals (Puhl & Suh, 2015).
Reducing Anti-fat Attitudes
A number of studies have examined the effect of changing the perception of weight
controllability. In one of the first experimental studies regarding the belief in weight
controllability, DeJong (1980) found that anti-fat attitudes are higher when obese people are seen
as responsible for their condition than when obesity is explained by other reasons, such as a
thyroid disease (DeJong, 1980). Later, Crandall (1994) used an experimental design in which the
experimental group was exposed to written materials stressing the genetics of weight control. By
contrast, the control group was presented with an unrelated essay about the role of psychological
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 4
stress on illness. The results showed lower anti-fat attitudes in the experimental group as
compared to control group, as measured by the Anti-fat Attitudes Questionnaire (AFA). Overall,
the results of experimental studies examining the effect of manipulation of weight controllability
beliefs on anti-fat attitudes show mixed findings (Daníelsdóttir, O’Brien, & Ciao, 2010).
Presenting information to participants about the inability to fully control one’s weight has been
found to be related to lower anti-fat attitudes among experimental groups compared to control
groups in several studies (Blaine & Williams, 2004; Diedrichs & Barlow, 2011; O’Brien, Puhl,
Latner, Mir, & Hunter, 2010; Weiner, Perry, & Magnusson, 1988), but not in others (Harris,
Walters, & Waschull, 1991; Teachman, Gapinski, Brownell, Rawlins, & Jeyaram, 2003;
Thorsteinsson, Loi, & Breadsell, 2016). For example, a study by Teachman et al. (2003) did not
find lower implicit or explicit bias when priming participants (N = 144) with an article describing
genetic causes of obesity compared with control group. Mixed findings in past literature could be
due to diverse methodologies. Regarding this study, we decided to improve the design of
previous studies, as described in detail in the procedure section below.
The Present Study
The present study was done in the Czech Republic, a country that was a part of the
communist block until 1989. In the post-communist era, the country has transitioned to
democracy and has experienced several changes that seem to correspond with “the progressive
integration of economies and societies” that primarily characterizes globalization (Kawachi &
Wamala, 2007). These shifts included a variety of transitions comprising epidemiologic,
demographic, and nutritional aspects that may partially explain the increases in the prevalence of
obesity (Ancona Lopez & Ridel Juzwiak, 2010; Garman, 2007). Data collected by the WHO
indicate that obesity among adults in the Czech Republic have increased sharply from 14% in
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 5
2000 to 21% in 2011 (OECD, 2013), whereas the prevalence among adolescents almost doubled
from 5.5% in 1998 to 10.4% in 2010 (Sigmundová, El Ansari, Sigmund, & Frömel, 2011). When
including both overweight and obese, the prevalence is around 55% among adults in the Czech
Republic (Ng et al., 2014).
Anti-fat attitudes in the Czech Republic appear to be similar to those in other Western
countries. For instance, a study on body image attitudes found that Czech girls as young as 6
years old already express negative attitudes toward overweight bodies (Fraňková & Chudobová,
2000). This negative attitude toward fat people has also been recently observed in another study
that found that 38% of adolescents reported being victims of weight-related teasing (Almenara &
Ježek, 2015). In particular, this study found that overweight Czech adolescents were more than
three and a half times more likely to report body shape and body weight teasing as compared to
their non-overweight counterparts (Almenara & Ježek, 2015).
Although these studies suggest that the situation in the Czech Republic is similar to other
countries in rates of obesity as well as anti-fat attitudes, no study so far has performed an
experimental manipulation of anti-fat attitudes in individuals from a non-Western country.
Therefore, considering the role of belief in weight controllability on anti-fat attitudes, the present
study strived to experimentally assess the effect of a text containing either a weight-
controllability or a weight-uncontrollability message on individuals’ anti-fat attitudes in a sample
of Czech adults.
Regarding the second goal, we tested the following hypotheses. First, we expected that
the rates of anti-fat attitudes in the sample would be similar to previous studies done on this
topic. Second, by significantly improving the design of the experimental manipulation by making
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 6
the experimental message resemble a news article and carefully concealing the true nature of the
study, we expected that this would yield a significant effect on anti-fat attitudes. Specifically, we
hypothesized that the group exposed to a message describing weight as controllable would show
higher anti-fat attitudes compared to a control group. By contrast, we expected that the group
exposed to a message describing weight as uncontrollable would show lower anti-fat attitudes
compared to a control group.
Method
Participants
In total, 673 participants were initially recruited. Of these, 51 participants were dropped
due to their failure to provide sufficient evidence that they have read the priming text carefully.
Then, another 42 participants were removed as they were able to guess the true purpose of the
study (see Procedure section below). Therefore, the final sample included 580 participants, 208
men (35.9%) and 372 women (64.1%). These were randomly assigned into three groups (see
below) in the following way: willpower condition (n = 185), genetic condition (n = 193), and
control condition (n = 202).
