Download Prejudice Against Fat People: Ideology and Self-Interest

Journal of Personality and Social Psychology
1994, Vol. 66. No. 5, 882-894
Copyright 1994 by the American Psychological Association, Inc.
0022-3514/94/J3.00
Prejudice Against Fat People: Ideology and Self-Interest
Christian S. Crandall
Prejudice against fat people was compared with symbolic racism. An anti-fat attitudes questionnaire
was developed and used in several studies testing the notion that antipathy toward fat people is part
of an "ideology of blame." Three commonalities between antifat attitudes and racism were explored:
(a) the association between values, beliefs, and the rejection of a stigmatized group, (b) the oldfashioned antipathy toward deviance of many sorts, and (c) the lack of self-interest in out-group
antipathy. Parallels were found on all 3 dimensions. No in-group bias was shown by fat people.
Fatism appears to behave much like symbolic racism, but with less of the negative social desirability
of racism.
One of the most interesting and controversial areas of inquiry
in attitudes and intergroup relations over the past 20 years has
been the issue of symbolic attitudes and beliefs. This work has
followed the functional approach to attitudes sponsored by
Katz (1960) and Smith, Bruner, and White (1956), and two general classes of attitude functions have been identified: instrumental and symbolic (Herek, 1986). Instrumental attitudes are
based on an evaluation of the attitude object in terms of its
utility for the person and are rational, prosaic, and relatively
immediate in nature. Attitudes with a symbolic basis are characterized by the expression of beliefs and values relevant to self
and identity. Based in early childhood learning, they represent
long-standing values about society, are affectively laden, and can
be independent of self-interest (Abelson, 1982; Herek, 1987;
Pryor, Reeder, & McManus, 1991; Sears, Lau, Tyler, & Allen,
1980).
do not emulate traditional American values, such as discipline,
self-control, and self-reliance (Kinder, 1986).
This article extends the notion of symbolic racism to antifat1
attitudes (Allison, Basile, & Yuker, 1991). I build the analogy
between racial attitudes and antifat attitudes, with the intention
of revealing the structure of antifat attitudes, as well as testing
the generality of the symbolic racism idea. Symbolic racism theory is briefly reviewed, along with some criticisms of the idea,
and some evidence of prejudice against fat people in the United
States is detailed. The studies in this article were designed to
show that the conception of symbolic belief applies to the rejection of fat people. In this research, I attempt to avoid the problems in conception, design, and analysis for which the symbolic
racism research program has been criticized.
Symbolic racism can be characterized by three hypotheses
(Kinder, 1986). The first hypothesis is that anti-Black affect
stems from the belief that Blacks do not support the traditional
values of hard work, self-contained independence, obedience to
authority, and self-discipline. The second is that this valuebased rejection occurs in addition to old-fashioned atavistic racial prejudice. The third hypothesis is that self-interest does not
form the basis of symbolic racism (Sears, Hensler, & Spear,
1979).
There are several important implications of the symbolic attitude notion. One is that abstract ideologies can play an important role in the prediction of behavior (e.g., voting). Another is
that people are motivated more by ideology and beliefs than
by self-interest. This suggestion is in contrast to realistic group
conflict theory (Sherif & Sherif, 1963), which suggests that competition over resources leads to prejudice toward out-group
members.
Symbolic Racism
Most of the research on symbolic attitudes has been on symbolic politics, particularly symbolic racism. For example, Sears,
Kinder, and their associates suggested that a substantial amount
of racism is based on symbolic beliefs about how Black people
fail to live up to classic American Protestant values (Kinder &
Sears, 1981; Sears et al., 1980). The symbolic racism perspective argues that prejudice toward Blacks is made up of two components: old-fashioned racial hatred and the belief that Blacks
Portions of these data were presented at the Fourth Nags Head Conference on Stereotypes and Intergroup Relations, Highland Beach, Florida.
I thank Colby Cohen, Lisa Hardy, Holly Harris, Rebecca Martinez,
Brooke Meyers, Jennifer Prendergast, Suzanne Shellabarger, and Stacey
Wyman for their assistance in collecting data. I especially appreciate
the contributions of Mikey Maclean Thomas in developing the AFA
questionnaire. Monica Biernat, Fletcher Blanchard, and several anonymous reviewers gave extensive and helpful comments on an earlier version of this article.
Correspondence concerning this article should be addressed to Christian S. Crandall, Department of Psychology, University of Kansas, Lawrence, Kansas 66045. Electronic mail may be sent to [email protected].
Criticism of Symbolic Racism
It is not surprising that the symbolic approach to racism has
been controversial and criticized on both conceptual and em1
The words fat and antifat were chosen because they are descriptive
and because they do not imply a medical condition (e.g., obese), nor do
they refer to some normative standard that may be genetically determined (e.g., overweight). The term fat is not used in a pejorative sense.
882
883
ANTIFAT ATTITUDES
pirical grounds. For example, Sniderman, Piazza, Tetlock, and
Kendrick (1991) have argued that although a variety of data
consistent with the symbolic racism hypothesis have been collected, there is no research suggesting that valuing independence, self-reliance, or respect for authority are essential aspects
of predicting the rejection of Blacks by Whites.
On similar empirical grounds, several authors have criticized
the program for relying on ill-defined or changing measures of
symbolic racism (Bobo, 1983; Sniderman & Tetlock, 1986). In
addition, there is criticism for measuring attitudes, racism, and
self-interest on the basis of single survey items (Bobo, 1983;
Sniderman & Tetlock, 1986).
Another area of criticism has been the conceptualization of
self-interest (see Sears & Funk, 1991). Kinder and Sears (1981)
defined self-interest in terms of having children potentially subject to busing; they assumed that it is always against parents'
self-interest to have their children bused. Crano (1992) argued
that parents need not be ineluctably opposed to busing; rather,
they may be in favor of it. Bobo (1983) argued that this definition is overly narrow. He suggested that a threat to the individual's group, or threats to the status quo that favors Whites in
general, should be interpreted as within the definition of selfinterest. This is an important argument, as it appears that
group-level discontent, rather than individual discontent, leads
to politicization and social protest (Dube & Guimond, 1986).
Symbolic Prejudice and Fat People
If antifat attitudes are structured in much the same manner
as symbolic racism, then prejudice toward fat people will be
associated with (a) beliefs and values that reflect self-determination and the Puritan work ethic and (b) measures of intolerance, such as racism and authoritarianism. In addition, (c) selfinterest should be uncorrelated with antifat attitudes in terms
of one's own fatness. To test these hypotheses, I developed a
psychometrically adequate measure of antifat attitudes and correlated it with values and beliefs relevant to the Protestant ethic
and intolerance.
In addition to addressing methodological and conceptual
criticisms, expanding symbolic attitudes to other groups has the
value of establishing the generality of the idea. Because BlackWhite relations in the United States have been a major focus of
the social sciences, political discourse, literature, education,
and popular culture for decades, there remains the possibility
that Black-White attitudes have an independent and unique attitude system, with their own structure and set of functions. By
applying the theory to other stigmatized groups, one may be
able to show that attitudes of Whites toward Blacks function
much like other prejudicial attitudes.
There is evidence of strong antifat sentiment in the United
States (e.g., AHon, 1982; Yuker & Allison, in press), and there is
reason to believe that factors related to symbolic attitudes apply
to antifat prejudice (Crandall & Biernat, 1990). Furthermore,
the lesser focus on fat people in culture, science, and politics
makes it unlikely that an independent attitude system for fat
people exists.
An added advantage of studying antifat attitudes is that they
are subject to lower levels of social desirability suppression,
when compared with racial attitudes. Sears and his colleagues
have hypothesized that symbolic racism developed from the
changing norms about the desirability of expression of antiBlack prejudice (e.g., Kinder, 1986; McConahay, 1986; Sears,
1988). Symbolic racism arose, they argued, out of the simultaneous change in public mores about racism and the individual's
early training in racial antipathy, resulting in a somewhat disguised version of racism (Gaertner & Dovidio, 1986). Studying
attitudes toward fat people, with the apparent lack of normative
pressure to suppress expression, allows us to examine the role
of social desirability suppression in the structure of prejudicial
attitudes.
Evidence of Prejudice Against Fat People
Although some reviewers have suggested that the harmful
effects of obesity stigma have not been convincingly demonstrated (Jarvie, Lahey, Graziano, & Framer, 1983), the majority
of research has suggested that being fat is associated with a wide
variety of negative characteristics. Fat people are seen as unattractive (Harris, Harris, & Bochner, 1982), aesthetically displeasing (Wooley & Wooley, 1979), morally and emotionally
impaired (Keys, 1955), alienated from their sexuality (Millman, 1980), and discontent with themselves (Maddox, Back, &
Liederman, 1968; Rodin, Silberstein, & Streigel-Moore, 1984).
