The Journal of Social Psychology, 2005, 145(3), 335–362
The Multidimensional Nature of Ageism:
Construct Validity and Group Differences
DEBORAH E. RUPP
Department of Psychology and Institute of Labor and Industrial Relations
University of Illinois at Urbana-Champaign
STEPHEN J. VODANOVICH
Department of Psychology
University of West Florida, Pensacola, FL
Department of Psychology
University of Illinois at Urbana-Champaign
ABSTRACT. The authors investigated the factor structure and construct validity of the
Fraboni Scale of Ageism (FSA; M. Fraboni, R. Saltstone, & S. Hughes, 1990) and the age
and gender differences in ageism scores. Confirmatory factor analyses supported the mul-
tidimensional nature of FSA scores and generally corroborated the initial factor structure
reported by M. Fraboni, with some notable exceptions. Essentially, the present findings
were aligned with theoretical models of ageism that emphasize both cognitive facets and
affective facets. That is, on the basis of their factor analytic findings, the authors redefined
Fraboni’s original factors of Antilocution, Avoidance, and Discrimination as Stereotypes,
Separation, and Affective Attitudes, respectively, because of the clustering of items with-
in factors. The revised 3-factor structure accounted for 36.4% of the variance in FSA
scores. FSA factor scores significantly related to other scores from other measures of age-
related attitudes, with higher correlations among factors that were similar in terms of their
cognitive nature versus their affective nature. Finally, younger individuals and men had
significantly higher ageism scores on the FSA than older individuals and women. The
authors discussed the importance of adequately assessing ageism, with particular empha-
sis devoted to the understanding of age bias.
Key words: age bias, ageism, group differences, older workers
THE TERM AGEISM was first used to describe prejudice and discrimination
directed toward older persons by Butler (1969). Ageism has been referred to as
the third great ism of our society (following racism and sexism; Butler, 1995).
Palmore (1999) explained that ageism involves both prejudice and discrimina-
tion, both stereotypes and attitudes, and therefore both cognitive and affective
processes. Research has indicated that ageism is quite prevalent in today’s soci-
ety (Palmore, 2001), possibly even more prevalent than sexism and racism (Bana-
ji, 1999), although it is typically much more difficult to detect (Levy & Banaji,
Research has shown ageism to be a strong antecedent of age bias (Rupp,
Vodanovich, & Credé, in press). Two recent meta-analytic investigations on
ageism have found that older individuals are generally perceived less favorably
than those who are younger (Gordon & Arvey, 2004; Kite & Stockdale, 2004).1
These facts, coupled with the rise of both the number of older people in society
(Williams & Nussbaum, 2001), as well as the number of age discrimination
claims being filed (McCann & Giles, 2002), suggest that more research is need-
ed that explores the construct of ageism and its measurement.
The persistence of age-related stereotypes is curious given the existence of
considerable evidence that older individuals are generally as capable as their
younger counterparts. Workplace researchers have found chronological age not
to be a valid (negative) predictor of performance for many tasks (Cleveland &
Landy, 1983; Laczko & Philipson, 1991; Liden, Stilwell, & Ferris, 1996; Seg-
rave, 2001; Wilkening, 2002). A meta-analysis by Waldman and Avolio (1986)
detected significant positive correlations between age and productivity for both
“professional” and “nonprofessional” jobs. On the other hand, negative relation-
ships (suggestive of age bias) existed for supervisory ratings of performance, par-
ticularly for those in nonprofessional positions.
Research on the construct of ageism also appears to be warranted given the
potential negative impact of ageism on both individuals and organizations. For
individuals, ageism can lead to ageist discourse, expressed ageist attitudes, and
discriminatory practices based on age (McCann & Giles, 2002), which have been
shown to cause lowered self-efficacy, decreased performance, and cardiovascu-
lar stress (Levy, Ashman, & Dror, 2000; Levy, Hausdorff, Hencke, & Wei, 2000).
For organizations, ageism can lead to costly age discrimination suits (McCann &
Giles). Between 1994 and 2000, the median award in U.S. age discrimination law-
suits was $268,926 (Employment Practice Liability, 2001), and recent settlements
(e.g., Westinghouse, Lennox, Continental Airlines, First Union) have ranged from
$6.2 million to $58.8 million (McCann & Giles). This rise in age discrimination
is evident in both U.S. organizations and those of other nations (Bennington,
2001; Chiu, Chan, Snape, & Redman, 2001; Ho, Wei, & Voon, 2000; McMullin
& Marshall, 2001; Taylor & Walker, 1997; van den Heuvel, 1999).
336 The Journal of Social Psychology
We presented portions of this paper at the 18th annual meeting of the Society for Indus-
trial and Organizational Psychology, April 2003, Orlando, FL.
We thank Silke Holub, Seth Spain, Koren Aragaki, and Demetria Gallagher for their
help with various elements of the project.
Address correspondence to Deborah E. Rupp, Department of Psychology and Insti-
tute of Labor and Industrial Relations, University of Illinois at Urbana-Champaign, 603
East Daniel Street, Champaign, IL 61820; email@example.com (e-mail).
Because the aforementioned figures are larger than those in cases involving
sex and race discrimination, there seems to be an increasing important need to
gain a better understanding of ageism and its measurement (Kite & Wagner, 2002;
Levy et al., 2000). In the words of E. S. Cohen (2001, p. 576), “ageism has moved
from the arena of morality and moral obligation to the arena of legal obligation.”
However, despite this evidence, few researchers have investigated ageism, its
measurement, its structure, and both individual and group differences in the con-
struct of ageism.
The Construct of Ageism, its Structure, and its Measurement
Early measures of age-related attitudes were developed to assess mostly uni-
dimensional constructs involving commonly held opinions about older people.
For example, Tuckman and Lorge (1953) developed the Old People Question-
naire, which assessed the extent to which individuals possess misconceptions or
stereotypes about older persons. The measure consisted of 137 items classified
into 13 evaluative categories (e.g., conservativeness). As a follow-up to this mea-
sure, Golde and Kogan (1959) developed a 20-item, qualitative sentence-com-
pletion measure for which participants formed sentences about “old people” and
“people in general.” The scale was intended to measure general attitudes about
older individuals. Kogan (1961) then created a quantitative version of the mea-
sure, termed the Attitudes Toward Old People Scale (OP), which required indi-
viduals to rate such statements.
The Facts on Aging Quiz (Palmore, 1977) comprises 25 true–false items that
measure participants’ actual level of knowledge regarding the aging process.
Although not a direct measure of ageism, this scale has been useful for research
on overall perceptions of the aged in that it measures participants’ actual level of
knowledge regarding the aging process.
The Aging Semantic Differential (ASD; Rosencranz & McNevin, 1969) has
been primarily used in gerontological research. As the first multi-dimensional
measure of age-related attitudes, this scale consists of 32 bipolar adjective pairs
on which participants rate different age groups. The scale was designed to mea-
sure attitudes about older persons’ level of competence, autonomy, and accept-
ability. However, a confirmatory factor analysis of the scale (Intrieri, von Eye, &
Kelly, 1995) revealed that a four-factor model measuring instrumentality, auton-
omy, acceptability, and integrity was superior to the original three-factor solution.
The Fraboni Scale of Ageism
Fraboni, Saltstone, and Hughes (1990) argued that the earlier ageism scales
were limited to assessing only the cognitive components of ageism (only one
aspect of ageism as defined by Butler, 1980). Therefore, the Fraboni Scale of
Ageism (FSA) was developed to measure antagonistic, discriminatory attitudes
Rupp, Vodanovich, & Credé 337
and the tendency toward avoidance, to represent a more complete measure of
ageism. In the present study, Allport’s (1958) levels of prejudice were used to
guide the writing of the items. On the basis of Allport’s levels and Butler’s defi-
nition of ageism, three factors were proposed: Antilocution (antagonism and
antipathy fuelled by misconceptions, misinformation, or myths about older per-
sons), Avoidance (withdrawal from social contact with older persons), and Dis-
crimination (discriminatory opinions regarding the political rights, segregation,
and activities of older persons).