The mean age of the final sample was 22.19 years (SD = 5.26). There were 464 Czech
(80.0%) and 116 Slovak (20.0%) respondents. Regarding education, the majority reported high
school as the highest attained education (84.3%), followed by some college degree (15.2%), and
elementary school (0.5%). Most respondents indicated that they were studying (70.3%),
followed by both working and studying (22.8%), working (5.3%), and neither studying nor
working (1.6%).
Procedure
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 7
The online survey and random assignment was possible using Qualtrics - an online
website specially designed for online surveys (www.qualtrics.com). Invitations to participate in
a study were sent as a link via the Internet. Three strategies were used, namely creating an event
on Facebook, by posting an invitation on an Internet forum, and by sending an email with an
Internet link to the online survey to a random set of students from a large Czech university. The
link was accompanied by few sentences asking the recipients to participate in the study and
stressing the complete anonymity of their responses. Local ethical guidelines were followed in
the treatment of the study sample.
Participants accessing the online survey were given a description of the survey as
“opinions about current issues in medicine,” and were asked to give informed consent while
being informed about the anonymity of their participation. Then, participants were randomly
assigned to one of three conditions: willpower condition, genetic condition, and control group.
In each condition, participants at the beginning of the survey were asked to “read a one-page
scientific article about the topic of health”.
Each condition was described as follows: (1) Willpower condition, in which the content
of the article indicated that obesity is the result of lacking willpower and lack of exercise (i.e.,
body weight is controllable); (2) Genetic condition, where the article emphasized that the
primary cause of obesity is genetically based (i.e., body weight is not controllable); and (3)
Control condition, where the article described chlamydia infection (i.e., unrelated topic). A
similar procedure has been used before (Teachman et al., 2003), but the procedure in the current
study was expanded upon in the following ways. First, we provided longer and more detailed
articles (priming texts), approximately one-page long. Second, the articles were matched by
their content: in both experimental conditions (willpower/genetics), there was an equal number
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 8
of phrases conveying the idea that weight is a result of willpower/genetics. The style of the
articles was adjusted to resemble a news report containing information about the latest scientific
findings. Third, two distractors were used to mask the real purpose of the study, as this
knowledge might bias participants’ responses. First, at the beginning of the online survey,
participants were presented with the following message: “please wait for the system to choose
an article for you,” followed by a 5-second waiting time to simulate a “loading screen”. This
mechanism was meant to convey to participants that the article was being randomly chosen
from among many others. Second, the survey showed a heading (e.g., “Version 5” or “Version
12”) to further reinforce the idea that the article was randomly selected from a pool of other
seeming possibilities. Finally, to ensure that participants had read and understood the text, they
were asked to answer yes/no questions related to the content of the article they were presented
(e.g., “Obesity is mostly a result of genetic factors”). At this point, 51 participants were
excluded because they did not correctly answer at least four out of five questions.
Following the first phase outlined above, participants continued with completing the
attitudinal measurements. First, they were presented with the Anti-Fat Attitudes Questionnaire
(AFA; Crandall, 1994). To further obscure the real purpose of the study, the AFA items were
mixed up with other items asking about attitudes towards smokers and alcoholics. The majority
of these “fill” items were adapted from two studies, one (Vargas & Luis, 2008) about the
attitudes toward alcohol, alcoholics, and alcoholism (11 questions out of 13), and the other
(Park, Mitra, & Asch, 2012) about attitudes toward smokers and smoking (4 questions out of
11). The rest of the items were developed by the first author of this study. This procedure did
not affect the results, as no significant effect of group belonging to attitudes towards alcohol (p
< .962) or smoking (p < .841) was found.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 9
After this section, participants were asked to try to describe the purpose of the study in
their own words, by imagining how to describe it to a friend. This item served as a check to
determine whether participants were aware of the real purpose of the study. The results of this
item suggested that 42 participants were able to guess the purpose of the study (7.1% of the
sample). Thus, these 42 participants were also excluded from further analyses.
Finally, participants were given a thorough debriefing describing the real purpose of the
study along with a contact email of the first author of the study should they have concerns or
questions.
Measures
Anti-fat attitudes. Anti-fat Attitudes (AFA) questionnaire was originally developed by
Crandall (1994) to measure attitudes toward fatness and fat people. The AFA scale is a 13-item
questionnaire comprising three subscales: Dislike, Fear of fat, and Willpower (see the online
supplement for the full measure). The first subscale, Dislike (7 items), is an evaluative subscale
measuring dislike of fat people (e.g., “I really don't like fat people much”). Second subscale,
Fear of fat (3 items), measures concerns about own weight and body shape (e.g., “I feel
disgusted with myself when I gain weight”). Finally, Willpower subscale (3 items) reflects the
beliefs about the controllability of weight and fat (e.g., “Fat people tend to be fat pretty much
through their own fault.”). The items are scored on a Likert-type scale ranging from 1 (very
strongly disagree) to 9 (very strongly agree). Scores for each subscale result from averaging
responses to their corresponding items. Higher scores indicate stronger anti-fat attitudes.