Their physicians describe them as "weak-willed" (Monello &
Mayer, 1963), and their peers rate them as unlikable (Goodman,
Richardson, Dornbusch, & Hastorf, 1963).
Fat people are denigrated by thin people, health care workers,
employers, peers, potential romantic partners, their parents,
and even by themselves (Allon, 1982; Crandall & Biernat,
1990). Employers are unwilling to hire fat people, even if their
fatness would not interfere with performance (Roe & Eickwort,
1976). When employed, fat people report significant discrimination on the job (Rothblum, Brand, Miller, & Oetjen, 1990).
Fat executives are less likely to be promoted into higher paying
positions (Larkin & Pines, 1979).
Fat people are less likely to attend college (Canning & Mayer,
1966), perhaps because their parents are less willing to pay for
their expenses (Crandall, 1991). These difficulties with advancement in the job force and higher education may account for the
fact that fat people tend to be downwardly economically mobile
from their parents (Goldblatt, Moore, & Stunkard, 1965); in
the United States, fatness is associated with lower socioeconomic status (Sobal & Stunkard, 1989).
Is Fatness a Function of Willpower?
It is critical to point out at this juncture that the research does
not point to gluttony or sloth as the primary cause of fatness.
The majority of research evidence supports the notion that
body weight is the result of genetic and metabolic factors and
is only modestly related to dietary habits. For instance, in an
adoption study, Stunkard and his colleagues (Stunkard et al.,
1986) found that the body fat of children was highly correlated
with the biological parents' body fat, but uncorrelated with the
adoptive parents' body fat. To directly examine the overeating
hypothesis, Garrow (1974) reviewed 13 studies of body weight
and food intake, and in 12 of these, obese subjects ate the same
amount or less than normal weight subjects.
884
CHRISTIAN S. CRANDALL
In addition, a number of studies have shown that a variety of
physiological factors make dieting both difficult and ineffective.
For example, after dieting, energy metabolism becomes increasingly efficient (Brownell, Greenwood, Stellar, & Shrager, 1986)
and high-caloric foods become increasingly palatable (Nisbett,
1972). In short, the belief that fat people got that way primarily
from overeating and a lack of self-control does not properly represent the scientific data.
Instead, I propose the notion that holding antifat attitudes
serves a value-expressive function (Katz, 1960), reinforcing a
worldview consistent with the Protestant work ethic, self-determination, a belief in a just world, and the notion that people
get what they deserve. If ideology leads a person to chronically
attribute controllable causality to others, he or she will tend to
blame fat people for their weight and stigmatize them for it.
A similar argument has been made for racism in particular.
Many Whites hold Blacks accountable for their relatively poor
economic status (Ryan, 1971). The belief that individuals in disadvantaged groups are responsible for any negative aspects of
their situation is known as the "ultimate attribution error" (Pettigrew, 1979). Several researchers have shown relatively consistent individual differences in causal attributions (e.g., Anderson, 1983; Peterson & Seligman, 1984). These consistent attributional styles might in turn lead to the derogation of any
disadvantaged group.
Outline
This article is in three sections. In the first section (Study 1),
an anti-fat attitudes questionnaire is developed. In the second
section (Studies 2-4), attitudes and beliefs consistent with the
values of self-determination, self-control, and a tendency to reject deviance are correlated with antifat attitudes, and an experiment designed to change beliefs about control is described. In
the third section (Studies 5-6), antifat attitudes and racism are
explicitly compared in terms of social pressures about the expression of prejudice and the role of self-interest.
Study 1: Development of the Antifat Attitudes
Questionnaire
Method
Subjects were 251 undergraduates (135 women, 53.8%) who filled out
a questionnaire in a psychology class. The questionnaire included a
wide variety of items covering personal relevance, causes of fatness, willingness to interact with fat people, and so forth; 26 questions were answered on a 0-9 Likert scale. Items were selected for inclusion on the
basis of loading substantially on any factor with an eigenvalue greater
than 1.0. The text of the 13 items selected on the basis of the analysis
described below, are displayed in Table 1.
Results
A principal-components factor analysis of the items, with
varimax rotation of factors with eigenvalues, greater than 1.0,
indicated that three dimensions meaningfully described the
data. The results are displayed in Table 1.
Thefirstfactor reflects an evaluative scale and is labeled Dislike (a = .84). The second factor represents the individual's con-
cerns about weight and fat's self-relevance and is labeled Fear of
Fat (a = .79). The third factor reflects subjects' beliefs about the
controllability of weight and fat and is labeled Willpower (a =
.66). (In the studies reported below with adequate sample size,
the factor structure was replicated with only minor deviations
from Table 1.) The relevant items for each subscale were
summed and divided by the number of items to create three
scales ranging from 0 to 9. These three scales constitute the
Antifat Attitudes (AFA) questionnaire.
A series of/ tests by sex revealed that men scored higher than
women on Dislike (Ms = 2.47 and 1.85, respectively), /(243) =
3.42, p < .01, and slightly higher on Willpower (Ms = 6.12 and
5.65), t(248) = 1.79, p < .08. Women scored notably higher on
Fear of Fat (M = 6.78) than men (M = 3.55), /(249) = 9.93, p <
.001.
Dislike and Willpower were correlated (r = .43, n = 244, p <
.001); believing weight is due to willpower and denigrating fat
people go hand in hand. Fear of Fat was uncorrelated with Dislike (r = .01, ns). This suggests that self-relevant concerns about
fatness are not a major aspect of disliking fat people. Fear of Fat
was not correlated with Willpower (r = .01, n = 250, ns).
Discussion
The three scales each represent a different aspect of antifat
attitudes: prejudice toward fat people, belief in the controllability of weight, and the individual's self-relevant concerns about
fatness. The negative affect and the attribution of controllability
were substantially correlated, and neither of these was associated with self-relevant concerns (Fear of Fat).
The correlation between Willpower and Dislike is consistent
with DeJong's (1980) finding that a belief in controllability is
an important contributor to antifat attitudes. A similar finding
was reported in a recent article that came to our attention while
preparing this article. Allison et al. (1991) developed two independent scales, a measure of antifat attitudes (Attitude Toward
Obese Persons Scale) and a measure of attribution, fault, or
blame (Beliefs About Obese Persons Scale); they found a very
similar correlation between attributions of controllability and
negative attitudes toward obese people (across three samples,
average r = .42). In Study 2, an attempt is made to generalize
thisfindingto a broader set of attributional variables, connecting attitudes and beliefs about fat people to a network of related
attitudes: an ideological system.
Study 2: Antifat Attitudes in an Ideological System
If one believes that fatness is fat people's fault, then one will
denigrate and stigmatize them (DeJong, 1980; Weiner, 1986;
Weiner, Perry, & Magnusson, 1988). The argument is attributional; if ideology leads to controllable attributions to targets,
one will blame Black people for their economic situation, beautiful people for their good looks (Dion & Dion, 1987), and fat
people for their weight.
/ suggest that the classic Protestant values that are hypothesized to underlie symbolic racism can be described as a general
tendency to make controllable attributions. Attributions and
values are conceptually linked, in that both reflect characteristic ways of giving meaning to the social world. To the extent that
ANTIFAT ATTITUDES
885
Table 1
Factor Analysis ofAntifat Attitude Items
Factor
loadings
Dislike"
.792
.776
.689
.678
.678
.645
.630
Fearoffatb
.883
.824
.791
Willpower0
.824
.733
.651
Item text
I really don't like fat people much.
I don't have many friends that are fat.
I tend to think that people who are overweight are a little untrustworthy.
Although some fat people are surely smart, in general, I think they tend not to
be quite as bright as normal weight people.
I have a hard time taking fat people too seriously.
Fat people make me feel somewhat uncomfortable.
If I were an employer looking to hire, I might avoid hiring a fat person.
I feel disgusted with myself when I gain weight.
One of the worst things that could happen to me would be if I gained 25 pounds.
I worry about becoming fat.
People who weigh too much could lose at least some part of their weight through
a little exercise.
Some people are fat because they have no willpower.
Fat people tend to be fat pretty much through their own fault.
Note. All undisplayed factor loadings < .30.
* Eigenvalue = 3.6 b Eigenvalue = 2.1. c Eigenvalue =1.9.
one endorses the virtue of hard work and self-determination,
one will tend to celebrate the victories of heroes and, conversely,
blame victims for their fates. For example, the politically conservative are more likely to make person attributions in explaining poverty, whereas the politically liberal are more likely to
make situational attributions (Converse, 1964; Feather, 1985;
Zucker & Weiner, 1993). Thus, beliefs, values, and ideologies
should be closely linked with individual differences in the tendency to make internal, controllable attributions.
If antifat attitudes serve a value-expressive function, reinforcing a worldview consistent with this ideology, then they should
be correlated with a variety of attitudes that serve this same
function. For example, antifat attitudes should be correlated
with the belief in the just world (Lerner, 1980). The belief that
hard work is rewarded is consistent with classic American values, with making internal or controllable attributions, with having a positive attitude toward those who work hard, and with a
tendency to reject those who are perceived as lazy or noncontributing.
