Cross-Cultural Differences in The Acceptance of Barnum Profiles Supposedly Derived From Western Versus Chinese Astrology
Cross-Cultural Differences in The Acceptance of Barnum Profiles Supposedly Derived From Western Versus Chinese Astrology
Cross-Cultural Differences in The Acceptance of Barnum Profiles Supposedly Derived From Western Versus Chinese Astrology
discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/45637820
CITATIONS READS
3 232
2 authors, including:
Paul Rogers
Goldsmiths, University of London
22 PUBLICATIONS 288 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Paul Rogers on 30 March 2016.
Dear Author/Editor,
Greetings, and thank you for publishing with Sage Publications. Your article has been copyedited, and we
have a few queries for you. Please respond to these queries by marking up your proofs with the necessary
changes/additions.
No
The present study examines cross-cultural differences in the Barnum effect. At Stage 1 of the
study, 287 respondents (comprising 149 Westerners and 138 Chinese nationals) provided birth
details and completed a belief-in-astrology questionnaire. At Stage 2 a week later, 258 of these
(130 Westerners, 128 Chinese) then completed a second belief-in-astrology questionnaire
before receiving a Barnum profile supposedly derived from either Western or Chinese astrol-
ogy, which they rated for, among other things, perceived accuracy (a) for themselves and
(b) for other people in general. Although preliminary analysis offered initial support for a
universal Barnum effect, this disappeared after respondent gender, age, general education, and
psychological knowledge were controlled for. Further analyses revealed little support for
cross-cultural differences in either astrological beliefs or susceptibility to the Barnum effect,
although surprisingly, Chinese nationals who believed in astrology did perceive their own
(Barnum) profile to be more accurate for people in general than did Chinese skeptics. Finally,
Barnum acceptance was not influenced by the apparent source of profiles. The role these
factors play in relation to Barnum susceptibility and methodological limitations of the present
study are discussed.
D espite limited scientific evidence for astrology (Culver & Ianna, 1984; Eysenck &
Nias, 1982; Groome, 2001), thousands of people continue to read their daily horo-
scopes and/or consult astrologers for advice and personal guidance (Roe, 1998[AQ: 1]).
Correspondingly, Chinese astrology seems just as popular in China (e.g., Yip, Lee &
Cheng, 2002). According to skeptics, the apparent validity of astrology stems from peoples’
tendency to accept vague, ambiguous, and generalized personality descriptions as being
Authors’ Note: The authors would like to thank various staff in the University of Central Lancashire’s
Department of Languages and International Studies, most notably, Dr. Isobel Donnelly and Dr. Brian
Austerfield, for allowing us to sample their Chinese students. Special thanks also go to Ms. Lingling Peng and
Dr. Xiaohu Guo for (back-) translating study materials and to Dr. Thanzami Van Lai plus several anonymous
reviewers for their advice regarding earlier drafts of this article. Correspondence should be addressed to Dr.
Paul Rogers, Department of Psychology, Darwin Building, University of Central Lancashire, Preston,
Lancashire PR1 2HE, United Kingdom; e-mail: [email protected].
1
2 Journal of Cross-Cultural Psychology
These findings suggest that even a person’s knowledge of—rather than merely his or her
belief in—astrology is not sufficient to offset Barnum susceptibility.
Other work suggests that Barnum acceptance cannot be explained as mere gullibility
(Dean, 1991; Piper-Terry & Downey, 1998; Standing & Keays, 1985), demand character-
istics (Johnson, Cain, Falke, Hayman, & Perillo, 1985; MacDonald & Standing, 2002), or
a greater motivation to find self-referent matches within profiles (Wiseman & Smith, 2002).
In sum, susceptibility to the Barnum effect reflects a general information processing bias
whereby individuals consistently fail to discriminate unique from universal character
descriptions and thus fall victim to a fallacy of personal validation (Forer, 1949). This can
lead to the assimilation of profile content into one’s own self-concept (Hamilton, 2001;
Lillqvist & Lindeman, 1998; Van Rooij, 1999) and ultimately to confirmation (Davies,
1997[AQ: 2]) and self-serving biases (MacDonald & Standing, 2002).