Since AFA has never been adapted to Czech language, it was first necessary to complete
a translation. The back-translation method was used (Brislin, 1970). First, three native Czechs
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 10
with expertise as English translators completed a translation of the AFA. Then, the first author
of this study compared and combined these three Czech versions into one. Second, a new
professional translator was asked to translate the final Czech version back to English without
any knowledge of the original version. This new English version was then compared with the
original one in order to detect possible shifts of meaning. No major changes were found.
Finally, six graduate students were contacted by email and asked to submit any comments on
potential uncertainties in the meaning or comprehension of the items. Their comments were
used to develop the final version of the questionnaire.
This newly-created Czech version of the AFA was then tested in a confirmatory factor
analysis framework (CFA) to validate its factor structure. Based on the original proposition of
the AFA (Crandall, 1994), the proposed model for CFA analysis consisted of three inter-
correlated latent variables (three AFA subscales) with 13 items as their respective indicators.
The results showed an adequate model fit: χ2(62) = 221.230, p < .001; CFI = .930, TLI = .912,
RMSEA = .067, 90% RMSEA CI [.057, .076], SRMR = .048, which confirms the AFA factor
structure based on commonly accepted criteria of model fit (Hu & Bentler, 1999). Standardized
estimates parameters (loadings and uniqueness) and latent factor correlations are presented in
Figure 1.
Covariates. Participants reported their sex (0 = male, 1 = female), age, level of
education (1 = elementary school, 2 = high school, 3 = college degree), work status (1 = work, 2
= study, 3 = work and study, 4 = neither work nor study), and nationality (0 = Czech, 1 =
Slovak).
Statistical Analysis
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 11
The data were examined using SPSS 21 software package. First, descriptive analyses,
correlations, and group comparison tests were completed. Then, two-step hierarchical regression
analyses were employed in order to evaluate the effect of the experimental manipulation on each
outcome variable (AFA subscales), above and beyond the effect of demographic variables
(Cohen, Cohen, West, & Aiken, 2003). A dummy code was used to code the experimental versus
control groups, where the control group served as the reference group. Where applicable,
covariate variables were entered in the first step of regression as a set. In the second step,
dummy codes for the willpower group (1= willpower, 0 = other) and genetic group (1 = genetic,
0 = other) were entered. The unstandardized regression coefficients for dummy codes represent
the score difference between means of the respective experimental group and the control group
on a given subscale. For example, B = .500, p <.001 means that the experimental group shows a
mean that is a significantly higher by .500 on a given dependent variable compared to control
group (and vice versa for a negative B value).
Results
Descriptive Statistics
In the first step, correlations were run among the study variables. A correlational matrix
of the variables of interest along with mean levels and reliabilities is provided in Table 1. The
three AFA subscales were significantly correlated: Fear of fat with Willpower (r = .22, p <.001)
and with Dislike (r = .20, p <.001), and Willpower with Dislike (r = .35, p <.001). Age was not
significantly related to any of the outcome variables. The subscales also showed good internal
consistency, as indexed by Cronbach’s alpha: Dislike = .80, Fear of fat = .85, Willpower = .64.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 12
Then, a series of t-tests was carried out to examine possible mean-level differences by
gender and nationality in outcome variables. Post-hoc Bonferroni corrections with a critical p-
level adjustment by the number of comparisons was used in order to control for the risk of Type
I error. A significant gender difference was observed only for the Fear of fat subscale t (578) =
-8.69, p < .001, indicating that females scored significantly higher (M = 6.18) than males (M =
4.40). No significant differences were found between Czech and Slovak participants in any of
the outcome variables. In addition, ANOVAs were executed to examine possible differences in
outcome variables for education or level of employment. A significant effect was found for
employment status for the Fear of fat subscale, F (3, 576) = 3.597, p = .013. For this reason,
employment status was included in the regression model for this outcome, represented by three
dummy codes with studying only status used as the reference group.
Effect of Belief in Weight Controllability
In the next step, three hierarchical regression analyses were run, one for each AFA
subscale. Since gender showed a significant effect on the Fear of fat subscale, it was included in
the first step of a two-step hierarchical regression in order to control for this covariate.
Moreover, work status dummy codes were also included in the first step of this regression.
Since work status is partially a function of age, age was also included to control for its effect.
Then, two dummy variables were added in the second step of the hierarchical regression,
representing participants’ inclusion in either the “willpower” or “genetic” group condition. The
results from these analyses are presented in Table 2.
For the Willpower subscale, the regression was significant F (2, 577) = 20.340, p < .001;
moreover, the estimates for both experimental conditions were also significant (B willpower =
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 13
0.63, p < .001, B genetic = -0.40, p = .012). These results indicate that people exposed to a
message in agreement with the weight controllability belief score significantly higher on
Willpower subscale than people exposed to an unrelated topic (i.e., chlamydia infection). By
contrast, people exposed to a message challenging the weight controllability belief score
significantly lower on Willpower subscale than people exposed to an unrelated topic. Significant
results were also obtained for the Fear of fat subscale, F (2, 573) = 3.144, p = .044. As shown in
Table 2, there was a significant difference between the willpower group and the control group
(B willpower = 0.53, p = .026) as well as between the genetic group and control group (B
genetic = 0.49, p = .038). These results indicate that people exposed to experimental conditions
where the content is related to weight control have higher levels of fear of fat than people
exposed to an unrelated content (i.e., chlamydia infection).