Antifat attitudes may also reflect a general orientation of intolerance and dislike of social deviance from an ideal norm of
any sort. This argument is akin to the symbolic notion that racism is based in part on old-fashioned racial prejudice (Kinder,
1986). Therefore, antifat attitudes should be associated with the
classic general individual-differences measure of old-fashioned
intolerance—authoritarianism (Altmeyer, 1984).
Finally, it is interesting to note that the stereotypes associated
with fat people and Blacks are remarkably similar—both
groups are regarded as lazy, sinful, and lacking discipline and
self-denial (Allon, 1982).2 If the endorsement of this traditional,
conservative set of beliefs underlies both racist attitudes and
antifat attitudes, then racism should be correlated with antifat
prejudice.
Method
Subjects. Five separate samples were collected. All of the samples
filled out the AFA questionnaire. In a just world sample, participants
were 43 men and 70 women from an introductory psychology class. In
an authoritarianism sample, participants were 53 men and 52 women
enrolled in an introductory psychology course. In a racism sample, participants were 543 students (270 men and 273 women) in several introductory psychology classes whofilledout a series of questionnaires for
course credit. The scales were embedded into a much longer questionnaire, which was designed for selection of subjects into experiments. In
a poverty sample, 74 undergraduates were selected from public places
around a university campus. Finally, in a values sample, subjects were
126 female, 153 male, and 10 gender-unspecified undergraduates in a
social psychology class who participated for course credit.
Measures. Subjects in the just world sample filled out the Just
World Scale (JWS; Rubin & Peplau, 1975). The JWS probes the belief
that the world is ultimately fair. The authoritarianism sample responded to a Right Wing Authoritarianism scale from Altmeyer (1984).
Christie (1991) has argued that this scale is "the best current measure"
of authoritarianism available. The racism sample answered a sevenitem Modern Racism Scale taken from McConahay (1986), a scale that
has been used in previous symbolic racism research. The values sample
responded to the Protestant Ethic Values Scale taken from Katz and
Hass(1988).
The Modern Racism Scale and the JWS items both ranged from 1 to
2
Finding a correlation between the measure of racism and antifat
attitudes does not necessarily speak directly to the issue of the similarity
between the structure of antifat attitudes and symbolic racism in the
way that attitudes toward Cracker Jack and attitudes toward baseball
may share the same structure but be uncorrelated. It is only to the extent
that they share the same underlying causal structure (e.g., similar values, beliefs in causality and justice, and authoritarianism) that we hypothesize that measures of symbolic racism and antifat attitudes should
be correlated.
886
CHRISTIAN S. CRANDALL
5, with higher scores indicating greater agreement. Each scales' items
were separately summed and divided by the number of items in the
scale, resulting in 1 to 5 scales. The authoritarianism scale was created
in the same fashion, but its items ranged from 0 to 9. Political orientation was included in both the just world and the authoritarianism samples and was measured by a two-item political ideology scale taken from
the Gallup Poll (Gallup, 1981), ranging from 0-9, with higher numbers
indicating greater conservatism.
In the poverty sample, subjects filled out a survey that included the
J WS and a single 7-point attribution item for the causes of poverty. This
poverty control item read, "To what extent do you believe that poverty
is caused by factors that poor people can control or factors outside of
poor people's control?" Higher numbers indicate belief that "poor people can control it." (Correlations between the AFA scales and the JWS
are based on the combined data of the just world and poverty samples.)
Finally, the Protestant Ethic Values Scale ranged from 0-5, with higher
scores indicating more value endorsement.
Results
Subjects averaged 2.56 (SD = 0.59) on the JWS, 2.26 (SD =
0.81) on the Modern Racism Scale, 4.56 (SD = 1.17) on the
Authoritarianism Scale, and 5.21 (SD = 2.07) on political conservatism. Subjects scored across the entire range of belief in
the controllability of poverty, averaging very close to the middle
of the scale (M = 3.80, SD = 1.68). As shown in Table 2, ideological measures were significantly correlated with antifat attitudes. All of the social ideology variables showed the same essential pattern of correlations; they were correlated with Dislike
and Willpower and were not correlated with Fear of Fat. Again,
Willpower was significantly correlated with Dislike (just world
sample, r=.52,n= 113; poverty sample, r = .60, n = 74; values
sample, r = .46, n = 289; racism sample, r = .43, n = 540;
authoritarianism sample, r = .37, n = 100; allps < .001).
Men scored slightly higher on the Modern Racism Scale (2.4
vs. 2.1), /(537) = 4.03, p < .001, the Protestant Ethic Values
Scale (3.0 vs. 2.78), r(277) = 2.98, p < .01, and the Authoritarianism Scale (4.9 vs. 4.2), /(101) = 3.44, p < .01, than women,
and there was no sex difference on the JWS, ^(112) = .22, ns.
There were no sex differences in the size of correlations between
antifat attitudes and any of the ideological variables, and partialing out the effect of gender had no substantive effect on the
correlations reported in Table 2. Authoritarianism was negatively correlated with Fear of Fat (r = -.32, n= 104, p < .01).
This correlation was based entirely on men (r = —.34, n = 53,
p < .02); among women there was no such correlation (r = —.04,
n = 51, ns). Authoritarianism appears to lead men away from
weight concerns, although the reason why is unclear.
Discussion
The pattern of correlations is consistent with the hypothesis
that rejection of fat people is based on the underlying ideological
assumption that people get what they deserve, or deserve what
they get, and that deviance from these relatively narrowly defined values should result in social rejection. Clearly, antifat attitudes are correlated with attitudes that reflect this ideology.
This ideology appears in its essential aspects to be the same kind
of values that may underlie symbolic racism. Antifat attitudes
may be a logical consequence of an ideological network or attitude system (McGuire, 1985), although the causal direction remains to be determined.
However, to some extent, the power of these data are muted
by the fact that undergraduate subjects are used—a sample that
is likely to be poorly informed about political ideology and to
have limited knowledge about racism, poverty, and whether the
world is indeed a just place (Sears, 1986). For this reason, the
"known groups" technique was used, and I chose as subjects
individuals with well-defined attitudes, a sort of "ideological
elite," who should have a meaningful and coherent attitude system, with more internal logical consistency.
Study 3: College Democrats and College Republicans
To further test the degree to which antifat attitudes are associated with a personal worldview or ideology, two student
groups with a known political ideology were given the AFA
questionnaire: the College Democrats and the College Republicans.
Hypotheses
Republicans will be more likely than Democrats to believe
that people are fat through their own fault, and as a consequence
they will be more willing to denigrate fat people. Because fear
of fat is more a private issue, depending less on a world view
than on aesthetics, there should be no difference between the
two political groups on Fear of Fat.
Table 2
Correlation Between Social Ideology and Antifat Attitudes
Scale
Conservative
politics
.47***
.48***
.09
.28***
.25***
.09
Dislike
Willpower
Fear of fat
n
Belief in
just world
217
190
Poverty
control
Protestant
ethic
29***
.32***
.04
.56***
.67***
-.03
73
288
Modern
racism
.23***
.28***
-.08
542
Authoritarianism
.29**
.16*
-.31**
100
Note. Protestant ethic = Protestant ethic values endorsement. For poverty control, high scores indicate
belief that poverty is internally controllable.
* p < . 1 0 . **/><.01. ***p<.005.
ANTIFAT ATTITUDES
Method
Subjects were 26 members of the University of Florida College Democrats and 30 members of the College Republicans (about 90% of the
active membership of each group). The groups were paid $2 per questionnaire. The questionnaires contained the three AFA scales, as well as
the political ideology scale used in Study 2.
Results
College Republicans scored higher in political conservatism
than did College Democrats (8.7 vs. 2.1), t(54) = 27.71, p <
.0001; Table 3 shows the attitudes by group. A 3 X 2 (Scale X
Group) mixed model analysis of variance (ANOVA) revealed
that the College Republicans report more antifat attitudes, F( 1,
54) = 8.15, p < 0.01. Subsequent t tests indicated that College
Republicans scored higher than College Democrats in both Dislike, f(54) = 2.21, p < .05, and Willpower, *(54) = 1.74, p < .05,
one-tailed, but not in the self-relevant Fear of Fat scale, z(54) =
1.33, p > .20. These effects did not change when either gender
or body mass index (BMI) was covaried.
As there is variability of political conservatism within group
(e.g., conservative Democrats and liberal Republicans), a more
accurate method of testing the hypothesis is to correlate the
AFA scale scores with the continuous measure of political beliefs. Disregarding group membership, the political conservatism scale was significantly correlated with both Dislike {r =
.36, p < .01) and Willpower (r = .28, p < .05) but not Fear of
Fat (r = .07, ns).