2002), coupled with evidence that astrological believers are more prone to ethnic prejudice
(Dambrun, 2004), it seems reasonable to suggest that Chinese students will be especially
prone to accepting Barnum profiles seemingly derived from Chinese astrology.
Method
Respondents
A total of 365 respondents (208 Westerners, 157 Chinese) were sampled from classes at
the University of Central Lancashire (UCLan). Of these, 287 respondents (149 Westerners,
138 Chinese) returned usable questionnaires at this first stage of sampling (see Procedure
subsection below), a preliminary response rate of 78.6%. Most (n = 258) also returned usable
questionnaires at Stage 2, for a final response rate of 70.7%. This final sample of 258
respondents comprised 130 Westerners plus 128 Chinese individuals,3 the majority of whom
(69.5%) were female. Overall, respondents were ages 18 to 48 years (M = 22.0 years; SD =
4.9 years) with the vast majority (97.3%) educated to at least A level or equivalent.
As expected, virtually all Chinese respondents (99.2%) claimed English was not their
first language, with the vast majority (98.4%) also stating they had not lived in the United
Kingdom, Europe, or America prior to coming to UCLan. Any claiming otherwise or with
missing language and/or residence data were removed from further analyses. The remain-
ing Chinese respondents had spent an average of 1.68 years (SD = 1.00 years) studying in
the United Kingdom.
Design
This study had a 2 (sample type: Western vs. Chinese respondents) × 2 (apparent profile
source: Western vs. Chinese astrology) × 2 (belief in astrology: believer vs. nonbeliever)
between-subjects design. Dependent variables were two measures of perceived profile accu-
racy (i.e., for self and for other people in general) plus three support-for-astrology measures
(i.e., profile helpfulness, profile tone, and “astrology works” ratings) as described below.
Materials
Several new questionnaires were constructed for use in the present study. All versions for
the Chinese sample were rewritten into Chinese by an experienced translator (LP), a Chinese
postgraduate student from the UCLan’s Department of Languages and International Studies.
The equivalence of English and Chinese materials was later verified through back-translation
(Brislin, 1970) by a Chinese member of staff (XG) recruited through the same department.4
BDQ. The BDQ included a brief historical description of either Western or Chinese
astrology before asking participants to give their name plus date, time, and place of birth
6 Journal of Cross-Cultural Psychology
to create the illusion that a real astrological profile was to be generated. No BDQ data were
actually analyzed.
BIAQ. The BIAQ comprised 11 items assessing people’s beliefs about the use of astrol-
ogy for predicting various aspects of life, such as a one’s personality, relationship out-
comes, career or business decisions, and the likelihood war or natural disasters. Two further
items assessed how scientific and how trustworthy astrology was seen to be. These items
were rated along a 7-point Likert-type scale from 1 (strongly disagree) to 7 (strongly
agree), with higher scores reflecting more belief in (the validity of) astrology.
The BIAQ also included five astrology experience and knowledge items. These assessed
the frequency with which respondents read their horoscope (from 1 = never to 7 = all the
time), visited a practicing astrologer and/or called an astrology telephone helpline (both from
1 = never to 7 = more than 12 times per year), how much respondents felt they knew about
astrology (from 1 = absolutely nothing to 7 = I can interpret an entire astrological chart),
how much money (in British pounds) they had spent on astrology in the past year, and
finally, whether they had participated in a similar study of astrology (yes or no). Although
no distinction was made between Western versus Chinese astrology on any specific item, one
of the two systems was implicated via the historical introduction outlined above.
Profile Accuracy Questionnaire (PAQ). The PAQ comprised five dependent measures,
two assessing the perceived accuracy of profiles (i.e., for self and for other people in gen-
eral) plus three assessing the extent to which profiles appeared to offer support for the
validity of astrology (i.e., profile helpfulness, profile tone, and “astrology works”; see
Appendix B). All dependent variables were rated along a 7-point Likert-type scale from 1
(not at all or none) to 7 (extremely or totally).
Demographics. Finally, respondents were asked to indicate their gender, age, ethnicity,
educational qualification, current occupational status (full-time student, part-time student,
or other), and degree subject; whether English was their first language; and how long they
had studied in the United Kingdom, in Europe, and/or in America. Each participant was
also given a unique identification code.
Procedure
All respondents were students at UCLan, a large university in the northwest of England.