Although the FSA has demonstrated potential as a well-balanced measure of
ageism with which investigators can study age bias, research that validates the
psychometric qualities of the FSA has been sparse. In the construction of the
FSA, Fraboni et al. (1990) used data from 100 high school students to pilot test
their measure. Subsequently, those authors conducted an exploratory principle-
components analysis on a sample of 230 Canadian college students and workers
(averaging 66% female, mean age = 31.2 years, mean education level = 14.2
years) to select their final items. Finally, 100 participants from their original sam-
ple were reused to conduct an exploratory factor analysis on their final 29 items,
which provided preliminary evidence for their proposed factor structure. Because
early results on the FSA suggested its potential as a promising and more com-
plete measure of ageism, in the present study we sought to further explore its reli-
ability and factor structure.
In addition, another objective was to investigate the construct validity of FSA
subscale scores. Following the nomological net approach to construct validity
(Cronbach & Meehl, 1955), we proposed that the cognitively focused FSA fac-
tor Discrimination would have a stronger relationship with scores on some of the
previously mentioned age-related attitude and belief measures than would the
affectively oriented FSA factors (Antilocution and Avoidance).
Hypothesis 1: FSA scores obtained from a new, non-Canadian, adult sample will
show adequate internal consistency reliability and will support the factor structure
proposed by Fraboni et al. (1990), and cognitively focused FSA subscales will be
more strongly related to past measures of age-related attitudes than will the more
affectively or behaviorally toned subscales (providing preliminary construct validity
Group Differences in Ageism
Lastly, in the present study we sought to explore group differences in ageism.
Specifically, we tested for age and gender effects in ageism scores.
Age differences. Past empirical research has found that younger people are more
ageist than older people. That is, young individuals generally possess more neg-
ative attitudes toward the older person than do their older counterparts (Bell &
Stanfield, 1973a, 1973b; Kogan, 1961; Kogan & Shelton, 1962). Organizational
338 The Journal of Social Psychology
research has found a similar age effect on attitudes about the aged (Hassell &
Perrewe, 1995) as well as an age effect regarding the severity of actions taken
against older workers for performance errors (with younger workers making
harsher recommendations; Erber, Szuchman, & Rothberg, 1990; Finkelstein,
Burke, & Raju, 1995; Rupp et al., in press). In addition, Kalavar (2001) found a
significant, negative correlation (−.19) between age scores and ageism scores. In
contrast, other researchers have indicated that older people are more biased
toward their own age group than are younger people (Hellbusch, Corbin, Thorson,
& Stacy, 1994). Still other researchers have failed to detect an age effect of any
kind (Berg & Sternberg, 1992). Finally, meta-analytic studies have indicated that
younger raters possess more ageist attitudes than do older raters (Gordon &
Arvey, 2004; Kite & Stockdale, 2004). However, the findings of Kite and
Stockdale suggest that this relationship may not be linear. That is, middle-aged
participants (on average) were found to have the highest ageism scores.
Given these contradictory findings, in the present study we sought to inves-
tigate the effect of chronological age on ageism scores using the Fraboni Scale
of Ageism, which represents the most balanced measure of ageism to date. Based
on the research on Social Identity Theory (SIT; Tajfel & Turner, 1979), we pre-
dicted that younger individuals would be more ageist than would older individu-
als. SIT indicates that individuals are motivated to perceive their own group in
more positive terms relative to out-groups. Indeed, several researchers have used
the basic tenets of SIT as a possible explanation for age effects on ageism scores
(e.g., Cuddy & Fiske, 2002; Kite & Wagner, 2002; McCann & Giles, 2002).
Hypothesis 2: A main effect for participant age will exist for ageism scores such that
younger participants will score higher in ageism than will older participants.
Gender differences. Kogan and Shelton (1962) have discussed the importance of
considering gender differences in ageism. Although their main finding was that
younger participants had more negative beliefs about older people, These
authors also reported that significant age differences were related to the gender
of the participant. That is, age differences on two items were found for male par-
ticipants only, and age differences on three other items were found for female
Some research has found women to be less ageist than men (Fraboni et al.,
1990; Kalavar, 2001). For instance, using a primarily female (66%), Caucasian
(93%), and young (M= 20.2 years, SD = 4.9 years) sample of university stu-
dents (N= 200), Kalavar found evidence that male participants (M= 70.6, SD
= 13.3) possessed greater ageism scores on the FSA than did their female coun-
terparts (M= 62.9, SD = 14.1). This finding is consistent with that of Fraboni
et al., who reported that men (M= 61.0, SD = 11.6) had significantly greater
ageism scores on the FSA than did women (M= 56.4, SD = 11.8). Finally, Kite
and Stockdale (2004) suggested that men generally give lower ratings to older
Rupp, Vodanovich, & Credé 339
individuals on performance dimensions labeled as “competence” and “behav-
ior/behavioral intentions” than do women. However, such gender differences
were not consistently found across the studies that were examined. Although the
evidence supporting a gender effect on ageism scores is somewhat inconclusive,
we predicted that women would display less systematic ageism than would men.
Such a finding would be congruent with the work of Deaux (1985) that found
women to be more warm, caring, and empathic, whereas men were more com-
petitive and critical.
Hypothesis 3: A main effect for participant gender will exist for ageism scores such
that male participants will score higher in ageism than will female participants.
Two samples (Sample A, N= 353; Sample B, N = 201) of undergraduate
students at a public university in the southeastern United States were used for
this study. Sample A was 70.5% female, and mean and median ages were 22.6
years and 20.0 years, respectively. Ages of the participants ranged from 17 years
to 58 years with 37.3% of the sample under the age of 20 years, 49.0% between
the ages of 20 years and 29 years, 7.55% between the ages of 30 years and 39
years, and 6.15% over 40 years. The racial makeup of Sample A was 18.07%
Sample B was 71.5% female, and mean and median ages were 22.15 years
and 20.0 years, respectively. Ages of the participants ranged from 17 years to 54
years, with 38.0% of the sample under 20 years, 54% of the sample between 20
years and 29 years, 4.75% of the sample between 30 years and 39 years, and
3.25 % of the sample over 40 years. Sixteen percent of Sample B belonged to a
racial or ethnic minority.
The students volunteered to participate during class time and received extra
credit for their participation. They signed a consent form assuring them that
their participation was voluntary, that they could withdraw from participation
at any time, and that their responses would be kept confidential. At the end of
the study, participants received both oral and written debriefings explaining the
purpose of the study and the broad research questions and providing them with
references to pertinent papers that they could read if interested in the topic.
They were also asked not to share any aspects of the present study or their par-
ticipation in it with anyone until the end of the semester so as not to bias future
Sample A completed the FSA (Fraboni et al., 1990), the ASD (Rosencranz
& McNevin, 1969), the OP (Kogan, 1961), and a short measure of demograph-
ics (i.e., age, gender, race). Sample B completed the FSA only.
340 The Journal of Social Psychology
The FSA (Fraboni et al., 1990) consists of 29 items designed to assess both
cognitive and affective components of ageism. Participants responded to the items
using a Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree).
Fraboni et al.’s original study investigating the psychometric properties of the FSA
consisted of a total of 231 participants. Of these, 109 were university students
(76% women, 24% men). The other participants (N= 122; 56% women, 44%
men) were participants from disparate occupations (e.g., social workers, custodi-
ans, mechanics, sales, education). As mentioned earlier, the mean age of the entire
sample was reported to be 31.2 years with an average education of 14.2 years.