Discussion
A conservative political ideology is associated with antifat attitudes. This is made more persuasive by the fact that in these
groups political attitudes are likely to be genuine attitudes (Converse, 1963) that are centrally held and relatively important to
the respondents (Krosnick, 1988).
Because these two groups are defined by political attitudes,
it is likely that political attitudes are more central and more
fundamental than antifat attitudes in this sample. It is doubtful
that respondents joined political organizations for the opportunity to express attitudes toward fat people. Although still correlational, these data suggest that holding a conservative
worldview may lead to antifat attitudes, on the basis of the presumption that fat people are responsible for being fat, just as
they are responsible for their social position, income, material
success, and so forth (Lane, 1962).
Studies 2 and 3 indicate that antifat attitudes exist within an
Table 3
Antifat Attitudes by Political Group Membership
Antifat attitude
Dislike
Willpower
Fear of fat
Republicans
Democrats
2.88
5.82
4.20
1.71*
4.72*
4.08
Note. Overall test of group difference, F\ 1,54) = 8.15, p < .01.
* p < .05, between groups.
887
attitude system. They are consistent with studies by Crandall
and Biernat (1990) that found antifat attitudes to be associated
with a wide variety of conservative social attitudes and with the
study by Weigel and Howes (1985) that found that the derogation of out-groups tends to be an individual-differences characteristic.
The dislike of fat people was also correlated with a measure
representing a general intolerance of others (authoritarianism).
These data explicitly bring antifat attitudes into the mainstream
of research on prejudice and intolerance. Authoritarianism has
also been shown to be associated with a wide range of intolerance and low acceptance of members of any out-group in the
form of antisemitism, racism, and sexism (e.g., L. B. Brown,
1973; R. Brown, 1965). Fat people are no exception.
Not only are antifat attitudes hypothesized to be an expression of a nonspecific rejection of deviance, but also a consequence of a sociopolitical value system. McGuire (1985) argued
that the ramifications of beliefs in attitude systems tend to be
vertical in nature; for example, in his scheme, belief in the just
world would serve as a premise in a formal syllogism, leading
one to the logical conclusion that fat people had done something
to deserve their fate—presumably indolence and gluttony. A
tendency to blame the victim may play a causal role in antifat
attitudes.
Study 4: Changing Beliefs About the Causes of Obesity
To gather support for the notion that an underlying set of beliefs and values causally precede antifat attitudes, I designed an
experiment to directly manipulate these beliefs. In this study,
subjects were persuaded that fatness was not caused by a lack of
self-control, but rather was subject to uncontrollable physiological and genetic factors.
In the spirit of McGuire's (1960) work on attitude consistency, I attempted to persuade subjects about the conclusion of
a logical syllogism by attacking one of its premises. McGuire
(1960) persuaded his subjects that any recreation constituting
a serious health menace would be outlawed by the city health
authority and then indicated that local beaches were polluted.
Although they did not initially believe swimming would be outlawed by the city, and the attitude was not directly attacked,
subjects ultimately came to agree that swimming would be outlawed, indicating that they had worked out the implication of
the change in the premise for the syllogism's conclusion.
An analogous experiment was designed to persuade subjects
that weight and fatness are a function of physiology. To the extent that belief in self-control is essential to antifat attitudes,
persuading subjects against this premise should lead to the conclusion that people are not responsible for their fatness. Subjects
so persuaded should score lower on the Willpower scale, and
thus lower on the Dislike scale, but with no effect on Fear of Fat.
Method
Subjects were 11 men and 31 women enrolled in a psychology course
who received course credit for their participation.
Subjects came into the lab in groups of two tofiveand assembled in
one room. The experiment was billed as a test of memory for written
versus spoken prose. One of the subjects withdrew a slip of paper from
a container that read either "Sailing and Weight Control" or "Sailing
CHRISTIAN S. CRANDALL
and Stress." Assignment to condition was made on the basis of this
drawing. Subjects were told that there were three conditions, one in
which the messages were verbal, another in which the messages were
written, and another in which they were both verbal and written, and
that they were in this third condition. The experimenter (one of two
women) then proceeded to read a page of material describing safety
factors associated with sailing, followed by either the experimental persuasion materials or the control materials. Experimenters were kept
blind to the hypotheses.3
Subjects in the "persuade" condition (n = 24) were read a two-page
persuasive message concerning weight metabolism and genetics. These
materials stressed the genetics of weight control, describing (a) twin
studies, (b) the genetic component of fatness, and (c) the effects of dieting on metabolism. The studies were described in terms of set-point
theory, and human and animal studies were described. Subjects in the
control condition (n = 18) were read a two-page message concerning the
role of psychological stress on illness. The message was similar to the
persuasion message in its psychobiological content, length, emphasis on
human and animal studies, and complexity of materials, but did not
refer to weight.
Subjects were then given two "fact sheets," purportedly composed by
the writer of the essay, which restated the essential facts of the essay.
They received two fact sheets, corresponding to the two messages they
had just heard. Immediately after they read the fact sheets, subjects were
given a questionnaire packet containing the dependent measures. The
packet was separated into four sections, one each for the several messages the subject "might" have heard: sailing, physiology of weight control, stress and health, and baseball (a distractor—no one heard about
baseball).
Two kinds of questions were asked. Each section began with 10 truefalse factual questions. These answers served as a check that the subjects
were attending to the message. These were followed by opinion items,
answered on a 0-9 Likert scale. Typical factual questions concerning
weight were "When people (and rats) lose weight, they become irritable" (true), "It is hard to either lose or gain a large amount of weight"
(true), and "Obese people eat many more calories than normal weight
people" (false). For the weight control section, the opinion items were
from the AFA questionnaire.
Subjects always began the packet with the sailing section and thus
were used to the opinion-type question by the time they came on the
AFA questionnaire. This format allowed for some ability to disguise the
true purposes of the study.
Results and Discussion
On average, subjects in the control condition answered 7.0
true-false items correctly (with 5.0 representing chance); in the
persuade condition they correctly answered 9.5, t(40) = 8.67, p
<.001.
To test the effects of the persuasive message, a 3 X 2 (Persuasion X AFA Scale) mixed model ANOVA was calculated. The
results are displayed in Table 4; an effect of condition was found,
Study 5: Prejudice, Social Desirability, and the
Distribution of Attitudes
Table 4
Antifat Attitudes by Persuasion
Antifat attitude
Willpower
Dislike
Fear of fat
F{\, 40) = 4.41, p < .05. Subsequent / tests revealed that subjects in the persuade condition scored lower in Willpower than
the control group, suggesting that the subjects were persuaded
by the arguments, t(40) = 2.06, p < .05. Consistent with the
hypothesis, subjects in the persuade condition also showed less
Dislike, /(40) = 2.14,p < .05.
No significant difference emerged on the Fear of Fat scale,
/(40) = .46, ns, indicating that the self-relevant aspect of antifat
attitudes is not based on beliefs about self-control and willpower. Furthermore, if these results were due primarily to demand factors, it would also be reasonable to assume that they
would operate on Fear of Fat. That the Fear of Fat scale did not
show a difference between groups suggests that the differences
may have been due to the persuasive effect of the information.
This experiment cannot definitively demonstrate that beliefs
about the controllability of weight and obesity are the cause of
the correlation between the Willpower and Dislike scales. However, it does show that assumptions about discipline and selfcontrol can play a causal role in antipathy toward fat people.
Changing this belief reduces antifat attitudes.
This experiment can be contrasted with one by Wiese, Wilson, Jones, and Neises (1992). Weise et al. attempted to reduce
medical students' endorsement of the obese stereotype. They
used a broad-band manipulation based on the elaboration likelihood model (Petty & Cacioppo, 1986) that increased sympathy for obese patients, which included a videotaped interview
with an attractive and articulate obese nurse describing her
problems with dieting, two role-playing exercises, and a National Public Radio special on the causes of obesity. Wiese et al.
succeeded in reducing endorsement of the stereotype, but they
did not succeed in reducing the stigma of obesity. Although they
increased empathy with the obese and changed their subjects'
stereotypes, they did not change attributions, perhaps because
these medical students held largely accurate beliefs about the
causes of obesity before the experimental manipulations. I
would argue that because attributions did not change, Weise et
al. did not successfully reduce their subjects' social rejection of
fat people. This suggests that attributions, rather than stereotypes, lead to the rejection of fat people (see Brigham, 1973;
Stangor, Sullivan, & Ford, 1991).
Kinder and Sears (Kinder, 1986; Kinder & Sears, 1981) have
argued that an essential component of anti-Black attitudes is the
belief that Blacks have earned their fate and that their economic
and social position has resulted from controllable factors. This
same logic applies to antifat prejudice. When subjects are persuaded that fat people are not responsible for their condition,
they become more accepting of fat people.
Persuade
Control
4.65
1.75
6.19
5.57*
2.52*
6.50
Note. Test of group difference, F( 1,40) = 4.41, p< .05.