In contrast, the results for the Dislike subscale were non-significant, F (2, 577) = 0.406, p
= .666, showing no effect of any of the experimental manipulations (R2 = 0.0). Thus, this
indicates that the entire sample displays similar levels of anti-fat attitudes toward fat people,
independently of the experimental exposure.
Finally, due to gender having a significant effect on Fear of fat, moderation effects of
gender were examined. No significant gender*experimental manipulation effect was found for
any of the AFA subscales.
Discussion
The goal of this study was to experimentally evaluate the effect of a message describing
either controllability or non-controllability of weight on individuals’ anti-fat attitudes in a sample
of Czech adults.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 14
The moderate intercorrelations of the subscales confirm the notion that the three
subscales are not part of a general factor but rather independent, yet related constructs. The
correlations between variables of interest indicated some gender differences as women scored
higher on the Fear of fat subscale. This finding is in line with previous studies which found that
women generally face stronger sociocultural pressures to be thin, as compared to men (Schaefer
et al., 2015).
We hypothesized that the group exposed to a message describing weight as controllable
would show higher anti-fat attitudes (i.e., higher scores in AFA subscales) compared to a control
group. On the other hand, we expected that the group exposed to a message describing weight as
uncontrollable would show lower anti-fat attitudes compared to a control group. In regards to the
Willpower subscale, our results confirmed our hypothesis. They show that participants in the
willpower condition scored significantly higher than the control group, whereas the scores of
participants in the genetic condition were significantly lower. In other words, people exposed to a
message emphasizing willpower as the main cause of being fat (weight controllability belief),
displayed higher agreement with statements supporting this belief compared to a control group.
Conversely, people exposed to a message emphasizing how genes influence body weight
displayed lower levels of weight controllability belief compared to a control group. In contrast to
traditional views of stereotypes as somehow stable (Allport, 1954), stereotypes are seen
nowadays as flexible, dynamic, and susceptible to change (Eagly & Diekman, 2005). Although
this study did not evaluate a real change in individuals’ attitudes, our results suggest that the
exposure to information that is congruent/incongruent with the weight controllability belief
would lead to higher/lower adherence to this stereotype, respectively and compared with people
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 15
not exposed to this kind of information. Indeed, this assumption is in line with the results of most
of the interventions striving to reduce weight bias (Lee et al., 2014).
Regarding the Fear of fat subscale, the results were not completely as hypothesized. As
expected, participants in the willpower condition showed significantly higher scores in Fear of
fat than the control group. However, contrary to our predictions, scores for Fear of fat in the
genetic condition were not lower, but actually higher, than in the control group. Thus,
participants who were exposed to an experimental condition where the content of the priming
article described the basis of obesity (either controllable or uncontrollable), had higher levels of
fear of fat than people exposed to an unrelated content (chlamydia infection). In some previous
studies, it has been observed that people exposed to supraliminal stimuli (i.e., consciously
perceived stimuli) pertaining to physical appearance activate appearance-related self-schemas
(i.e., cognitive assumptions about the importance of one’s appearance in life; Cash, 2011;
Hausenblas et al., 2013; López-Guimerà, Levine, Sánchez-Carracedo, & Fauquet, 2010).
Furthermore, there is evidence of an association between attentional biases to appearance-related
stimuli and body dissatisfaction (Rodgers & DuBois, 2016). For this reason, it is possible that
participants exposed to our priming articles aimed to manipulate weight-controllability beliefs,
activated their appearance-related self-schemas which made them more body-conscious than
participants in the control group. In turn, this attention bias toward body weight could have
elicited weight and shape concerns as revealed by their higher scores in Fear of fat, as compared
to the control group.
On the other hand, we did not find any significant effect of the experimental conditions
on the scores in the Dislike subscale. This finding is in line with experimental studies that found
no differences between experimental and control groups in overall anti-fat prejudice, regardless
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 16
of experimental manipulation (e.g., Teachman et al., 2003). Considering that other interventions
have been successful in reducing weight-biased attitudes and increasing acceptance of obese
persons (e.g., Puhl, Schwartz, & Brownell, 2005), the divergence in findings could be a matter of
methodological differences (e.g., the type of stimuli used). More sophisticated, interactive and
longer interventions could possibly create a major effect. Nevertheless, it should be noted that a
recent review of interventions to reduce weight bias found that mean effects across intervention
types were small, and thus new strategies for reducing weight bias should be designed (Lee et al.,
2014). For instance, there is a growing interest in social equality, social justice and social change
within the study of disadvantaged groups (e.g., de Lemus & Stroebe, 2015; Stroebe, Wang, &
Wright, 2015). According to some authors, if we study intergroup relations (e.g., obese with non-
obese), we should also speak about group equality and focus on social justice rather than solely
on prejudice reduction, group harmony and social cohesion (Wright & Lubensky, 2009). This
perspective, named collective action, suggests that strategies like fostering direct action of
members of majority groups (e.g., non-obese) to remove social inequality, and engaging
disadvantaged group members (e.g., obese individuals) in these actions, can promote social
justice (Wright & Lubensky, 2009). Thus, perhaps it is necessary to broaden our perception of
weight bias as a problem solely of prejudice (microlevel), to perceive weight bias also as a social
justice issue (macrolevel). This could help us to adopt a more integrated perspective in the field
of eating- and weight related problems (see Russell-Mayhew & Grace, 2016).