Fraboni et al. (1990) found FSA scores to have adequate internal-consisten-
cy reliability with a Cronbach’s alpha coefficient of .86. The FSA’s items were
designed to measure three of Allport’s (1958) five levels of prejudice as related
to ageism: Antilocution (e.g., “Many old people just live in the past”), Avoidance
(e.g., “I don’t like it when old people try to make conversation with me”), and
Discrimination (e.g., “Most old people should not be trusted to take care of
infants”). A preliminary exploratory principle-components analysis supported
these factors, accounting for 23.3%, 7.2%, and 7.0% of the variance, respective-
ly. Cronbach’s coefficient alpha reliabilities of the Antilocution, Avoidance, and
Discrimination subscales were reported as .76, .77, and .65, respectively. In the
present study, the alpha reliability estimates of these subscales were .75, .61, and
.77, respectively. Fraboni et al. also presented evidence regarding the construct
validity of the FSA. That is, they found scores on the FSA to possess significant,
negative correlations with the Facts on Aging Quiz (−.28) and the Acceptance of
Others Scale (−.22).
The ASD (Rosencranz & McNevin, 1969) was constructed to measure the
valences of stereotypic attitudes about age. The ASD consists of 32 bipolar adjec-
tive scales on which participants judge an indicated social object on one of seven
response levels, with lower scores indicating a more positive attitude toward the
object. In the original construction of the ASD, the social objects were persons
of all ages, a man between the ages of 20 years and 30 years, a man between the
ages of 40 years and 55 years, and a man between the ages of 70 years and 85
years. An initial factor analysis of the ASD indicated three major dimensions:
Instrumental–Ineffective, Autonomous–Dependent, and Personal Acceptability–
Subsequent investigators used a generalized social object, such as a nonspe-
cific old person (Underwood, Eklund, & Whisler, 1985). Further examination of
the ASD factor structure by Holtzman, Beck, and Kerber (1979) revealed four
dimensions: Instrumentality,Autonomy, Acceptability, and Integrity. A subsequent
confirmatory factor analysis by Intrieri et al. (1995) found the modified four-fac-
tor model as having the best fit. Moderate-to-high intercorrelations existed among
the factors (ranging from .56 to .96), and the internal consistency reliability of the
Rupp, Vodanovich, & Credé 341
subscales varied from .75 to .85. In the present study, we used the four-factor solu-
tion in subsequent statistical analyses. We found the internal consistency (alpha)
of Instrumentality, Autonomy, Acceptability, and Integrity scores to be .78, .80,
.78, and .80, respectively.
Kogan’s (1961) OP measures individuals’attitudes toward older people. The
scale originally consisted of 34 “old people” items in the form of positive–nega-
tive pairs. That is, 17 items expressed negative statements about older people
(OP−; e.g., “most old people get set in their ways and are unable to change”), and
17 items expressed analogous statements written in the positive direction (OP+;
e.g., “most old people are capable of new adjustments when the situation demands
it”). Participants responded to items using a Likert-type response scale that
ranged from 1 (strongly disagree) to 6 (strongly agree).
Kogan (1961) reported internal consistency reliability coefficients (alphas)
ranging from .66 to .85 for the 34-item scale across three samples. Also, Pearson
product-moment coefficients between positively and negatively worded scales
were found to be significant, ranging from .46 to .52. Because Kogan found the
OP−scale to possess greater reliabilities than the OP+ scale, we used the OP−
scale in the present study, and its alpha reliability estimate was .86.
FSA Factor Structure and Construct Validity
We tested the three-factor model suggested by Fraboni et al. (1990) using a
confirmatory factor analysis with polychoric correlations. The model consisted
of three intercorrelated first-order factors with each item loading onto one of the
three latent factors. We did not allow error variances to correlate, and, for each
latent factor, we constrained the path from one item to a given factor to 1.00 for
purposes of statistical identification (Byrne, 1998). Item loadings are presented
in Table 1. Results indicated that this model did not provide a very good fit to the
data, χ2(374) = 877.75, RMSEA = .062, NNFI = .80, CFI = .81, GFI = .85, AGFI
= .83. Therefore, we conducted a follow-up exploratory factor analysis with a
varimax rotation on Sample B to revisit the factor structure of the measure. Items
were included within a factor if their loadings were .40 or greater. A three-factor
solution emerged, which accounted for 36.4% of the variance in FSA scores.
Table 2 shows the factor loadings for each item as well as its similarity or dis-
crepancy from the Fraboni et al. three-factor solution.
The results from the preceding analysis revealed a factor structure that was
somewhat different from the one suggested by Fraboni et al. (1990). However, our
factor structure appears to be more consistent with the original purpose of the FSA
(i.e., to measure both the affective components of ageism and the cognitive ones)
than Fraboni et al.’s original factor structure. Factor 1 (α= .79) consisted of 10
items that describe beliefs about older persons as a group; see Table 2. Factor 1
342 The Journal of Social Psychology
Rupp, Vodanovich, & Credé 343
TABLE 1. Item Loadings for Initial Confirmatory Factor Analysis (N= 353)
Item Antilocution Avoidance Discrimination
1. Many old people are stingy
and hoard their money and
3. Many old people just live in
the past. .83
4. Most old people should not
be trusted to take care of
5. Many old people are happiest
when they are with people
their own age. .79
9. I would prefer not to go to an
open house at a senior’s club,
if invited. .56
16. Old people should feel
welcome at the social
gatherings of young people. .52
25. Old people deserve the same
rights and freedoms as do
other members of our society. .88
27. Old people can be very
28. I would prefer not to live with
an old person. .85
29. Old people do not need much
money to meet their needs. .58
2. Many old people are not
interested in making new
friends, preferring instead the
circle of friends they have
had for years. –.15
8. Old people complain more
than other people do. .46
17. Old people don’t really need
to use our community sports
18. It is best that old people live
where they won’t bother
20. It is sad to hear about the
plight of the old in our society
these days. .96
344 The Journal of Social Psychology
TABLE 1. Continued
Item Antilocution Avoidance Discrimination
21. Old people should be
encouraged to speak out
22. Most old people are interest-
ing, individualistic people. .46
23. I personally would not want
to spend much time with an
old person. .71
24. There should be special clubs
set aside within sports
facilities so that old people
can compete at their own level. .80
6. Most old people would be
considered to have poor
personal hygiene. .91
7. Most old people can be
irritating because they tell the
same stories over and over
10. Teenage suicide is more
tragic than suicide among the
11. I sometimes avoid eye contact
with old people when I see
12. I don’t like it when old people
try to make conversation with
13. Complex and interesting
conversation cannot be ex-
pected from most old people. .96
14. Feeling depressed when
around old people is probably
a common feeling. .78
15. Old people should find
friends their own age. 1.00a
19. The company of most old
people is quite enjoyable. .55
26. Most old people should not
be allowed to renew their
drivers licenses. .79
aPaths were constrained to be equal to 1.0.