* p < .05, between groups.
I have argued that a useful analogy can be drawn between
antifat attitudes and symbolic racism. However, there is one
3
The experimenters were told about the hypotheses shortly before
the end of the study, so that in the case of seven subjects (two sessions),
the experimenters were not blind. Removing these subjects from analyses had no appreciable effect.
889
ANTIFAT ATTITUDES
critical difference between racism and weightism, that is, strong
social norms suppress the public expression of racism. It has
been found that American Whites' attitudes toward Blacks have
become more favorable over the last four decades (Case & Greeley, 1990; Firebaugh & Davis, 1988). On the other hand, these
apparently favorable trends may simply reflect an increase in
the normative constraints against the public display of racial
prejudice, thus suppressing the accuracy of self-reported racial
attitudes (Dovidio & Gaertner, 1986).
For the past 20 years or more, modern theories of interracial
relations (e.g., ambivalent racism, Katz, Wackenhut, & Hass,
1986; automatic and controlled processes in stereotyping, Devine, 1989; aversive racism, Dovidio & Gaertner, 1986; modern
racism, McConahay, 1986; symbolic racism, Kinder & Sears,
1981) have in common the notion that people's underlying
"true" racism is suppressed by a second factor, which governs
the public expression of the prejudice (Crosby, Bromley, &
Saxe, 1980; Sigall & Page, 1971). These two factors amount to
a feeling of ambivalence about Blacks, and whether positive or
negative behaviors are carried out usually depends on the social
norms about the expression of racism (Blanchard, Crandall,
Brigham, & Vaughan, in press; Blanchard, Lilly, & Vaughan,
1991).
Why suppress the expression of racist emotions? For two reasons: social norms and egalitarian values (Katz et al., 1986).
Americans by and large express strongly egalitarian and proBlack attitudes (Case & Greeley, 1990). Overtly racist behaviors
are suppressed because they are normatively inappropriate and
bring about guilt, punishment, and rejection for "incorrect" behavior. These theories all suggest that when using self-report
techniques to measure racial attitudes, social norms about public behavior, egalitarian values, and social desirability concerns
all contaminate the process. How powerful are these norms
with respect to antifat attitudes?
A closer look at the distributions of the Modern Racism Scale
and the Dislike scale from Study 2 is instructive. The average
Modern Racism Scale score was 2.26 on a scale running from 1
to 5 (SD = 0.81); the distribution was highly positively skewed
(7! = 1.92), with most subjects self-reporting low levels of racism. On the other hand, on a 0-9 scale, the Dislike scale was
symmetrically and normally distributed (M = 2.51, SD = 1.51;
7, = 0.48). Both distributions were divided into three groups:
subjects at least one standard deviation below the mean, subjects within one standard deviation of the mean, and subjects
greater than one standard deviation above the mean. Using the
normal distribution as the expected value, the Modern Racism
Scale significantly deviated from normal, x 2 (2, N = 542) =
23.26, p < .001, but the Dislike scale did not, x 2 (2, N = 542) =
0.41, ns.
The difference in distributions can be interpreted in two
ways. One possibility is that there is very little racism extant
among these subjects, causing almost all of the subjects to score
at the low end of the Modern Racism Scale. A more likely possibility is that there is a normative pressure (e.g., social desirability), which suppresses agreement with the racism items and
leads to the skewing of responses at the lower end of the scale.
To test this, I assembled several samples of undergraduates
who had filled out the Modern Racism Scale4 and the Dislike
scale, and I calculated what I call the Political Correctness Index
Table 5
PC Index of the Dislike and the Modern Racism
Scales Across Samples
PC Index
Scale
Modern Racism
Dislike
1,147
1,259
9.68
2.94
Note. PC Index represents the percentage of respondents who chose
the most politically correct response on every item of the scale, x2(l) =
47.21, p<. 001.
(PC Index). To be counted as politically correct, an undergraduate must have disagreed as strongly as possible with each item
of the scale. For the Dislike scale, subjects had to chose a 0 on
all seven items; for the Modem Racism Scale, subjects must
have chosen a 1 on all seven items. The PC Index represents the
percentage of respondents answering all items in a politically
correct (antidiscriminatory) fashion.
A total of 1,147 Modern Racism and 1,259 Dislike completed questionnaires were obtained. The results of the analysis
are displayed in Table 5. Where the measure of racism elicited
nearly 10% politically correct responses, a measure of antifat
prejudice elicited less than one third of that proportion, x2( 1, N
= 2,406) = 47.21, p < .001. There may be a normative pressure
among some of these respondents to disavow all feelings of prejudice toward Blacks, but this pressure seems to be weaker for
disavowing prejudice against fat people.5
A study by Crandall and Thompson (1993) supports this interpretation. For a study of "group decision making," Crandall
and Thompson had subjects in small groups read the folders of
four equally qualified applicants to psychology graduate school,
which included photographs that indicated the applicants were
a lean White man, a lean White woman, a lean Black man, and
a fat White man. Subjects expressed their opinions about the
applicants in turn; the last person to discuss the folders was a
female confederate who either made a race-prejudiced or fatprejudiced remark. In the control condition, she agreed with the
subjects' comments. In the race-prejudiced condition, she said
"I would definitely not pick the Black guy. I don't think Black
people make good counselors. I know I wouldn't go to a Black
guy. Personally I just don't like Black people." In the fat-prejudiced condition, she made exactly the same comment, except
the v/ordfat was substituted for Black.
Subsequently, subjects made favorability ratings of the group
members on a series of questions on a 1-7 scale. When expressing no prejudice, the confederate was rated positively (M =
4
1 thank Monica Biernat for making some of the modern racism data
available. The sample of Dislike questionnaires came from studies described in this article, from the American sample from Crandall and
Martinez (1993), and from other unpublished data sets.
5
Because the Dislike scale has 10 response categories, and the Modern Racism Scale has only 5, by chance we might expect as much as
twice as many politically correct responses on Modern Racism questionnaires. To account for this, I doubled the PC Index rate for the Dislike questionnaires, which was still highly significantly lower than the
rate for Modern Racism, x 2 (l,A r = 2,406) = 12.21, p< .0005.
890
CHRISTIAN S. CRANDALL
5.31); when expressing race prejudice, her ratings dropped substantially (M = 3.72); and when she expressed fat prejudice, her
ratings were between the control and Black prejudice conditions
(M = 4.44). All three conditions differed significantly from each
other. The the expression of prejudice resulted in social rejection, for both anti-Black and antifat prejudice. Festinger,
Schachter, and Back (1950) suggested that the more important
a norm is, the greater its effect will be on liking. It appears that
the race prejudice norm was significantly more powerful than
the fat prejudice norm; the expression of racist remarks resulted
in a significantly lower rating than the expression of antifat remarks.
In this article I simultaneously argue two things: that antifat
attitudes are similar to symbolic racism and that antifat attitudes are dissimilar to racism. There is support for both of these
lines of argument. Both symbolic racism and antifat attitudes
take their place in a wide ideological net of conservative attitudes and values, authoritarianism, and rejection of deviance.
On the other hand, the social suppression of antifat sentiment
is not as strong or well-developed as the pressure to suppress
racist attitudes. To the extent that the processes underlying racism are the same as fatism, excepting that antifat attitudes appear in a relatively pristine, unsuppressed form, racism (and the
rejection of any deviant group) might be well be understood by
comparing it with attitudes toward fat people.
Study 6: Self-interest, In-Group Bias, and Antifat
Attitudes
The "negative hypothesis" of symbolic racism is especially
interesting, that self-interest does not form the basis of symbolic
racism. In-group bias is one of the broadest possible measures
of self-interest. Because both a sense of self and self-esteem depend to a large extent on group membership (Brewer, 1979; Tajfel & Turner, 1985), it is in one's self-interest to evaluate one's
own groups positively. The sixth andfinalstudy reported in this
article is based on the secondary analysis of data from the previous studies, to ascertain whether heavier subjects reported
more accepting attitudes toward fat people. If antifat attitudes
are based, even in part, on self-interest, then fatter respondents
should have more positive attitudes toward fat people and thus
should score lower on the Dislike scale.
In all of the samples described above, subjects reported their
height and weight. To determine "fatness," the BMI was calculated as (weight[kg]/height2[m]). Kraemer, Berkowitz, and
Hammer (1990) argued that BMI is the best noninvasive measure of fatness; it is highly correlated with weight and not correlated with height.