Taken together, our results have important implications for clinical interventions aimed at
problematic eating- and weight-related attitudes and behaviors. First, clinical interventions
should consider that stimuli activating appearance-related cognitive schemas can motivate the
avoidance of fatness (i.e., fear of fat), as our results suggest. Then, one needs to bear in mind that
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 17
the avoidance of overweight stigma (i.e., fear of fat) has been found to be closely correlated with
the perceived rewards of being thin (approach bias towards thinness), and both were found to be
associated with problematic eating- and weight-related attitudes and behaviors (Dalley & Buunk,
2009; Mussap, 2007; Woud, Anschutz, Van Strien, & Becker, 2011). Thus, interventions should
consider that the manipulation of cognitive biases such as weight controllability beliefs might
actually backfire by rising weight concerns. Indeed, a recent review suggests that interventions in
the field of obesity and weight-management should consider avoiding messages centered on
weight control, and instead they should consider messages centered on helping people to
improve their general health and well-being beyond the trivial importance of physical appearance
(see Tylka et al., 2014). Moreover, it would be beneficial for these interventions if future studies
further examine the effect of diverse types of appearance-related stimuli on relevant attitudes and
behaviors such as weight bias, fear of fat, dietary restraint, and excessive preoccupation with
food, among others.
Limitations
It is important to stress that this study employed only a single assessment of the anti-fat
attitudes. Although the experimental manipulation led to significant differences among
experimental groups and control group for two out of three dependent variables, we did not
assess change within participants. Thus, it would be rather imprecise to interpret it as evidence
for either an increase or decrease in AFA. More studies employing repeated measures (pre- and
post-test) would be beneficial in shedding a light on the possibility of change.
Another potential issue might be that this study did not control for participant’s own
weight and/or their body perception, which might have influenced how they responded.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 18
However, existing studies on the effect of participant’s weight on their weight stigma show
mixed findings, where some studies report inverse relationship between weight status and anti-
fat attitudes (e.g., Schwartz, Vartanian, Nosek, & Brownell, 2006), while others studies did not
find this effect (e.g., Puhl et al., 2015). Moreover, this study is limited by the sample, which was
not representative of the whole population. Future studies in this area should employ samples
from a more general population or focus on other age groups, such as adolescents.
Nevertheless, the current study has also some strengths worth mentioning. Compared
with previous experimental studies with similar goals, this study was improved in several ways.
First, a large random sample was collected, providing adequate power to detect the effect of
manipulation and increases the generalizability of the findings. Second, the study used several
techniques (description of the study, “loading screen”, covering of the target items) to conceal
the true nature of the experiment in order to counter social desirability, which proved successful,
as only 35 participants were able to guess the true purpose of the study. Third, the participants
were presented with information about both controllability and uncontrollability of weight,
adapted to resemble a news report, in order to assess whether the controllability belief can be
affected in both directions. Finally, hierarchical linear regression provided several advantages
over traditionally used procedures in this type of research, such as the conventional analysis of
variance (ANOVA). Regression analyses gave us the ability to precisely estimate the direction of
the experimental manipulation, size of the effect, as well as control for confounding variables.
Conclusions
Overall, the results showed that attitudes toward fat people in the Czech Republic mirror
the findings from studies done in Western countries (e.g., Crandall, 1994; Crandall et al., 2001).
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 19
Particular care should be given to messages that can perpetuate negative or incorrect information
about fat people among the general population. This study attests to a significant effect of a
simple one-page article on participant’s perception of weight controllability and fear of fat. It is
necessary to realize that such an article, while seemingly “neutral”, might lead to increase in
anti-fat attitudes through its description of body weight etiology. We hope that this study will
lead to higher awareness of this issue, which would possibly translate to a more nuanced
reporting based on medical facts and free of possible implicit biases.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 20
Conflict of interest
Conflict of interest: none.
Acknowledgements
The authors acknowledge the support of the Central European Longitudinal Study of Pregnancy
and Childhood (CELSPAC) (MUNI/M/1075/2013), and the THINLINE Project (GA15-05696S)
which are co-financed by Masaryk University and the state budget of the Czech Republic.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 21
References
Allport, G. W. (1954). The nature of prejudice. Cambridge, MA: Addison-Wesley.
Almenara, C. A., & Ježek, S. (2015). The source and impact of appearance teasing: An
examination by sex and weight status among early adolescents from the Czech Republic.