Rupp, Vodanovich, & Credé 345
TABLE 2. Varimax Factor Loadings for Ageism Dimensions (N= 201)
Item 1 2 3
Factor 1: Stereotypes
1. Many old people are stingy and
hoard their money and possessions. .57 .00 .24 Antilocution
2. Many old people are not interested
in making new friends, preferring
instead the circle of friends they
have had for years. .64 .00 .00 Antilocution
3. Many old people just live in the
past. .71 .00 .00 Antilocution
4. Most old people should not be
trusted to take care of infants. .42 .28 .00 Antilocution
5. Many old people are happiest
when they are with people their
own age. .45 .20 .00 Antilocution
6. Most old people would be
considered to have poor personal
hygiene. .55 .18 .26 Antilocution
7. Most old people can be irritating
because they tell the same stories
over and over again. .64 .21 .26 Antilocution
8. Old people complain more than
other people do. .56 .18 .25 Antilocution
9. I would prefer not to go to an open
house at a senior’s club, if invited. .48 .29 .35 Avoidance
10. Teenage suicide is more tragic than
suicide among the old. .40 .00 .21 Antilocution
Factor 2: Separation
11. I sometimes avoid eye contact with
old people when I see them. .25 .70 .00 Avoidance
12. I don’t like it when old people try
to make conversation with me. .22 .79 .10 Avoidance
13. Complex and interesting
conversation cannot be expected
from most old people. .18 .50 .15 Antilocution
14. Feeling depressed when around
old people is probably a common
feeling. .32 .42 .00 Avoidance
Fraboni et al.
346 The Journal of Social Psychology
TABLE 2. Continued
Item 1 2 3
Factor 2: Separation (continued)
15. Old people should find friends
their own age. .00 .57 .00 Avoidance
16. Old people should feel welcome
at the social gatherings of young
people. .00 .46 .26 Avoidance
17. Old people don’t really need to
use our community sports facilities. .28 .42 .17 Discrimination
18. It is best that old people live
where they won’t bother anyone. .41 .51 .16 Discrimination
Factor 3: Affective attitude
19. The company of most old people
is quite enjoyable. .25 .21 .62 Discrimination
20. It is sad to hear about the plight
of the old in our society these
days. .00 .00 .66 Discrimination
21. Old people should be encouraged
to speak out politically. .14 .13 .59 Discrimination
22. Most old people are interesting,
individualistic people. .13 .20 .65 Discrimination
23. I personally would not want to
spend much time with an old
person. .32 .39 .47 Avoidance
Original item excluded from revised measure
24. There should be special clubs set
aside within sports facilities so
that old people can compete at
their own level. .16 –.27 –.10 Discrimination
25. Old people deserve the same rights
and freedoms as do other members
of our society. .00 .33 .27 Discrimination
26. Most old people should not be
allowed to renew their drivers
licenses. .38 .15 .00 Antilocution
Fraboni et al.
appears to measure the cognitive component of ageism identified in past research
(e.g., Kogan, 1961; Tuckman & Lorge, 1953). Consequently, we relabeled Factor
1, which was similar to Fraboni et al.’s Antilocution factor, Stereotypes.
Factor 2 and Factor 3 appeared to measure ageism’s affective component (see
Table 2). Factor 2, which is comparable to Fraboni et al.’s (1990) Avoidance factor,
consisted of 10 items (α= .76) that primarily assessed the desire of individuals to
separate themselves from older people. Consequently, this factor was relabeled Sep-
aration. The items that loaded on Factor 3 (k = 5, α= .70) were mostly reflective
of emotionally related attitudes toward older people (like Fraboni et al.’s Discrim-
ination factor). Therefore, we relabeled Factor 3 Affective Attitudes. Six items
included in the original FSA did not load highly on any of the three factors and
were therefore excluded from the measure in subsequent analyses (see Table 2).
To further validate this revised three-factor model, we conducted a second
confirmatory factor analysis on the original sample of 353 participants using
the revised factor structure. Results showed a significant improvement in fit
over the original structure, χ2(227) = 479.41, RMSEA = .056, NNFI = .87, CFI
= .88, GFI = .89, AGFI = .87 (see Table 3 for item loadings) and relatively low
correlations among factors; see Table 4. This revised three-factor model was
compared to (a) a single-factor model, χ2(230) = 677.58, RMSEA = .079, NNFI
= .76; CFI = .78, GFI = .85, AGFI = .82, (b) a three-factor model with uncor-
related latent factors, χ2(230) = 772.21, RMSEA = .082, NNFI = .73; CFI = .75,
GFI = .84, AGFI = .87, and (c) three two-factor models, each of which con-
tained one of the original latent factors as the first factor, with the combination
Rupp, Vodanovich, & Credé 347
TABLE 2. Continued
Item 1 2 3
Original items excluded from revised measure
27. Old people can be very creative. .33 .23 .38 Avoidance
28. I would prefer not to live with an
old person. .39 .18 .39 Avoidance
29. Old people do not need much
money to meet their needs. .30 .34 .12 Antilocution
Note. Bold entries signify the strongest factor loadings.
Fraboni et al.
348 The Journal of Social Psychology
TABLE 3. Item Loadings for Confirmatory Factor Analysis Using Revised
Factor Structure (N= 353)
Item from Fabroni Scale of Ageism Stereotypes Separation attitudes
Teenage suicide is more tragic than
suicide among the old. .52
Many old people are stingy and hoard
their money and possessions. .80
Many old people are not interested in
making new friends, preferring
instead the circle of friends they have
had for years. .61
Many old people just live in the past. .80
I would prefer not to go to an open
house at a senior’s club, if invited. .85
Most old people should not be trusted to
take care of infants. .59
Many old people are happiest when
they are with people their own age. .57
Most old people would be considered to
have poor personal hygiene. .88
Most old people can be irritating
because they tell the same stories
over and over again. 1.00a
Old people complain more than other
people do. .84
I sometimes avoid eye contact with old
people when I see them. .89
I don’t like it when old people try to
make conversation with me. 1.00a
Complex and interesting conversation
cannot be expected from most old
Feeling depressed when around old
people is probably a common feeling. .52
Old people should find friends their
own age. .54
Old people should feel welcome at the
social gatherings of young people. .33
Old people don’t really need to use our
community sports facilities. .54
It is best that old people live where they
won’t bother anyone. .72
of the two remaining latent factors as the second factor; see Table 5 for fit index-
es. Chi-square difference tests indicated that the revised three-factor structure,
with correlated factors, was the solution demonstrating the best fit. A post hoc
examination of modification indexes did not suggest further meaningful modi-
fications to the structure of the FSA. Given this evidence in support of the mul-
tidimensionality of ageism, we used the revised FSA factor scores (as opposed
to total scores) in all subsequent analyses. All remaining analyses were com-
puted using this revised factor structure on the first group of 353 participants
Table 6 provides the summary statistics, coefficient alphas, and intercorrela-
tions for the measures used in the present study. As the table indicates, significant
correlations existed between the FSA factors and other measures of age-related atti-
tudes. As expected, FSA Stereotypes, measuring ageism’s cognitive component,
was most strongly related to the more cognitively focused ASD and OP subscales.
Rupp, Vodanovich, & Credé 349
TABLE 3. Continued
Item from Fabroni Scale of Ageism Stereotypes Separation attitudes
I personally would not want to spend
much time with an old person. 1.00
The company of most old people is
quite enjoyable. 1.00a
It is sad to hear about the plight of the
old in our society these days. .53
Old people should be encouraged to
speak out politically. .77
Most old people are interesting,
individualistic people. .90
aPaths were constrained to be equal to 1.0.