Table 6 shows the correlation between the Dislike scale from
the AFA questionnaire and the respondent's own BMI among
seven samples. A clear pattern emerges: there is no relation between one's own weight and attitudes about other people's
weight. Taking all seven data sets together, the correlation between BMI and the Dislike scale was — .01 (n = 1384, ns), a very
modest association.6
A potential threat to the straightforward interpretation of Table 6 is the possibility that correlations are low because no "true
members" the in-group are represented in the table. Despite the
fact that many college students believe that they are fat, partic-
Table 6
Correlations Between Dislike and Body Mass
Index: Lack ofIn-Group Bias
Study sample
n
Scale development
Political groups
Just world
Authoritarianism
Protestant ethic
Persuasion experiment
Modern racism
250
56
114
101
276
44
543
r
.05
-.07
-.01
.14
.08
.06
-.05
Note. All ps ns.
ularly women (Rodin et al., 1984), fat people are underrepresented among college students, perhaps because of prejudice
(Crandall, 1991, in press; Yuker & Allison, in press). To ascertain whether the sample included subjects who might be considered fat, I compared the BMIs of my subjects with tabled
norms of height, weight, and BMI taken from a large general
sample study of weight and health, the National Health and Nutrition Examination Study (NHANES; Cronk & Roche, 1982).
Light-to-medium-weight subjects were classified as being in the
lower 75% of the tabled norms, and heavy subjects were in the
top 25% of the norms.7
Similar to previous research (Crandall, 1991), college students were lighter on average than the general population for
their age group; 45 of 676 (6.7%) women fell in the top 25% of
the tabled norms of BMI for women, and 69 of 649 (10.6%)
men were in the top 25% for men. These deviations are highly
significant for both women and men, x 2 (l, iVs = 676 and 649)
= 121.31 and 71.46, respectively, both ps < .0001, suggesting
that college students, particularly women, x 2 (l, N = 1,325) =
6.65, p < .01, are on the lighter side for their age group. However, a t test by weight class on Dislike revealed that students in
the top 25% of the population in terms of BMI were no less
antifat than those in the lower 75% (Ms = 2.25 and 2.38, respectively), /(1323) = 0.83, p > .40. Allison et al. (1991) found a
similar result, using a sample that was composed of both normal weight and obese respondents (see also Harris & Smith,
1982). There is simply no evidence to suggest that fat people
display an in-group bias when it comes to expressing prejudice
about fat people.
This lack of an in-group bias among fat people is different
from what has been found with racism—Blacks and Whites
demonstrate in-group bias. For example, in Study 2, White respondents scored a 2.59 on the Modern Racism Scale, and
Black respondents scored a 1.93, t(509) = 6.51, p < .001, y =
.28. At the same time, Table 6 is consistent with the negative
hypothesis of symbolic racism, that self-interest in relation to
the stigmatized out-group is unrelated to the symbolic attitude.
Bobo (1983) has shown that when self-interest is reconstrued
6
The correlations between BMI and Dislike computed separately by
sex were not significant (both rs = .07).
7
These results do not change if the cutoff is made at the 90-10 percentile.
ANTIFAT ATTITUDES
as the broader group interest, symbolic racists show attitudes
and voting patterns consistent with their own group's (White
Americans) interests. This is consistent with Dube and Guimond's (1986) finding that the motivation to engage in social
protest is based not on an individual's sense of deprivation, but
a feeling that one's group is suffering relative deprivation. By
contrast, the data on antifat attitudes pass a fairly stringent test;
wefindno evidence of self-interested in-group bias. This is not
to say that fat people would never express self-interested attitudes; in certain domains they may do so (e.g., in the domains
of clothing, airplane seat size, industrial design, etc.). But in
domains measured by the AFA Dislike scale, there was no evidence of a bias motivated by self-interest.
The analogy between antifat attitudes and symbolic racism
has become more complete. The research using the AFA questionnaire supports (a) the beliefs-and-values hypothesis (classic
American values are part of the ideological network that includes antifat attitudes), (b) the old-fashioned antipathy hypothesis (antifat attitudes are correlated with authoritarianism
and racism), and (c) the negative hypothesis, that self-interest
does not appear to play a primary role in antifat prejudice.
General Discussion
I have attempted to avoid many of the methodological and
conceptual criticisms leveled at symbolic racism in the past. A
psychometrically sound measure of antifat attitudes was developed and used consistently across studies. The position antifat
attitudes have in a network of beliefs and values relevant to the
dominant American social ideology have been shown explicitly.
Finally, despite being given an opportunity to manifest a correlation between antifat attitudes and self-interest, in no case did
evidence of self-interested beliefs appear. These data extend the
class of symbolic beliefs to another social group, fat people, suggesting that symbolic attitudes are an important part of prejudice in America.
There is one important area of disagreement between the
data on antifat attitudes and symbolic racism theory. Symbolic
racism is thought to be a newly emerged form of racism, distinct
from old-fashioned racism (Kinder, 1986). On the other hand,
there is no reason to suspect that the antifat attitudes described
here are new. Antifat prejudice has been demonstrated by social
scientists for decades (see Allon, 1982;Bruch, 1941;Cahnman,
1968). Instead, antifat attitudes appear to be currently at the
stage that racism was some 50 years ago: overt, expressible, and
widely held. Symbolic racism is thought to have developed because of changing norms about the expression of racial hostility.
However, because antifat attitudes appear to be constructed
much like symbolic racism, the presence of norms suppressing
racist belief and speech do not appear to be a necessary component of symbolic attitudes. Anti-Black racism may have long
been constructed in the manner of symbolic racism (see Weigel
& Howes, 1985), well before the recent movement toward widespread endorsement of egalitarian values (Firebaugh & Davis,
1988).
In fact, evidence that symbolic racism has existed for a very
long time in the American consciousness can be found in a
speech given by Abraham Lincoln in 1860 to the Cooper Institute (Lincoln, 1991). Lincoln describes the slave states' attitudes
891
toward slavery in largely symbolic terms. He reviews the historical record and concludes that beliefs and attitudes about slavery
are largely symbolic, based on values rather than experience,
and virtually unrelated to concerns about realistic threat.
Group Identification
Perhaps it is because antifat attitudes are so closely linked to
a worldview, rather than self-interest, that fat and lean people
are equally likely to be antifat. This suggests that the self-esteem
of fat people should be closely linked to their antifat attitudes
and attributional habits. For example, Crandall and Biernat
(1990) found that heavy but antifat women suffered from low
self-esteem.
In an influential essay on how the stigmatized protect their
self-esteem, Crocker and Major (1991) described several selfprotective strategies that buffer the negative feedback stigmatized people receive, for example, attributing negative feedback
to prejudice toward the group and engaging in social comparison only with in-group members. However, the strategies they
reviewed are group oriented; for them to work, a sense of identification with the group is essential. Because fat people do not
show in-group bias, it may be that there is little group feeling
among fat people. If fat people feel individuated, these self-protective strategies would be unavailable to them. Crocker, Cornwell, and Major (1993) found that when fat women attributed a
lack of interest as a dating partner to their weight, they suffered
from anxiety and depression. Several studies suggest that fat
people have difficulties with self-regard (Rodin et al., 1984;
Wadden, Foster, Brownell, & Finley, 1984).
Ideological explanations may supersede self-interest, interfering with group identification and politicization, with sometimes
harmful consequences. Another reason fat people may not show
in-group bias is that identification with other fat people does not
improve their self-image. One primary reason for identifying
with a group is that association with other members in the
group can enhance self-esteem (Tajfel & Turner, 1985), and fat
people may see few opportunities for basking in reflected glory.
In addition, fat people may see themselves as capable of leaving
the group through dieting and exercise, further inhibiting group
identification. However, fat people might benefit from self-categorization as members of a group, avoiding the self-rejection
that Franz Fanon described for Blacks and citizens of colonized
countries in the 1960s (see Fanon, 1967, 1968).
The Culture ofAntifat Attitudes
The research reported here has been focused on individual
differences in antifat attitudes. However, antifat prejudices are
common among Americans and have been demonstrated across
a wide variety of settings (Yuker & Allison, in press). This is
in part due to a deep-seated historically conservative thread in
American values. One of the most fundamental of American
myths is a conservative one—that any boy can grow up to be
president one day. Another belief is that social and economic
positions are fluid, that is, that anyone can pull himself or herself up by his or her own bootstraps. These articles of faith are
part of the core of American socialization and are inherently
individualistic and conservative.
892
CHRISTIAN S. CRANDALL
These beliefs are not uniquely American, but self-determination and internal control are signal North American values (Betancourt, Hardin, & Manzi, 1992). In countries such as Brazil,
Chile, or Mexico, where these beliefs are not so strong (Kluckhohn & Strodtbeck, 1961), antifat attitudes should be uncorrelated with belief in willpower, and ultimately both the Willpower and Dislike scales should not be highly endorsed. Crandall and Martinez (1993) have found antifat attitudes to be
lower in Mexico than in the United States, and in Mexico antipathy toward fat people was unrelated to social ideology. Perhaps
as a result, obese people in Central America are not economically disadvantaged, as they are in the United States (Sobal &
Stunkard, 1989).
This cross-cultural analysis suggests that it is the simultaneous presence of two variables that leads to antifat attitudes.