Western and Chinese students were opportunistically sampled from a 1st-year undergraduate
Rogers, Soule / Cross-Cultural Difference in the Barnum Effect 7
Results
Demographics
Preliminary analyses revealed several demographic differences across the two sample
groups. First, the Western sample (75.4%) contained more females than the Chinese
(63.5%) sample (phi = –.13, p < .05, n = 256). Second, Western respondents (M = 21.5
years, SD = 6.5 years) were significantly younger than Chinese respondents (M = 22.7
years, SD = 2.24 years), t(160.2) = –2.05, p < .05. Third, Westerners who were qualified to
A level or equivalent (median = 3) had fewer qualifications, χ2(df = 5, N = 254) = 125.11,
p < .001, than their Chinese counterparts who were qualified to Higher National Diploma
or undergraduate degree level (median = 4). Finally, more Westerners (69.4%) had studied
at least some psychology (phi = –.72, p < .001, n = 250) than Chinese respondents (0.0%).
Future analyses will partial out respondent gender, age, general educational attainment, and
psychological knowledge as potential covariates.
Belief in Astrology
All analyses involving BIAQ scores were conducted across the two samples separately.
Reliability tests revealed that at preprofile (Stage 1) BIAQ scores had a high internal reli-
ability for both Western (alpha = .97) and Chinese (alpha = .90) samples. Similarly, post-
profile (Stage 2) BIAQ scores were again highly reliable for both Westerners (alpha = .95)
and Chinese (alpha = .93) groups. Correlational analysis across the two stages confirmed
that the BIAQ had high test–retest reliability for both Western (r = .83, p < .001, n = 83)
and Chinese (r = .76, p < .001, n = 82) respondents. Hereafter, all references to BIAQ rat-
ings refer to those at Stage 1 unless otherwise stated.5 Finally, analysis of covariance
(ANCOVA)—controlling for respondent gender, age, general education, and psychological
knowledge—revealed that Westerners and Chinese respondents did not differ in their BIAQ
scores at either Stage 1 or Stage 2 sessions.6 Mean BIAQ scores are given in Table 1.
Table 1
Belief in Astrology Questionnaire (BIAQ) Ratings
at Stages 1 and 2 Across Sample Group
Western (n = 130) Chinese (n = 128) All
Note: Significant sample effects found at the *p < .05, **p < .01, and ***p < .001 levels (two-tailed).
[AQ: 4]
received a full astrological reading from a practicing astrologer, only a fifth of Chinese
respondents (19.5%) made this claim. Sample group comparisons via ANCOVA—again
controlling for the aforementioned covariates—confirmed that on average, Westerns
(median = 1) had received significantly fewer full readings, F(1, 142) = 21.76, p <. 001,
eta2 = .13, than their Chinese counterparts (median = 3). In addition, most Western and
Chinese respondents had never telephoned an astrology help line (98.8% and 82.9%,
respectively), with the two groups spending comparatively little on astrology in the past
year (£2.06 and £1.81, respectively). These differences were not significant. Likewise, with
gender, age, education, and psychological knowledge all controlled for, Westerners’
reported general knowledge of astrology did not differ significantly from that of their
Chinese counterparts (median = 3 and 4, respectively).
Further analyses revealed significant positive correlations between BIAQ ratings and the
extent to which both Western and Chinese respondents claimed they read their daily news-
paper horoscopes (rho = .37, p < .001, n = 130, and rho = .35, p < .001, n = 128, respec-
tively), received full astrological readings (rho = .33, p < .001, n = 130, and rho = .26,
p < .005, n = 128, respectively) and spent money on astrology in that year (rho = .32, p <
.001, n = 130, and rho = .19, p < .05, n = 128, respectively). Western and Chinese BIAQ
ratings also correlated positively with respondents reported knowledge of astrology (rho =
.29, p < .005, n = 130, and rho = .39, p < .001, n = 125, respectively). Taken together, these
data provide good evidence of the BIAQ’s content and concurrent validity (Howitt &
Cramer, 2005) for both samples.