TABLE 4. Factor Correlations of the Subscales of the
Revised Fabroni Scale of Ageism (N= 353)
Variable 1 2
2. Separation .65
3. Affective attitudes .73 .66
350 The Journal of Social Psychology
TABLE 5. Fit Statistics for Confirmatory Factor Analysis Model Comparison
Variable χ2df RMSEA NNFI CFI GFI AGFI ∆χ2∆df
Fraboni three factor 877.75 374 0.062 0.80 0.81 0.85 0.83
Revised three factor 479.41 227 0.056 0.87 0.88 0.89 0.87
Single factor 734.35 230 0.079 0.76 0.78 0.85 0.82 254.94 3
(uncorrelated factors) 772.21 230 0.082 0.73 0.75 0.84 0.81 292.8 3
Two factor (stereotype
& separation combined) 663.97 229 0.073 0.79 0.81 0.86 0.83 184.56 2
Two factor (stereotype &
affective combined) 541.41 229 0.067 0.83 0.85 0.87 0.85 62 2
Two factor (separation &
affective combined) 593.16 229 0.067 0.82 0.83 0.87 0.85 113.75 2
Note. ∆χ2refers to the difference in chi square between the revised three-factor structure and the proposed alternative model. AGFI = adjusted good-
ness of fit index; CFI = comparative fit index; GFI = goodness of fit index; NNFI = nonnormed fit index; RMSEA = root mean squared error of
Rupp, Vodanovich, & Credé 351
TABLE 6. Intercorrelations Among Three Ageism Scales and Subscales
Subscale Items MSDCronbach’s α12 345 67
Fabroni Scale of Ageism
1. Stereotypes 10 2.04 0.436 .79
2. Separation 8 1.60 0.427 .76 .544
3. Affective attitudes 5 1.74 0.465 .70 .503 .463
Aging Semantic Differential Scale
4. Instrumental 6 3.79 0.919 .77 .507 .340 .411
5. Autonomy 8 3.36 0.906 .79 .317 .298 .305 .650
6. Acceptability 7 2.99 0.913 .78 .347 .307 .322 .596 .630
7. Integrity 5 3.28 1.045 .79 .332 .267 .197 .604 .675 .708
Attitudes Toward Old People
8. Negative attitudes
toward old people 14 2.28 0.679 .86 .733 .520 .508 .508 .385 .355 .322
Note. All correlations are significant at the level of p< .01. N= 353.
Likewise, the relationship between FSA Affective Attitudes and these measures was
relatively weak. Taken together, these analyses support Hypothesis 1.
Group Differences in Ageism
Prior to testing the effects of age and gender on ageism, we examined how
both ageism and age were distributed in our sample. In terms of ageism, none of
the scales exhibited meaningful skewness or kurtosis (max skewness = .686, max
kurtosis = .501), although low ageist attitudes are probably considered to be desir-
able. The distribution of age, however, exhibited both high skewness (2.556) and
high kurtosis (7.433). This is not surprising given the nature of our sample (col-
lege students). Although the presence of high skewness and kurtosis can attenu-
ate observed effects in a downward direction, we decided not to transform the age
variable. This is because transformed variables are often difficult to interpret, par-
ticularly when the original variable is easily interpretable, as was the case here.
Table 7 shows the results of the analyses of age and gender effects in ageism
scores. Ageism scores exhibited significant negative relationships with age for
five of the examined subscales, supporting Hypothesis 2, although the strength of
these effects was generally weak. Likewise, women, on average, exhibited lower
levels of ageism (Wilks’s Λ= .900, p< .001) on all eight of the examined sub-
scales; six of these differences were statistically significant and had moderate
effect sizes (J. Cohen, 1988), supporting Hypothesis 3.
In addition to the moderate linear relationship between age and ageism shown
by Table 7, we also found significant curvilinear relationships between participant
age and ageism scores. Table 8 illustrates that these curvilinear effects added sig-
nificant incremental validity over a simple linear effect for age for six of the eight
measures of age attitude. The direction of the curvilinear effects suggests an
asymptotic effect. That is, the decline in ageism scores that is associated with being
older becomes less pronounced at the higher end of the age spectrum.
Summary and Limitations
In summary, a series of confirmatory factor analyses provide support for the
existence of three FSA factors that are similar to the ones initially proposed by
Fraboni et al. (1990). However, given the present findings, we suggest that the fac-
tors are best labeled as stereotypes,separation, and affective attitudes. This struc-
ture appears to be a more accurate reflection of the factorial composition of the scale
and offers an improved characterization of the cognitive and affective facets of
ageism as defined by Butler (1969). This finding is bolstered by the specific rela-
tionships that we detected between FSA factors and scores on additional ageism
scales. That is, the cognitively oriented FSA factor of Stereotypes was significantly
352 The Journal of Social Psychology
Rupp, Vodanovich, & Credé 353
TABLE 7. Effect for Gender and Age on the Subscale Scores of the Fabroni Scale of Ageism, the Aging Semantic Differential
Scale, and the Attitudes Toward Old People Scale
Subscale M SD M SD t(350) pd p
Fabroni Scale of Ageism
1. Stereotypes 1.98 0.43 2.19 0.41 4.32 .000 –.50 –0.14 0.01
2. Separation 1.55 0.39 1.72 0.49 3.28 .001 –.38 –0.15 0.005
3. Affective attitudes 1.67 0.44 1.93 0.47 4.88 .000 –.59 –0.17 0.001
Aging Semantic Differential Scale
4. Instrumental 3.71 0.92 3.98 0.91 2.51 .013 –.28 –0.11 0.039
5. Autonomy 3.34 0.91 3.43 0.91 0.82 .412 –.10 –0.04 0.451
6. Acceptability 2.91 0.91 3.17 0.91 2.44 .015 –.29 0.03 0.604
7. Integrity 3.24 1.08 3.38 0.96 1.09 .275 –.14 0.06 0.257
Attitudes Toward Old People
8. Negative attitudes
toward old people 2.23 0.67 2.39 0.68 1.99 .048 –.24 –0.26 0.000
Note. N= 353.
354 The Journal of Social Psychology
TABLE 8. Curvilinear Relationships Between Participant Age and Scores on Measures of Ageism
Fabroni Scale of Ageism Aging Semantic Differential Scale
Variable Stereotypes Separation attitudes Instrumental Autonomy Acceptability Integrity
Age –1.01** –1.36** –.25 –1.08** –.79* –.38 –.65* –1.00**
Age20.881** 1.23** .08 0.98** .76* .42 .72* 0.76*
Initial R20.019** 0.022** .029** 0.012* .002 .001 .004 0.066**
∆R20.023** 0.045** .00 0.028** .017* .005 .015* 0.017*
Total R20.042** 0.067** .029** 0.041** .019** .006 .019* 0.083**
*p< .05. **p< .01.
correlated with other cognitive ageism measures. FSA factors that were more affec-
tive in nature (Separation and Affective Attitudes) were also significantly related to
other ageism measures, although to a lesser extent than the Stereotypes factor.
However, several limitations should be noted. First, the factor structure
reported here only accounted for approximately 36% of the variance in FSA
scores and that fit indices showed only moderate-to-good fit. Consequently, addi-
tional research on the FSA appears warranted. Specifically, it may be useful to
develop additional items for the factor Affective Responses. This factor possessed
the fewest number of items (k= 5) and the lowest internal consistency (α= .70).
This is especially important in that the Affective Responses factor adds breadth
to the cognitive nature of previous ageism measures.
A second limitation is the limited number and type of variables that have
been used to provide construct validity for the FSA. To completely understand
ageism’s placement in a larger nomological network (Cronbach & Meehl, 1955),
it would be beneficial for FSA scores to be compared with scores on many other
similar and dissimilar measures. Such useful constructs could include personali-
ty characteristics and general prejudice, stereotypes, and attitudes. Indeed, it
would be useful to know if a discriminatory profile exists or whether different
types of individuals hold negative attitudes toward different groups.