The first is a personal or cultural preference for thinness. The
second is the belief that weight is volitionally controlled. To generate dislike of fat people, one must think fat undesirable and
simultaneously blame the person for his or her situation. There
is no cause for antipathy when a person is not responsible and
no denigration when fat is not stigmatized. This argument is
true at both the cultural as well as the personal level (see Crandall & Martinez, 1993).
Fat people appear to be just one more on a long list of deviant
groups, including the elderly, homosexuals, ethnic minorities,
the handicapped, and the poor, who are stigmatized by the intolerant. Because rejection of any one of these groups tends to
be substantially correlated with any other (Altmeyer, 1984;
Weigel & Howes, 1985), further research is needed on prejudice
as an individual-differences variable. Although this approach
fell out of esteem in the 1970s (Christie, 1991), several constructs show promise, such as authoritarianism, politics, and
ideology. These data suggest that the consistent tendency to
make controllable attributions for others' misfortunes is another viable candidate.
Antifat attitudes provide a remarkable opportunity to come
to a greater understanding of the general processes of prejudice
and discrimination. They are pervasive, have an internal logic,
and result in discrimination. In contrast to racism and sexism,
the overt expression of antipathy toward fat people is currently
affected only modestly by normative pressure and concerns
about social desirability. The study of antifat attitudes, while
still in a relatively pristine form, provide an appealing counterpart to the study of race and gender in social psychology.
References
Abelson, R. P. (1982). Three modes of attitude-behavior consistency.
In M. P. Zanna, E. T. Higgins, & C. P. Herman (Eds.), Consistency
in social behavior: The Ontario symposium (Vol. 2, pp. 131-147).
Hillsdale, NJ: Erlbaum.
Altmeyer, R. (1984). Right-wing authoritarianism. Winnipeg, Canada:
University of Manitoba Press.
Allison, D. B., Basile, V. C , & Yuker, H. E. (1991). The measurement
of attitudes toward and beliefs about obese persons. Internationa!
Journal of Eating Disorders, 5, 599-607.
Allon, N. (1982). The stigma of overweight in everyday life. In B. Wolman (Ed.), Psychological aspects of obesity: A handbook (pp. 130174). New York: Van Nostrand Reinhold.
Anderson, C. A. (1983). Motivational and performance deficits in inter-
personal settings: The effects of attributional style. Journal of Personality and Social Psychology, 45, 1136-1147.
Betancourt, H., Hardin, C , & Manzi, J. (1992). Beliefs, value orientation, and culture in attribution processes in helping behavior. Journal
of Cross-Cultural Psychology, 23, 179-195.
Blanchard, E, Crandall, C. S., Brigham, J., & Vaughan, L. A. (in press).
Condemning and condoning racism. Journal of Applied Psychology.
Blanchard, F., Lilly, T, & Vaughan, L. A. (1991). Reducing the expression of racial prejudice. Psychological Science, 2, 101-105.
Bobo, L. (1983). Whites' opposition to busing: Symbolic racism or realistic group conflict? Journal of Personality and Social Psychology,
45, 1196-1210.
Brigham, J. (1973). Ethnic stereotypes and attitudes: A different mode
of analysis. Journal of Personality, 41, 206-223.
Brewer, M. (1979). Ingroup bias in the minimal group situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-324.
Brown, L. B. (1973). Ideology. Harmondsworth, England: Penguin
Books.
Brown, R. (1965). Social psychology. New York: Free Press.
Brownell, K. B., Greenwood, M. R. C , Stellar, E., & Shrager, E. E.
(1986). The effects of repeated cycles of weight loss and regain in rats.
Physiology and Behavior, 38, 459-464.
Bruch, H. (1941). Obesity in childhood and personality development.
American Journal of Orthopsychiatry, 11, 461-414.
Cahnman, W. (1968). The stigma of obesity. Sociological Quarterly, 9,
283-299.
Canning, H., & Mayer, J. (1966). Obesity—Its possible effect on college
acceptance. New England Journal of Medicine, 275, 1172-1174.
Case, C. E., & Greeley, A. W. (1990). Attitudes toward racial equality.
Humboldt Journal of Social Relations, 16,67-94.1.
Christie, R. (1991). Authoritarianism and related constructs. In J. P.
Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 501-571). San Diego,
CA: Academic Press.
Converse, P. E. (1963). Attitudes and non-attitudes. In E. R. Tufte (Ed.),
The Quantitative analysis of social problems (pp. 168-189). Reading,
MA: Addison-Wesley.
Converse, P. E. (1964). The nature of belief systems in mass publics. In
D. E. Apter (Ed.), Ideology and discontent (pp. 206-261). New York:
Free Press.
Crandall, C. S. (1991). Do heavyweight students have more difficulty
paying for college? Personality and Social Psychology Bulletin, 17,
606-611.
Crandall, C. S. (in press). Do parents discriminate against their heavyweight daughters? Personality and Social Psychology Bulletin.
Crandall, C. S., & Biernat, M. R. (1990). The ideology of anti-fat attitudes. Journal ofApplied Social Psychology, 20, 227-243.
Crandall, C. S., & Martinez, R. (1993). Culture, beliefs, and anti-fat
altitudes. Manuscript submitted for publication.
Crandall, C. S., & Thompson, E. A. (1993). Creating hostile environments: Name-calling and social norms. Manuscript submitted for
publication.
Crano, W. D. (1992, October). Vested interest: Alternative formulations.
Paper presented at the Annual Convention of the Society for Experimental Social Psychology, San Antonio, TX.
Crocker, J., Cornwell, B., & Major, B. (1993). The affective consequences of attributional ambiguity: The case of overweight women.
Journal of Personality and Social Psychology, 64, 60-70.
Crocker, J., & Major, B. (1991, June). Reactions to stigma: The moderating role of justifications. Paper presented at the Seventh Ontario
Symposium on Personality and Social Psychology, Waterloo, Ontario, Canada.
Cronk, C. E., & Roche, A. S. (1982). Race- and sex-specific reference
ANTIFAT ATTITUDES
data for triceps and subscapular, skinfolds and weight/stature2. American Journal ofClinical Nutrition, 35, 347-354.
Crosby, F., Bromley, F., & Saxe, L. (1980). Recent unobtrusive studies of
black and white discrimination and prejudice. Psychological Bulletin,
87, 546-563.
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.
Devine, P. G. (1989). Stereotypes and prejudice: Their automatic and
controlled components. Journal ofPersonality and Social Psychology,
56,5-18.
Dion, K. L., & Dion, K. K. (1987). Belief in the just world and physical
attractiveness stereotyping. Journal of Personality and Social Psychology, 52, 775-780.
Dovidio, J. F, & Gaertner, S. L. (1986). Prejudice, discrimination and
racism: Historical trends and contemporary approaches. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination and racism
(pp. 1-34). San Diego, CA: Academic Press.
Dube, L., & Guimond, S. (1986). Relative deprivation and social protest: The personal-group issue. In J. M. Olson., C. P. Herman, &
M. P. Zanna (Eds.), Relative deprivation and social comparison: The
Ontario symposium (Vol. 4, pp. 201 -216). Hillsdale, NJ: Erlbaum.
Fanon, F. (1967). Blackfaces, white masks. New York: Grove Press.
Fanon, F. (1968). The wretched ofthe earth. New York: Grove Press.
Feather, N. (1985). Attitudes, values, and attributions: Explanations of
unemployment. Journal of Personality and Social Psychology, 48,
876-889.
Festinger, L., Schachter, S., & Back., K. W. (1950). Social pressures in
informal groups. New York: Harper.
Firebaugh, G., & Davis, K. E. (1988). Trends in anti-black prejudice,
1972-1984: Region and cohort effects. American Journal of Sociology, 94.251-212.
Gaertner, S. L., & Dovidio, J. F. (1986). The aversive form of racism. In
J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination and
racism (pp. 61-90). San Diego, CA: Academic Press.
The Gallup: Political opinion 1980. Wilmington, DE: Scholary Services, Inc.
Garrow, J. (1974). Energy balance and obesity in man. New York: Elsevier.
Goldblatt, P. B., Moore, M. E., & Stunkard, A. J. (1965). Social factors
in obesity. Journal of the American Medical Association, 192, 10391044.
Goodman, N., Richardson, S. A., Dornbusch, S. M., & Hastorf, A. H.
(1963). Variant reactions to physical disabilities. American Sociological Review, 25,429-435.
Harris, M. B., Harris, R. J., & Bochner, S. (1982). Fat, four-eyed and
female: Stereotypes of obesity, glasses and gender. Journal ofApplied
Social Psychology, 6, 503-516.
Harris, M. B., & Smith, S. D. (1982). The relationships of age, sex,
ethnicity, and weight to stereotypes of obesity and self-perception.
International Journal ofObesity, 7, 361-371.
Herek, G. M. (1986). The instrumentality of attitudes: Toward a neofuctional theory. Journal ofSocial Issues, 42, 99-114.