Respondent Subgroups
Stage 1 BIAQ ratings were dichotomized via median split analysis for Western (median =
2.63) and Chinese (median = 3.27) samples separately, thereby creating a believer-versus-
nonbeliever dichotomy for both groups. Eight respondents who scored median BIAQ
ratings for their respective sample were removed from further analyses. Of the remaining
250 respondents, 161 returned useable profile (PAQ) scores, a Stage 2 response rate of
64.4%. This smaller sample comprised 82 Westerners (40 Western believers and 42 Western
nonbelievers) plus 79 Chinese respondents (43 Chinese believers and 36 Chinese nonbe-
lievers). All subsequent analyses are performed on this revised sample.
Rogers, Soule / Cross-Cultural Difference in the Barnum Effect 9
Support-for-Astrology Ratings
Separate 2 (sample type: Western vs. Chinese) × 2 (apparent profile source: Western vs.
Chinese astrology) × 2 (belief in astrology: believer vs. nonbeliever) between-subjects
ANCOVA—controlling for the same four covariates—were performed on each of the three
remaining support-for-astrology dependent variables (i.e., profile helpfulness, profile tone, and
“astrology works”). A single significant covariate was found, with respondent gender posi-
tively associated with profile tone, F(1, 132) = 4.54, p < .05, eta2 =.03, such that males per-
ceived the profile to be less positive than females. As Table 2 shows, a significant main effect
for belief was found in ratings for helpfulness, F(1, 133) = 24.68, p < .001, eta2 = .16; profile
tone, F(1, 132) = 5.79, p < .05, eta2 = .04; and “astrology works,” F(1, 132) = 21.00, p < .001,
eta2 =.14; astrological believers rated profiles as more helpful, more positive, and providing
more evidence that astrology works than did nonbelievers. Similarly, a significant sample
effect was found in scores for both helpfulness, F(1, 133) = 15.11, p < .001, eta2 =.10, and
“astrology works,” F(1, 141) = 5.94, p < .05, eta2 =.04, with Westerners rating profiles less
helpful and as providing less evidence that astrology works than did their Chinese counter-
parts. No other significant main or interaction effects were found for these three measures.
Accuracy Believer 4.73 (1.44) 4.50 (1.29) 4.61 (1.34) 4.95 (1.32) 4.45 (1.43) 4.70 (1.38) 4.86 (1.35) 4.47 (1.35) 4.66 (1.36) B **
for self Nonbeliever 5.48 (0.87) 4.72 (1.18) 5.13 (1.08) 4.65 (1.27) 5.33 (0.98) 4.97 (1.18) 5.11 (1.13) 5.00 (1.12) 5.06 (1.12) A × S × B *
All 5.17 (1.18) 4.61 (1.23) 4.89 (1.23) 4.81 (1.29) 4.83 (1.32) 4.82 (1.29) 4.99 (1.24) 4.72 (1.27) 4.85 (1.26)
Accuracy Believer 4.67 (1.29) 4.28 (1.36) 4.45 (1.33) 4.15 (1.18) 3.50 (1.28) 3.83 (1.26) 4.37 (1.24) 3.87 (1.36) 4.11 (1.32)
for Nonbeliever 4.57 (1.12) 4.72 (1.18) 4.64 (1.14) 5.00 (1.00) 4.87 (1.06) 4.94 (1.01) 4.76 (1.08) 4.79 (1.11) 4.77 (1.08)
others
All 4.61 (1.18) 4.50 (1.28) 4.56 (1.22) 4.54 (1.17) 4.09 (1.36) 4.32 (1.28) 4.58 (1.17) 4.30 (1.32) 4.44 (1.25)
Profile Believer 2.27 (1.33) 2.39 (1.46) 2.33 (1.38) 3.80 (1.40) 3.55 (1.54) 3.68 (1.46) 3.14 (1.56) 3.00 (1.59) 3.07 (1.57) B ***
helpful- Nonbeliever 4.05 (1.43) 3.28 (1.64) 3.69 (1.56) 4.76 (0.97) 4.38 (0.96) 4.58 (0.97) 4.37 (1.28) 3.79 (1.45) 4.10 (1.39) S ***
ness All 3.31 (1.64) 2.83 (1.59) 3.07 (1.62) 4.24 (1.30) 3.92 (1.36) 4.08 (1.33) 3.78 (1.54) 3.38 (1.57) 3.58 (1.56)
Profile tone Believer 4.13 (0.64) 3.83 (1.47) 3.97 (1.16) 4.10 (1.25) 4.05 (1.36) 4.08 (1.29) 4.11 (1.02) 3.95 (1.39) 4.03 (1.22) B ***
Nonbeliever 4.43 (1.16) 4.17 (0.86) 4.31 (1.03) 4.82 (0.95) 4.87 (1.06) 4.84 (0.99) 4.61 (1.08) 4.48 (1.00) 4.55 (1.04)
All 4.31 (0.98) 4.00 (1.20) 4.15 (1.10) 4.43 (1.17) 4.40 (1.29) 4.42 (1.22) 4.37 (1.07) 4.20 (1.25) 4.28 (1.16)