Journal of School Health, 85, 163–170. http://doi.org/10.1111/josh.12236
Ancona Lopez, F., & Ridel Juzwiak, C. (2010). Social impact of modern diet and tendencies. In
R. Pinho Ferreira Guine (Ed.), Food, diet and health: Past, present and future tendencies
(pp. 27–58). New York, NY: Nova Science Publishers.
Blaine, B. E., & Williams, Z. (2004). Belief in the controllability of weight and attributions to
prejudice among heavyweight women. Sex Roles, 51, 79–84.
http://doi.org/10.1023/B:SERS.0000032315.95890.d9
Brislin, R. W. (1970). Back-translation for cross-cultural research. Journal of Cross-Cultural
Psychology, 1, 185–216. http://doi.org/10.1177/135910457000100301
Cash, T. F. (2011). Cognitive-behavioral perspectives on body image. In T. F. Cash & L. Smolak
(Eds.), Body image: A handbook of science, practice, and prevention (pp. 39–47). New
York, NY: The Guilford Press.
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation
analysis for the behavioral sciences (3rd ed.). Mahwah, NJ: Routledge.
Cramer, P., & Steinwert, T. (1998). Thin is good, fat is bad: How early does it begin? Journal of
Applied Developmental Psychology, 19, 429–451. http://doi.org/10.1016/S0193-
3973(99)80049-5
Crandall, C. S. (1994). Prejudice against fat people: Ideology and self-interest. Journal of
Personality and Social Psychology, 66, 882–894. http://doi.org/10.1037/0022-3514.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 22
Crandall, C. S., D’Anello, S., Sakalli, N., Lazarus, E., Nejtardt, G. W., & Feather, N. T. (2001).
An attribution-value model of prejudice: Anti-fat attitudes in six nations. Personality and
Social Psychology Bulletin, 27, 30-37. http://doi.org/10.1177/0146167201271003
Dalley, S. E., & Buunk, A. P. (2009). “Thinspiration” vs. “fear of fat”. Using prototypes to
predict frequent weight-loss dieting in females. Appetite, 52, 217–221.
http://doi.org/10.1016/j.appet.2008.09.019
Daníelsdóttir, S., O’Brien, K. S., & Ciao, A. (2010). Anti-fat prejudice reduction: a review of
published studies. Obesity Facts, 3, 47–58. http://doi.org/10.1159/000277067
de Lemus, S., & Stroebe, K. (2015). Achieving social change: A matter of all for one? Journal of
Social Issues, 71, 441–452. http://doi.org/10.1111/josi.12122
DeJong, W. (1980). the stigma of obesity: The consequences of naive assumptions concerning
the causes of physical deviance. Journal of Health and Social Behavior, 21, 75–87.
http://doi.org/10.2307/2136696
Diedrichs, P. C., & Barlow, F. K. (2011). How to lose weight bias fast! Evaluating a brief anti-
weight bias intervention. British Journal of Health Psychology, 16, 846–861.
http://doi.org/10.1111/j.2044-8287.2011.02022.x
Eagly, A. H., & Diekman, A. B. (2005). What is the problem? Prejudice as an attitude-in-context.
In J. F. Dovidio, P. Glick, & L. A. Rudman (Eds.), On the nature of prejudice: Fifty years
after Allport (pp. 19–35). Malden, MA: Blackwell Publishing.
Fairburn, C. G., & Brownell, K. D. (2002). Eating disorders and obesity: a comprehensive
handbook. New York, NY: Guilford Publications.
Fraňková, S., & Chudobová, P. (2000). Development of body image in preschool girls.
Homeostasis in Health and Disease, 40, 161–169.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 23
Garman, S. (2007). Trends and transitions: The sociopolitical context of public health. In A.
Scriven & S. Garman (Eds.), Public health: Social context and action (pp. 23–33). New
York, NY: Open University Press.
Geier, A. B., Schwartz, M. B., & Brownell, K. D. (2003). “Before and After” diet advertisements
escalate weight stigma. Eating and Weight Disorders, 8, 282–288.
http://doi.org/10.1007/BF03325027
Harris, M. B., Walters, L. C., & Waschull, S. (1991). Altering attitudes and knowledge about
obesity. The Journal of Social Psychology, 131, 881–884.
http://doi.org/10.1080/00224545.1991.9924675
Hausenblas, H. A., Campbell, A., Menzel, J. E., Doughty, J., Levine, M., & Thompson, J. K.
(2013). Media effects of experimental presentation of the ideal physique on eating
disorder symptoms: A meta-analysis of laboratory studies. Clinical Psychology Review,
33, 168–181. http://doi.org/10.1016/j.cpr.2012.10.011
Himes, S. M., & Thompson, J. K. (2007). Fat stigmatization in television shows and movies: A
content analysis. Obesity, 15, 712–718. http://doi.org/10.1038/oby.2007.635
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55.
http://doi.org/10.1080/10705519909540118
James, W. P. T. (2008). WHO recognition of the global obesity epidemic. International Journal
of Obesity, 32, S120–S126. http://doi.org/10.1038/ijo.2008.247
Kawachi, I., & Wamala, S. (2007). Globalization and health: Challenges and prospects. In I.