Herek, G. M. (1987). Can functions be measured? A new perspective on
the functional approach to attitudes. Social Psychology Quarterly, 50,
285-303.
Jarvie, G. J., Lahey, B., Graziano, W., & Framer, E. (1983). Childhood
obesity: What we know and what we don't know. Developmental Review, 2, 237-273.
Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quarterly, 24, 163-204.
Katz, l.,&Hass, R. G. (1988). Racial ambivalence and American value
conflict: Correlational and priming studies of dual cognitive structures. Journal ofPersonality and Social Psychology, 55, 893-905.
893
Katz, I., Wackenhut, J., & Hass, R. G. (1986). Racial ambivalence,
value duality, and behavior. In J. F. Dovidio & S. L. Gaertner (Eds.),
Prejudice, discrimination and racism (pp. 35-60). San Diego, CA:
Academic Press.
Keys, A. (1955). Obesity and heart disease. Journal ofChronic Diseases,
1,456-460.
Kinder, D. (1986). The continuing American dilemma: White resistance to change 40 years after Myrdal. Journal of Social Issues, 42,
151-172.
Kinder, D., & Sears, D. O. (1981). Prejudice and politics: Symbolic racism versus racial threats to the good life. Journal of Personality and
Social Psychology, 40, 414-431.
Kluckhohn, F, & Strodtbeck, F. (1961). Variations in value orientation.
Evanston, IL: Row, Peterson.
Kraemer, H. C, Berkowitz, R., & Hammer, L. (1990). Methodological
difficulties in studies of obesity: I. Measurement issues. Annals ofBehavioral Medicine, 12, 112-118.
Krosnick, J. A. (1988). The role of attitude importance in social evaluation: A study of policy preferences, presidential candidate evaluations, and voting behavior. Journal ofPersonality and Social Psychology, 55, 196-210.
Lane, R. (1962). Political ideology: Why the American common man
believes what he does. New York: Macmillan.
Larkin, J. E., & Pines, H. A. (1979). No fat persons need apply. Sociology of Work and Occupations, 6, 312-327.
Lerner, M. (1980). The belief in a just world: A fundamental delusion.
New York: Plenum.
Lincoln, A. (1991). Address at Cooper Institute, New York, February
27, 1860. In Grafton, J. (Ed.), Great Speeches/Abraham Lincoln (pp.
35-51). New York: Dover.
Maddox, G. L., Back, K. W., & Liederman, V. R. (1968). Overweight as
social deviance and disability. Journal ofHealth and Social Behavior,
9, 287-298.
McConahay, J. B. (1986). Modern racism, ambivalence, and the modern racism scale. In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice,
discrimination and racism (pp. 91-126). San Diego, CA: Academic
Press.
McGuire, W. J. (1960). A syllogistic analysis of cognitive relationships.
In C. I. Hovland & M. J. Rosenberg (Eds.), Attitude organization and
change (pp. 65-111). New Haven, CT: Yale University Press.
McGuire, W. J. (1985). Attitudes and attitude change. In G. Lindzey &
E. Aronson (Eds.), The handbook of social psychology (Vol. 2, pp.
233-346). New York: Random House.
Millman, M. (1980). Such a prettyface: Beingfat in America. New York:
Norton.
Monello, L. F, & Mayer, J. (1963). Obese adolescent girls: An unrecognized "minority" group? American Journal ofClinical Nutrition, 13,
35-39.
Nisbett, R. E. (1972). Hunger, obesity, and the ventromedial hypothalamus. Psychological Review, 79, 433-453.
Peterson, C, & Seligman, M. E. P. (1984). Causal explanations as a risk
factor for depression: Theory and evidence. Psychological Review, 91,
347-374.
Pettigrew, T. (1979). The ultimate attribution error: Extending Allport's
cognitive analysis of prejudice. Personality and Social Psychology
Bulletin, 5, 461-476.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model
of persuasion. In L. Berkowitz (Ed.), Advances in experimental social
psychology (Vol. 19, pp. 124-205). San Diego, CA: Academic Press.
Pryor, J. B., Reeder, G. D., & McManus, J. A. (1991). Fear and loathing
in the workplace: Reactions to AIDS-infected co-workers. Personality
and Social Psychology Bulletin, 17, 133-139.
Rodin, J., Silberstein, L., & Streigel-Moore, R. (1984). Women and
894
CHRISTIAN S. CRANDALL
weight: A normative discontent. Nebraska Symposium on Motivation, 27, 267-307.
Roe, D. A., & Eickwort, K. R. (1976). Relationships between obesity
and associated health factors with unemployment among low income
women. Journal ofAmerican Medical Women's Association 31 193—
204.
Rothblum, E. D., Brand, P. A., Miller, C. T., & Oetjen, H. A. (1990).
The relationship between obesity, employment discrimination, and
employment-related victimization. Journal of Vocational Behavior,
37, 251-266.
Rubin, Z., & Peplau, L. A. (1975). Who believes in a just world? Journal
of Social Issues, 31, 65-89.
Ryan, W. (1971). Blaming the victim. New York: Vintage.
Sears, D. O. (1986). College sophomores in the laboratory: Influences of
a narrow data base on social psychology's view of human nature.
Journal of Personality and Social Psychology, 51, 515-530.
Sears, D. O. (1988). Symbolic racism. In P. A. Katz & D. M. Taylor
(Eds.), Eliminating racism: Profiles in controversy (pp. 53-84). New
York: Plenum.
Sears, D. Q, & Funk, C. L. (1991). The role of self-interest in social and
political attitudes. In M. P. Zanna (Ed.), Advances in experimental
social psychology (Vol. 24, pp. 1-91). San Diego, CA: Academic
Press.
Sears, D. Q, Hensler, C. P., & Spear, L. K. (1979). Whites' opposition to
"busing": Self-interest or symbolic politics? American Political Science Review, 73, 369-384.
Sears, D. O., Lau, R. R., Tyler, T. R., & Allen, H. M. (1980). Self-interest
or symbolic politics in policy attitudes and presidential voting. American Political Science Review, 74, 670-684.
Sherif, M., & Sherif, C. W. (1963). Groups in harmony and tension. New
York: Harper & Row.
Sigall, H., & Page, R. (1971). Current stereotypes: A little fading, a little
faking. Journal of Personality and Social Psychology, 18, 247-255.
Smith, M. B., Bruner, J., & White. R. W. (1956). Opinions and personality. New York: Wiley.
Sniderman, P. M , Piazza, T, Tetlock, P. E., & Kendrick, A. (1991). The
new racism. American Journal of Political Science, 35, 423-447.
Sniderman, P. M., & Tetlock, P. E. (1986). Symbolic racism: problems
of motive attribution in political analysis. Journal of Social Issues
42, 173-188.
Sobal, J., & Stunkard, A. J. (1989). Socioeconomic status and obesity:
A review of the literature. Psychological Bulletin, 105, 260-275.
Stangor, C , Sullivan, L. A., & Ford, T. E. (1991). Affective and cognitive
determinants of prejudice. Social Cognition, 9, 359-380.
Stunkard, A. J., Sorenson, T. I. A., Hanis, C , Teasdale, T. W, Chakraborty, R., Schull, W. H., & Shulsinger, F. (1986). An adoption study
of human obesity. New England Journal of Medicine, 314, 193-198.
Tajfel, H., & Turner, J. C. (1985). The social identity theory of intergroup behavior. In S. Worchel & W G. Austin (Eds.), The social
psychology ofintergroup relations. Monterey, CA: Brooks/Cole.
Wadden, T. A., Foster, G. D., Brownell, K. D., & Finley, E. (1984). Selfconcept in obese and normal-weight children. Journal of Consulting
and Clinical Psychology, 52, 1104-1105.
Weigel, R. H., & Howes, P. W. (1985). Conceptions of racial prejudice:
Symbolic racism reconsidered. Journal of Social Issues, 41, 117-138.
Weiner, B. (1986). An atlributional theory of motivation and emotion.
New York: Springer-Verlag.
Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attributional analysis of reactions to stigmas. Journal of Personality and Social Psychology, 55, 738-748.
Wiese, H. J. C , Wilson, J. F , Jones, R. A., & Neises, M. (1992). Obesity
stigma reduction in medical students. International Journal of Obesity, 16, 859-868.
Wooley, S. C , & Wooley, O. W. (1979). Obesity and women—I: A closer
look at the facts. Women's Studies International Quarterly, 2, 69-79.
Yuker, H. E., & Allison, D. (in press). Obesity: socio-cultural perspectives. In L. A. Alexander (Ed.), Eating disorders anthology. New York:
Hemisphere.
Zucker, G. S., & Weiner. B. (1993). Conservatism and perceptions of
poverty: An attributional analysis. Journal of Applied Social Psychology, 23, 925-943.
Received August 20, 1991
Revision received September 7, 1993
Accepted September 13, 1993 •