Astrology Believer 2.53 (1.68) 2.44 (1.20) 2.48 (1.42) 3.75 (1.33) 3.55 (1.70) 3.65 (1.51) 3.23 (1.59) 3.03 (1.57) 3.12 (1.57) B ***
works Nonbeliever 4.19 (1.36) 3.28 (1.41) 3.77 (1.44) 4.53 (0.87) 4.73 (1.16) 4.63 (1.01) 4.34 (1.17) 3.94 (1.48) 4.15 (1.33) S *
All 3.50 (1.70) 2.86 (1.36) 3.18 (1.56) 4.11 (1.20) 4.06 (1.59) 4.08 (1.39) 3.81 (1.49) 3.45 (1.58) 3.63 (1.54)
the relationship between each group’s reported knowledge of astrology and their degree of
Barnum acceptance. Analyses for the Western sample revealed a single significant, positive
correlation between knowledge of astrology and accuracy for other ratings (r = .28, p < .05,
n = 82), with more knowledgeable Westerners perceiving Barnum profiles to be more accu-
rate for other people in general. For the Chinese sample, significant and positive correla-
tions were found between astrological knowledge and both profile tone (r = .23, p < .05,
n = 80) and “astrology works” (r = .32, p < .005, n = 80) ratings, with more knowledgeable
Chinese respondents perceiving profiles to be more positive and as providing more evi-
dence that astrology works. All other correlations were nonsignificant. To compare the
comparative strength of these correlations across the two samples, Pearson’s (r) correla-
tions were subjected to Fisher Z transformations followed by independent samples t tests.
Results confirmed that Westerners’ reported knowledge of astrology was just as strongly
associated with accuracy for others, profile tone, and “astrology works” ratings as they
were for Chinese respondents.
Discussion
Previous studies (e.g., French et al., 1991/1998; Glick et al., 1989; Sosis, Strickland, &
Hakey, 1980) have assessed belief in astrology using only a one- or a two-item measure.
Unlike these, the present study is the first to incorporate a psychometrically sound, multi-
ple-item measure of astrological belief. Several interesting findings emerged.
First, having controlled for four potential covariates (i.e., gender, age, general educa-
tional attainment, and psychological awareness), Chinese respondents were found to have
just as strong astrological beliefs as Western respondents. This unexpected finding is con-
trary to previous claims that Chinese nationals are more likely to endorse the existence of
paranormal phenomena than are people from other cultures (McClenon, 1988, 1990, 1993,
1994). It is possible that having studied in the United Kingdom for just longer than 20
months, the present Chinese sample was more “Westernized” and thus may have developed
a more skeptical view of astrology than those in previous studies. The possibility that
Chinese students here may be atypical of Chinese nationals is recognized throughout the
remainder of this discussion.
Second, respondents generally perceived their (fake) astrological profile to be at least
moderately accurate in describing their own personality. In line with early Barnum studies
(e.g., Forer, 1949; Stachnik & Stachnik, 1980), these findings suggest that respondents in
the present study were generally susceptible to the Barnum effect. But early studies gener-
ally failed to distinguish between measures of self versus other accuracy, relying on just the
12 Journal of Cross-Cultural Psychology
astrological systems (see footnote 2). In sum, it seems Barnum acceptance across cultures
extends to Chinese recipients and exists regardless of the alleged assessment procedure
and/or the apparent cultural source of profiles. Future research should explore whether
cross-cultural similarities exist for other ethnic groups (e.g., Afro-Caribbean, Asian,
Hispanic) and/or for profiles purportedly derived from other sources (e.g., graphology or
clinical assessment).