Kawachi & S. Wamala (Eds.), Globalization and health (pp. 3–15). New York, NY:
Oxford University Press.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 24
Lee, M., Ata, R. N., & Brannick, M. T. (2014). Malleability of weight-biased attitudes and
beliefs: A meta-analysis of weight bias reduction interventions. Body Image, 11, 251–259.
http://doi.org/10.1016/j.bodyim.2014.03.003
López-Guimerà, G., Levine, M. P., Sánchez-carracedo, D., & Fauquet, J. (2010). Influence of
mass media on body image and eating disordered attitudes and behaviors in females: A
review of effects and processes. Media Psychology, 13, 387–416.
http://doi.org/10.1080/15213269.2010.525737
Mussap, A. J. (2007). Motivational processes associated with unhealthy body change attitudes
and behaviours. Eating Behaviors, 8, 423–428.
http://doi.org/10.1016/j.eatbeh.2006.12.001
Ng, M., Fleming, T., Robinson, M., Thomson, B., Graetz, N., Margono, C., … Gakidou, E.
(2014). Global, regional, and national prevalence of overweight and obesity in children
and adults during 1980-2013: A systematic analysis for the Global Burden of Disease
Study 2013. The Lancet, 384, 766–781. http://doi.org/10.1016/S0140-6736(14)60460-8
O’Brien, K. S., Puhl, R. M., Latner, J. D., Mir, A. S., & Hunter, J. A. (2010). Reducing anti-fat
prejudice in preservice health students: a randomized trial. Obesity, 18, 2138–2144.
http://doi.org/10.1038/oby.2010.79
OECD. (2013). Health at a glance 2013: OECD Indicators. Paris: OECD Publishing.
Park, J. D., Mitra, N., & Asch, D. A. (2012). Public opinion about financial incentives for
smoking cessation. Preventive Medicine, 5, 41-45.
http://doi.org/10.1016/j.ypmed.2012.06.013
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 25
Puhl, R. M., Andreyeva, T., & Brownell, K. D. (2008). Perceptions of weight discrimination:
Prevalence and comparison to race and gender discrimination in America. International
Journal of Obesity, 32, 992–1000. http://doi.org/10.1038/ijo.2008.22
Puhl, R. M., & Brownell, K. D. (2006). Confronting and coping with weight stigma: an
investigation of overweight and obese adults. Obesity, 14, 1802–1815.
http://doi.org/10.1038/oby.2006.208
Puhl, R. M., & Heuer, C. A. (2009). The stigma of obesity: A review and update. Obesity, 17,
941–964. http://doi.org/10.1038/oby.2008.636
Puhl, R. M., Latner, J. D., O’Brien, K., Luedicke, J., Daníelsdóttir, S., & Forhan, M. (2015). A
multinational examination of weight bias: Predictors of anti-fat attitudes across four
countries. International Journal of Obesity, 39, 1166–1173.
http://doi.org/10.1038/ijo.2015.32
Puhl, R. M., Schwartz, M. B., & Brownell, K. D. (2005). Impact of perceived consensus on
stereotypes about obese people: a new approach for reducing bias. Health Psychology,
24, 517–525. http://doi.org/10.1037/0278-6133.24.5.517
Puhl, R. M., & Suh, Y. (2015). Health consequences of weight stigma: Implications for obesity
prevention and treatment. Current Obesity Reports, 4, 182–190.
http://doi.org/10.1007/s13679-015-0153-z
Richardson, S. A., Goodman, N., Hastorf, A. H., & Dornbusch, S. M. (1961). Cultural uniformity
in reaction to physical disabilities. American Sociological Review, 26, 241–247.
http://doi.org/10.2307/2089861
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 26
Rodgers, R. F., & DuBois, R. H. (2016). Cognitive biases to appearance-related stimuli in body
dissatisfaction: A systematic review. Clinical Psychology Review, 46, 1–11.
http://doi.org/10.1016/j.cpr.2016.04.006
Russell-Mayhew, S., & Grace, A. D. (2016). A call for social justice and best practices for the
integrated prevention of eating disorders and obesity. Eating Disorders, 24, 54–62.
http://doi.org/10.1080/10640266.2015.1113829
Schaefer, L. M., Burke, N. L., Thompson, J. K., Dedrick, R. F., Heinberg, L. J., Calogero, R. M.,
… Swami, V. (2015). Development and validation of the Sociocultural Attitudes Towards
Appearance Questionnaire-4 (SATAQ-4). Psychological Assessment, 27, 54–67.
http://doi.org/10.1037/a0037917
Schwartz, M. B., Vartanian, L. R., Nosek, B. A., & Brownell, K. D. (2006). The influence of
one’s own body weight on implicit and explicit anti-fat bias. Obesity, 14, 440–447.
http://doi.org/10.1038/oby.2006.58
Sigmundová, D., El Ansari, W., Sigmund, E., & Frömel, K. (2011). Secular trends: A ten-year
comparison of the amount and type of physical activity and inactivity of random samples
of adolescents in the Czech Republic. BMC Public Health, 11, 731-743.
http://doi.org/10.1186/1471-2458-11-731
Stroebe, K., Wang, K., & Wright, S. C. (2015). Broadening perspectives on achieving social
change. Journal of Social Issues, 71, 633–645. http://doi.org/10.1111/josi.12132
Teachman, B. A., Gapinski, K. D., Brownell, K. D., Rawlins, M., & Jeyaram, S. (2003).