Third, our findings indicate that men are more ageist than women. This find-
ing supports previous research on gender differences in ageism (e.g., Fraboni et
al., 1990; Kalavar, 2001; Kite & Stockdale, 2004). This result may be partly due
to females’ having higher scores on the personality dimension of expressiveness
(e.g., warmth, caring, empathy), whereas men generally possess higher instru-
mentality scores (Deaux, 1985). Kalavar argued that such an effect may be the
result of lifespan developmental processes and greater experience with and expo-
sure to older people. Nevertheless, investigators should interpret the gender effect
found herein with caution. That is, the difference in ageism scores attributed to
gender was relatively small (mean difference of .17) and could have been affect-
ed by a variety of factors, most notably the size of the current sample. A gender
difference of this size, albeit statistically significant, casts some doubt on the
pragmatic utility of the present finding. It is important to note, however, that our
effect sizes were moderate in strength and that they were strongest for the FSA
subscales. Further research is needed to determine the accuracy and generaliz-
ability of the gender–ageism relationship.
Last, the present results indicate that a significant negative relationship exists
between participants’ chronological age and ageism scores, indicating a tenden-
cy for younger raters to be more ageist than older raters. This finding supports
past research that has found similar effects (e.g., Finkelstein et al., 1995) and
counters past research that has not detected such an effect (e.g., Hellbusch et al.,
1994). In the present study, we also found evidence supporting the possibility that
a curvilinear relationship may exist between chronological age and ageism, which
is congruent with the research of Kite and Stockdale (2004). However, the skew-
Rupp, Vodanovich, & Credé 355
ness and kurtosis of the participants’ age was high. A log transformation could
have been used to increase the heteroscedasticity of the age variable, but results
that are based on such transformations are often difficult to interpret. Also, given
that the untransformed variable is a “real-life” variable, we believe that the atten-
uation of our effects that was due to skewness and kurtosis is a likely artifact of
using a college-age population. Future research should certainly attempt to repli-
cate our findings on a less skewed sample of adults.
Perhaps the general finding that greater ageism is found among younger
raters is a manifestation of out-group derogation (see Fiske, 2002, and the work
on SIT; Tajfel & Turner, 1979). Research in this area has generally found that
individuals are biased in their evaluations, attributions, and expectations of those
considered to be out-group members. Consistent with social comparison theory
(Festinger, 1954; Wood, 1989), it is also likely that young people seldom inter-
act with the aged and are primarily exposed to people of similar ages. This cir-
cumstance can lead to a confined view of older individuals, especially because of
their relatively rare—and often low-status—portrayal in books and the media
(e.g., Whitbourne & Hulicka, 1990).
These findings are interesting in comparison to the research on general
stereotyping and prejudice. Although research has shown that older persons (as
opposed to younger persons) are more prejudiced in general (von Hippel, Silver,
& Lynch, 2000), the present results suggest that this may not be the case when
the focus of the prejudice is one’s own group (i.e., older people). When age is the
target of prejudice, an opposite effect occurs. That is, younger individuals show
more ageism than do older individuals. This finding is consistent with our SIT
argument and is also certainly an avenue for future research.
A deeper understanding of ageism’s correlates and measurement would be
important to research on age bias—especially workplace age bias. Previous
research indicates that older workers receive harsher evaluations, have their errors
attributed to their age as opposed to situational factors, and generally fall victim
to the belief that job performance declines with age (Faley, Kleiman, & Leng-
nick-Hall, 1984; Finkelstein et al., 1995; Issacharoff & Harris, 1997; Perry, Kulik,
& Bourhis, 1996; Rupp et al., in press; Wilkening, 2002). It is important to note
that the belief that job performance declines with age exists despite the fact that
no differences have been found between older and younger workers when “objec-
tive” or productivity indices are used (Cleveland & Landy, 1983; Forteza & Pri-
eto, 1994; Laczko & Philipson, 1991; Landy, 1992; Mowery & Kamlet, 1996;
Park, 1994; Waldman & Avolio, 1986; Warr, 1994). Supporting the significance
of work-related ageism have been findings of recent meta-analytic studies indi-
cating that the impact of negative information on ageism is heightened within
employment contexts (Kite & Stockdale, 2004) and that both job applicants and
incumbents can be victims of ageism (Gordon & Arvey, 2004).
356 The Journal of Social Psychology
Another reason the study of ageism is warranted is that recent research has
found that ageism can be reduced (Braithwaite, 2002; Chiu et al., 2001; Ragan &
Bowen, 2001). Efforts have been aimed at both potentially ageist individuals (e.g.,
those who have the power to engage in age discrimination, such as raters in per-
formance evaluations or personnel-selection situations) and potential victims of
age bias (e.g., by providing opportunities for development in areas in which per-
formance seems to decline with age). Other interventions have focused on orga-
nization or community climate. Such a strategy attempts to sensitize individuals
to the possibility of unconsciously engaging in stereotyping (Braithwaite, 2002),
to educate individuals on the myths and realities of aging (Finkelstein et al.,
1995), and to emphasize the negative consequences of age bias. Examples of such
strategic actions include eliminating age types for jobs and holding organizational
decision makers accountable for age discrimination (Maurer & Rafuse, 2001;
Perry et al., 1996). This strategy reflects the more general literature on bias, which
has shown that bias can be reduced through education and constructive intergroup
contact, including contact of equal status, shared goals, cooperation in pursuit of
goals, and the support by authorities (Fiske, 2002).
Additional Future Research
Because of the importance of minimizing workplace age bias, one potentially
fruitful area of future research would be that of developing an organizationally
focused version of the FSA. This development would involve shifting the items’
focus from “the elderly” to “older workers.” Also, item content could be added to
refer explicitly to workplace situations. If such a measure was found to be reliable
and valid, research on workplace age bias could explore the impact of age-related
cognitive or affective beliefs and attitudes on discriminatory behaviors more direct-
ly. For instance, research has indicated that older employees and job applicants are
viewed more negatively (e.g., Finkelstein et al., 1995; Gordon & Arvey, 2004; Kite
& Johnson, 1988), are perceived as possessing less interpersonal skills, stamina,
competence, and dexterity (Finkelstein & Burke, 1998; Kite & Stockdale, 2004;
McMullin & Marshall, 2001) and as being less affected by training efforts (Cleve-
land & Shore, 1992; Maurer & Rafuse, 2001). Such views are most prominent for
jobs involving extensive physical requirements (Young, Rinhart, & Baits, 1997).
Also, recent research has suggested that older workers receive harsher sanctions for
job-related errors and that these may be the result of (a) ageist attitudes and (b) (sta-
ble) attributions of poor performance (Rupp et al., in press).
Future research might also consider integrating the findings of the present
study with that of research looking at individual differences in age perceptions held
about particular jobs. For example, Gordon and Arvey (1986) found that individ-
uals classify many jobs in terms of the average age of persons holding the posi-
tion. Further, these authors also found that this average-age perception was lower
for younger participants. In predicting outcomes such as performance ratings,
Rupp, Vodanovich, & Credé 357
future studies might consider exploring the interplay between supervisor age and
ageism, subordinate age, and the perceived age of the job as seen by subordinates.
It is important to note, however, that social and industrial/organizational psy-
chologists may assign very different meanings to the term “old” (Kite & Wagner,
2002). Whereas the ageism research conducted in the field of gerontology has
focused primarily on “the elderly” (i.e., those individuals over the age of 65 years),
ageism as studied by researchers in the organizational sciences may involve per-
ceptions about a slightly younger group of people. For example, in the United
States, the Age Discrimination in Employment Act (ADEA, 1967) stipulates that
employees over the age of 40 can file an age discrimination claim. Such an age is
hardly considered elderly by most standards. Thus, such legislation requires both
a broader definition of an “older” person or worker and caution in the application
of gerontological research findings to organizational situations.