Interestingly, although Chinese believers and nonbelievers did not differ in their percep-
tions of how accurate Barnum profiles were of their own personality, the former did per-
ceive profiles to be a more accurate description of other people in general. It seems Chinese
believers failed to recognize the vagueness and generality of fake astrological profiles to a
greater extent than Chinese skeptics. Such differences were not found for corresponding
Westerners and, further, were irrespective of apparent profile source. This is ironic, given
that Chinese horoscopes, which are derived from less specific birth data (cf. footnote 2),
really ought to be more generic. One interpretation is that Chinese believers maintained
particularly robust stereotypes about other peoples’ characters, which may in turn reflect a
greater susceptibility to two psychosocial processes, namely, (a) the strong expectancy
effects that underpin all belief in astrology (Munro & Munro, 2000) combined with
(b) societal pressures toward norm compliance and the suppression of individuality, which
are intrinsic of collectivist cultures, such as China’s (Nelson et al., 2004; Shiraev & Levy,
2007). More research is needed to explore this possibility.
Sixth, Chinese respondents also judged Barnum profiles to be more helpful and more sup-
portive of the validity of astrology than did Westerners. Surprisingly, this was regardless of
Chinese respondents’ degree of belief in astrology. Thus, in partial support of hypotheses, it
seems Chinese nationals—even those comparatively skeptical of astrology—were more open to
at least the possibility of astrology being a useful and valid tool for assessing personality. The
additional finding that Chinese respondents had received a greater number of full astrology
readings from a practicing astrologer than had Westerners supports this view.
There are several plausible explanations for these data. First, the median split in BIAQ
ratings meant some Chinese nonbelievers really ought to have been labeled low believers
in astrology. But this argument is weakened if one considers the same biases ought to apply
to Westerners, too. Alternatively, findings may simply reflect the tendency for Chinese
societies to endorse a more esoteric or metaphysical worldview (McClenon, 1988, 1990)
and/or make key decisions based on the Chinese zodiac (e.g., Yip et al., 2002). Third, it is
possible that Chinese respondents scored higher on some third mediating factor, such as
locus of control (Rotter, 1966). Previous studies suggest that people from non-Western
societies tend to have a more pronounced external locus of control and thus see events as
being caused by outside forces, such as fate or destiny (Shiraev & Levy, 2007). Given that
an external locus of control is also associated with stronger astrological beliefs (Sosis et al.,
1980) and a greater susceptibility to Barnum acceptance (Snyder & Larsen, 1972; Snyder
& Shenkel, 1976), current trends might reflect a tendency for Chinese respondents to
be particularly extreme in their external attributions. Similar considerations might also be
given to private self-consciousness, which tends to be lower in collectivist cultures
(Triandis, 1989) and is also associated with the acceptance of false personality feedback
(Davies, 1994).
14 Journal of Cross-Cultural Psychology
Seventh, the two groups did not differ in their reported knowledge of astrology, with
both claiming they could at least name the 12 (sun) signs of the zodiac. Westerners who
reportedly had more knowledge of—as opposed to belief in—astrology subsequently
judged their fake astrological (Barnum) profile to be more accurate of people generally than
did Westerners who reported less knowledge of astrology. By comparison, Chinese
respondents claiming more knowledge saw profiles as being more positive. Taken together,
these data offer partial support for the claim that one’s assumed knowledge of astrology is
no protection against misperceiving the generality of Barnum profiles (cf. French et al.,
1991/1998). Furthermore, the lack of sample group differences in the strength of these
associations suggests that this applies to recipients regardless of their cultural background.
Finally, the length of time Chinese nationals spent studying in the United Kingdom had
little impact on their perceptions of Barnum profiles. The implication here is that Chinese
nationals’ susceptibility to such misperceptions is robust enough to withstand exposure to
Western, or at least British, culture.
Methodological Limitations
Several methodological limitations are worthy of mention. First, as noted above, the
Barnum effect was only present when respondent gender, age, general education, and psy-
chological knowledge were not controlled for. This was surprising (cf. Furnham &
Schofield, 1987) and suggests that more research is needed to explore to the impact these
demographic factors have on Barnum acceptance.