Demonstrations of implicit anti-fat bias: The impact of providing causal information and
evoking empathy. Health Psychology, 22, 68–78. http://doi.org/10.1037/0278-
6133.22.1.68
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 27
Thorsteinsson, E. B., Loi, N. M., & Breadsell, D. (2016). The effect of weight controllability
beliefs on prejudice and self-efficacy. PeerJ, 4, e1764.
Tylka, T. L., Annunziato, R. A., Burgard, D., Daníelsdóttir, S., … Calogero, R .M. (2014). The
weight-inclusive versus weight-normative approach to health: Evaluating the evidence
for prioritizing well-being over weight loss. Journal of Obesity, 2014, e983495.
http://doi.org/10.1155/2014/983495.
Vargas, D. de, & Luis, M. A. V. (2008). Development and validation of a scale of attitudes
towards alcohol, alcoholism and alcoholics. Revista Latino-Americana de Enfermagem,
16(5), 895–902. http://doi.org/10.1590/S0104-11692008000500016
Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attributional analysis of reactions to
stigmas. Journal of Personality and Social Psychology, 55, 738–748.
http://doi.org/10.1037/0022-3514.55.5.738
Woud, M. L., Anschutz, D. J., Van Strien, T., & Becker, E. S. (2011). Measuring thinspiration
and fear of fat indirectly. A matter of approach and avoidance. Appetite, 56, 451–455.
http://doi.org/10.1016/j.appet.2010.12.012
Wright, S. C., & Lubensky, M. E. (2009). The struggle for social equality: Collective action
versus prejudice reduction. In S. Demoulin, J.-P. Leyens, & J. F. Dovidio (Eds.),
Intergroup misunderstandings: Impact of divergent social realities (pp. 291–310). New
York, NY: Psychology Press.
Table 1
Pearson’s Bivariate Correlations of the Study Variables
Gender Age Nationality Willpower Fear of fat
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 28
Gendera
–
Age .06 –
Nationalityb
-.10*-.18*** –
Willpower -.01 -.07 .07 –
Fear of fat .34*** -.05 -.05 .22*** –
Dislike .02 -.03 .01 .35*** .20***
Cronbach’s α .64 .85
Mean (SD) whole sample 6.35 (1.63) 5.54 (2.51)
Mean (SD) control group 6.29 (1.62) 5.23 (2.51)
Mean (SD) genetic group 5.89 (1.70) 5.65 (2.51)
Mean (SD) willpower group 6.91 (1.38) 5.76 (2.51)
a males = 0, females = 1.
b Czech = 0, Slovak = 1.
* p < .05. ** p < .01. *** p < .001
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 29
Table 2
Hierarchical Linear Regressions for the Three AFA Subscales with Unstandardized Estimates
AFA Subscales
Willpower Fear of fat
Predictors ΔR2 B p 95% CI ΔR2B p 95% CI Δ
Step 1 .127***
Gendera1.79 <.001 [1.38, 2.19]
Working -0.11 .794 [-0.97, 0.74]
Stud. and w. 0.20 .229 [-0.26, 0.67]
Not w. or s. -2.56 .001 [-4.12, -1.01]
Step 2 .063*** .007*
Gender 1.80 <.001 [1.40, 2.20]
Working -0.10 .820 [-0.96, 0.76]
Stud. and w. 0.20 .409 [-0.27, 0.66]
Not w. or s. -2.58 .001 [-4.13, -1.03]
Willpower 0.63 <.001 [0.31, 0.94] 0.53 .026 [0.06, 1.00]
Genetic -0.40 .012 [-0.71, -0.09] 0.49 .038 [0.03, 0.95]
Total Adj. R2.063 .134
Note. a males = 0, females = 1
Stud. and w. = studying and working; Not w. or s. = not working or studying
* p < .05. ** p < .01. *** p < .001.
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 30
Dislike
Fear of fat
Willpower
Item 1
Item 2
Item 3
Item 4
Item 5
Item 6
Item 7
Item 8
Item 9
Item 10
Item 11
Item 12
Item 13
.58
.36
.64
.70
.71
.68
.62
.78
.83
.82
.53
.57
.72
.23***
.28***
.50
.87
.60
.67
.50
.54
.62
.39
.31
.32
.72
.67
.48
.49***
WEIGHT CONTROLLABILITY AS A PREDICTOR OF ANTI-FAT ATTITUDES 31
Figure 1. Confirmatory Factor Analysis of the 13-item Anti-fat Attitudes Scale with Standardized
Estimates.
Note. *** p <.001.