There is also a need for research that reconciles the applied research show-
ing little decline in job performance with age (e.g., Wilkening, 2002), the cogni-
tive aging research showing a sharp decline (e.g., Salthouse, 2003), and the
ageism literature. Salthouse (2004) argued that although there is a sharp decline
in cognitive functioning as adults get older, the impairment rarely impacts every-
day functioning. This is because factors such as motivation, persistence, adapta-
tion, and experience inhibit performance declines. Also, many tasks (including
job tasks) rarely require the level of cognitive functioning required in laboratory
studies of this effect. However, it appears that individuals continue to possess age-
related attitudes regarding both people and jobs. An important goal for future
research would be to identify the antecedents of ageist attitudes and the mediat-
ing variables explaining how such attitudes lead to age bias. Rupp et al. (in press)
have offered causal attributions as one potential mediator, but much more
research is needed in this area.
Lastly, it would be fruitful for future research to study the construct of ageism
across cultures. Such studies should look toward the cross-cultural age bias liter-
ature for informative models (e.g., Bennington, 2001; Bennington & Wein, 2002;
Chiu et al., 2001; Ho et al., 2000; McMullin & Marshall, 2001; Taylor & Walk-
er, 1997; Van den Heuvel, 1999). Such a pursuit is especially relevant because
other countries are in the process of passing legislation outlawing age discrimi-
nation. For example, Reade (2003) has estimated that in England, where such leg-
islation is expected to pass by 2006, organizations not preparing themselves for
these new laws could expose themselves to £73 billion (over $118 billion) in
claims. Indeed, researchers must first seek to understand the construct of ageism
before trying to understand its effects on individuals and societies. The present
study represents one step in this direction.
In conclusion, the present study offers additional psychometric evidence
358 The Journal of Social Psychology
that the FSA constitutes a reliable, valid, and multidimensional measure of
ageism. The presence of both cognitive and affective components of ageism is
compatible with the theoretical work of Butler (1969) and extends the assess-
ment of ageism beyond previous measures that have been primarily cognitive in
nature. It is hoped that other researchers will find the FSA to be a useful instru-
ment for exploring the antecedents and consequences of ageism. The present
study also showed that men and younger individuals were significantly more
ageist than women and older individuals. More research is needed exploring why
such differences exist, as well as which additional variables might mediate or
moderate these effects.
1. It should be noted that this overall finding was shown (Gordon & Arvey, 2004) to
be moderated, in various degrees, by numerous factors such as publication date, rater
demographics, research design, amount and type of target information, and dimensions or
Age Discrimination in Employment Act of 1967, 42 U.S.C. § 6101 et seq.
Allport, G. W. (1958). The nature of prejudice. Garden City, NY: Doubleday Anchor
Banaji, M. R. (1999). Unconscious isms: Examples from racism, sexism, and ageism. Paper
presented at the Way Women Lean Conference, New Haven, CT.
Bell, B. D., & Stanfield, G. G. (1973a). The aging stereotype in experimental perspective.
Gerontologist, 13, 341–344.
Bell, B. D., & Stanfield, G. G. (1973b). Chronological age in relation to attitudinal judg-
ments: An experimental analysis. Journal of Gerontology, 28, 491–496.
Bennington, L. (2001). Age discrimination: Converging evidence from four Australian
studies. Employee Responsibilities and Rights Journal, 13(3), 125–134.
Bennington, L., & Wein, R. (2002). Aiding and abetting employer discrimination:The job
applicant’s role. Employee Responsibilities and Rights Journal, 14, 3–16.
Berg, C. A., & Sternberg, R. J. (1992). Adults’ conceptions of intelligence across the life
span. Psychology and Aging, 7, 221–231.
Braithwaite, V. (2002). Reducing ageism. In T. D. Nelson (Ed.), Ageism: Stereotyping and
prejudice against older persons (pp. 311–338). Cambridge, MA: MIT Press.
Butler, R. N. (1969). Age-ism: Another form of bigotry. Gerontologist, 9, 243–246.
Butler, R. N. (1980). Ageism: A foreword. Journal of Social Issues, 36, 8–11.
Butler, R. N. (1995). Ageism. In G. Maddox (Ed.), The encyclopedia of aging (2nd ed.,
pp. 38–39). New York: Springer.
Byrne, B. M. (1998). Structural equation modeling with LISREL, PRELIS, and SIMPLIS:
Basic concepts, applications, and programming. Mahwah, NJ: Erlbaum.
Chiu, W. C. K., Chan, A. W., Snape, E., & Redman, T. (2001). Age stereotypes and dis-
criminatory attitudes towards older workers: An East-West comparison. Human Rela-
tions, 54, 629–661.
Cleveland, J. N., & Landy, F. J. (1983). The effects of person and job stereotypes on two
personnel decisions. Journal of Applied Psychology, 68, 609–619.
Cleveland, J. N., & Shore, L. M. (1992). Self- and supervisory perspectives on age and
Rupp, Vodanovich, & Credé 359
work attitudes and performance. Journal of Applied Psychology, 77, 469–484.
Cohen, E. S. (2001). The complex nature of ageism: What is it? Who does it? Who per-
ceives it? Gerontologist, 41, 576–577.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hills-
dale, NJ: Erlbaum.
Cronbach, L. J., & Meehl, P. E. (1955). Construct validity in psychological tests. Psycho-
logical Bulletin, 52, 281–302.
Cuddy,A. J. C., & Fiske, S. T. (2002). Doddering but dear: Process, content, and function
in stereotyping of older persons. In T. D. Nelson (Ed.), Ageism: Stereotyping and prej-
udice against older persons (pp. 3–26). Cambridge, MA: MIT Press.
Deaux, K. (1985). Sex and gender. Annual Review of Psychology, 36, 49–81.
Employment practice liability: Jury award trends and statistics. (2001). Arlington, VA:
Erber, J. T., Szuchman, L. T., & Rothberg, S. T. (1990). Everyday memory failure: Age
difference in appraisal and attribution. Psychology and Aging, 5, 236–241.
Faley, R. H., Kleiman, L. S., & Lengnick-Hall, M. L. (1984). Age discrimination and per-
sonnel psychology: A review and synthesis of the legal literature with implications for
future research. Personnel Psychology, 37, 327–350.
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7,
Finkelstein, L. M., & Burke, M. J. (1998). Age stereotyping at work: The role of rater and
contextual factors on evaluations of job applicants. The Journal of General Psycholo-
gy, 125, 317–345.
Finkelstein, L. M., Burke, M. J., & Raju, N. S. (1995). Age discrimination in simulated
employment contexts: An integrative analysis. Journal of Applied Psychology, 80,
Fiske, S. T. (2002). What we know about bias and intergroup conflict, the problem of the
century. Current Directions in Psychological Science, 11(4), 123–128.
Forteza, J. A., & Prieto, J. M. (1994). Aging and work behavior. In H. C. Triandis & M.
D. Dunnette (Eds.), Handbook of industrial and organizational psychology (pp.
446–483). Palo Alto, CA: Consulting Psychologists Press.
Fraboni, M., Saltstone, R., & Hughes, S. (1990). The Fraboni Scale of Ageism (FSA):
An attempt at a more precise measure of ageism. Canadian Journal on Aging, 9,
Golde, P., & Kogan, N. (1959). A sentence completion procedure for assessing attitudes
toward old people. Journal of Gerontology, 14, 355–363.
Gordon, R. A., & Arvey, R. D. (1986). Perceived and actual ages of workers. Journal of
Vocational Behavior, 28, 21–28.
Gordon, R. A., & Arvey, R. D. (2004). Age bias in laboratory and field settings: A meta-
analytic investigation. Journal of Applied Social Psychology, 34, 1–27.
Hassell, B. L., & Perrewe, P. L. (1995). An examination of beliefs about older workers:
Do stereotypes still exist? Journal of Organizational Behavior, 16, 457–468.
Hellbusch, J. S., Corbin, D. E., Thorson, J. A., & Stacy, R. D. (1994). Physicians’attitudes
towards aging. Gerontology and Geriatrics Education, 15(2), 55–65.
Ho, L. S., Wei, X. D., & Voon, J. P. (2000). Are older workers disadvantaged in the Hong
Kong labor market? Asian Economic Journal, 14, 283–300.
Holtzman, J. M., Beck, J. D., & Kerber, P. E. (1979). Dimensional aspects of attitudes
toward the aged. Gerontologist, 19, 91.
Intrieri, R. C., von Eye, A., & Kelly, J. A. (1995). The aging semantic differential:A con-
firmatory factor analysis. Gerontologist, 35, 616–621.
Issacharoff, S., & Harris, E. W. (1997). Is age discrimination really age discrimination?
360 The Journal of Social Psychology
The ADEA’s unnatural solution. New York University Law Review, 72, 780–840.
Kalavar, J. M. (2001). Examining ageism: Do male and female college students differ?
Educational Gerontology, 27, 507–513.
Kite, M. E., & Johnson, D. T. (1988). Attitudes toward older and younger adults: A meta-
analysis. Psychology and Aging, 3, 233–244.
Kite, M. E., & Stockdale, G. D. (2004). Attitudes toward younger and older adults: An
updated meta-analytic review. Journal of Social Issues.
Kite, M. E., & Wagner, L. S. (2002). Attitudes toward older and younger adults. In T. D.
Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 129–161).
Cambridge, MA: MIT Press.
Kogan, N. (1961). Attitudes toward old people: The development of a scale and an exam-
ination of correlates. Journal of Abnormal and Social Psychology, 62, 44–54.
Kogan, N., & Shelton, F. C. (1962). Images of “old people” and “people in general” in an
older sample. The Journal of Genetic Psychology, 100, 3–21.
Laczko, F., & Philipson, C. (1991). Changing work and retirement: Social policy and the
older worker. Milton Keynes, England: Open University Press.
Landy, F. J. (1992). Alternative to chronological age in determining standards of suitability
for public safety jobs. Volume 1: Technical report. University Park, PA: Center for
Applied Behavioral Sciences.
Levy, B., Ashman, O., & Dror, I. (2000). To be or not to be: The effects of aging stereo-
types on the will to live. Omega—Journal of Death and Dying, 40(3), 409–420.
Levy, B. R., & Banaji, M. R. (2002). Implicit ageism. In T. D. Nelson (Ed.), Ageism: Stereo-
types and prejudice against older persons (pp. 49–75). Cambridge, MA: MIT Press.
Levy, B. R., Hausdorff, J. M., Hencke, R., & Wei, J. Y. (2000). Reducing cardiovascular
stress with positive self-stereotypes of aging. Journal of Gerontology, 55, 205–213.
Liden, R. C., Stilwell, D., & Ferris, G. R. (1996). The effects of supervisor and subordi-
nate age on objective performance and subjective performance ratings. Human Rela-
tions, 48, 327– 347.
Maurer, T. J., & Rafuse, N. E. (2001). Learning, not litigating: Managing employee devel-
opment and avoiding claims of age discrimination. Academy of Management Executive,
McCann, R., & Giles, H. (2002). Ageism in the workplace:A communication perspective.
In T. D. Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp.
163–199). Cambridge, MA: MIT Press.
McMullin, J. A., & Marshall, V. W. (2001). Ageism, age relations, and garment industry
work in Montreal. Gerontologist, 41, 111–122.
Mowery, D. C., & Kamlet, M. S. (1996). New technologies and the aging work force. In
S. A. Bass, F. G. Caro, & Y. Chen (Eds.), Achieving a productive aging society (pp.
81–95). Westport, CT: Auburn House.
Palmore, E. B. (1977). Facts on aging: A short quiz. Gerontologist, 17, 315–320.
Palmore, E. B. (1999). Ageism: Negative and positive. New York: Springer.
Palmore, E. B. (2001). The Ageism Survey: First findings. Gerontologist, 41, 572–575.
Park, D. C. (1994). Aging, cognition, and work. Human Performance, 7, 181–205.
Perry, E. L., Kulik, C. T., & Bourhis, A. C. (1996). Moderating effects of personal and
contextual factors in age discrimination. Journal of Applied Psychology, 81, 628–647.
Ragan, A. M., & Bowen,A. M. (2001). Improving attitudes regarding the elderly popula-
tion: The effects of information and reinforcement for change. Gerontologist, 41,
Reade, Q. (2003, April 8). Failure to prepare for ageism laws may cost firms billions. Per -
sonnel Today, 7.
Rosencranz, H. A., & McNevin, T. E. (1969). A factor analysis of attitudes toward the
Rupp, Vodanovich, & Credé 361
aged. Gerontologist, 9, 55–59.
Rupp, D. E., Vodanovich, S. J., & Credé, M. (in press). Age bias in the workplace: The
impact of ageism and causal attributions. Journal of Applied Social Psychology.
Salthouse, T. A. (2003). Interrelations of aging, knowledge, and cognitive performance. In
U. Staudinger & U. Lindenberger (Eds.), Understanding human development: Lifespan
psychology in exchange with other disciplines (pp. 265–287). Berlin, Germany: Kluw-
Salthouse, T. A. (2004). What and when of cognitive aging. Current Directions in Psy-
chological Science, 13, 140–144.
Segrave, K. (2001). Age discrimination by employers. Jefferson, NC: McFarland & Com-
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G.
Austin & S. Worschel (Eds.), The social psychology of intergroup relations. Monterey,
Taylor, P., & Walker, A. (1997). Age discrimination and public policy. Personnel Review,
Tuckman, J., & Lorge, I. (1953). Attitudes toward old people. The Journal of Social Psy-
chology, 37, 249–260.
Underwood, D. G., Eklund, S. J., & Whisler, S. (1985). A reexamination of the factor struc-
ture of the aging semantic differential using a generalized social object. Education
Gerontology, 11, 321–335.
van den Heuvel, A. (1999). Mature age workers: Are they a disadvantaged group in the
labour market? Australian Bulletin of Labour, 25(1), 11–22.
von Hippel, W., Silver, L. A., & Lynch, M. E. (2000). Stereotyping against your will: The
role of inhibitory ability in stereotyping and prejudice among the elderly. Personality
and Social Psychology Bulletin, 26, 523–532.
Waldman, D. A., & Avolio, B. J. (1986). A meta-analysis of age differences in job perfor-
mance. Journal of Applied Psychology, 71, 33–38.
Warr, P. (1994). Age and employment. In H. C. Triandis & M. D. Dunnette (Eds.), Hand-
book of industrial and organizational psychology (pp. 485–550). Palo Alto, CA: Con-
sulting Psychologists Press.
Whitbourne, S. K., & Hulicka, I. M. (1990). Ageism in undergraduate texts. American Psy-
chologist, 45, 1127–1136.
Wilkening, R. (2002). The age 60 rule: Age discrimination in commercial aviation. Avia -
tion, Space, and Environmental Medicine, 73(3), 194–202.
Williams, A., & Nussbaum, J. F. (2001). Intergenerational communication across the life-
span. Mahwah, NJ: Erlbaum.
Wood, J. V. (1989). Theory and research concerning social comparison of personal attrib-
utes. Psychological Bulletin, 106, 231–248.
Young, P., Rinehart, J. S., & Baits, D. M. (1997). Age discrimination: Impact of chrono-
logical age and perceived position demands on teacher screening decisions. Journal of
Research and Development in Education, 30, 103–112.
Received January 23, 2004
Accepted September 14, 2004
362 The Journal of Social Psychology