Second, the current Chinese sample may have been atypical compared to Chinese
nationals living in China. Furthermore, a rather crude measure of acculturation (time spent
in the United Kingdom) was used to explore levels of cross-cultural transmission. Future
studies should compare Westerners with Chinese recipients unexposed to Western culture
or use a more sensitive measure of how Chinese nationals integrate into Western culture.
Third, Barnum profiles and/or questionnaires may have suffered from limitations in cross-
cultural equivalence, a necessary condition for cross-cultural validity (Chen, Snyder &
Krichbaum, 2002; van de Vijver & Tanzer, 2004). According to van de Vijver and Tanzer
(2004), equivalence can be compromised by the inappropriate translation of culture-specific
terms (“item bias”); cross-cultural differences in the definition, associated behaviors, or cover-
age of construct domains (“construct bias”); or from the use of culturally incomparable sam-
ples and/or variation in instrument familiarity (“method bias”). In the present study, three
profile terms (i.e., affable, wary, and reserved) were back-translated from Chinese to English
with slightly different, albeit conceptually similar, interpretation (i.e., as courteous, suspicious,
and conservative, respectively), suggesting that item equivalence was not perfect. It is possible
these minor item biases had undue influence on Barnum acceptance. Similarly, it is possible
some of the psychological constructs described in a standard Barnum profile (e.g., pride, affa-
bility) were conceptually different for Western versus Chinese respondents. For example,
people living in collectivist cultures tend to generate more group-centric descriptions of self-
hood (Shiraev & Levy, 2007), place more emphasis on moral attributes (Cheng & Watkins,
2000), and associate shame with aspects of family life (Stipek, 1998) more than those living
in individualistic cultures. Future studies should ensure construct equivalence by modifying
Rogers, Soule / Cross-Cultural Difference in the Barnum Effect 15
Barnum profiles accordingly (see Zhang, Oian, & Yuling, 2003). Finally, because Chinese
respondents were recruited via seminars and Westerners via lectures, it is possible group dif-
ferences in class size may have affected Barnum ratings (e.g., via socially desirable respond-
ing), thus leading to some degree of method biases (cf. van de Vijver & Tanzer, 2004). Future
research should account for this.
A final methodological limitation concerns respondents’ focus on their own astrological
system. Although the present study manipulated the apparent profile source by providing a
brief historical overview of either Western or Chinese astrology (Stage 1) and by labeling
the Barnum profile accordingly (Stage 2), other BIAQ items did not differentiate between
the two astrological systems. It may be that respondents given an alien profile may have
answered their belief and knowledge items with their own astrological system in mind. As
such, future cross-cultural studies of astrology and/or the Barnum effect should provide
culture-specific instructions and questionnaire items where appropriate.
Appendix A
The Fake Astrology (Barnum) Profile
Appendix B
Dependent Measures
1. How accurately does this profile describe you personally? [“accuracy for self”]
2. How helpful will this profile be to your everyday life? [“profile helpfulness”]
3. How positive was the content of this profile overall? [“profile tone”]
4. How accurately does this profile describe people in general? [“accuracy for others”]
5. How much does this profile support the claim that astrology works? [“astrology works”]
Notes
1. Genuine horoscopes include characteristics traditionally associated with a particular sun sign (e.g., a
courageous Leo, an optimistic Sagittarian; see Diagram Group, 1993; White, 1993) regardless of the validity of
astrology. By comparison, false horoscopes include characteristics not traditionally associated with a given sign
(e.g., a cowardly Leo, a pessimistic Sagittarian), whereas a fake horoscope refers, simply, to a standard Barnum
profile (see French, Fowler, McCarthy, & Peers, 1991/1998).
2. Like Western astrology, Chinese astrology includes 12 zodiac signs, although the names and hieroglyphs
differ. Several other differences also exist. In Western astrology, an individual’s sun or zodiac sign is determined
by the precise time, date, year, and place of birth, whereas in Chinese astrology, this is determined solely by
the year in which a person is born. Thus, Chinese horoscopes are by definition more generic at least in terms
of to whom they apply. Second, Western astrology reflects the solar year (orbits of the earth around the sun),
whereas Chinese astrology reflects the lunar year (orbits of the moon around the earth). Third, Western astrol-
ogy incorporates four elements (earth, air, fire, and water), compared to Chinese astrology’s five (earth, fire,
water, wood, and metal). Finally, unlike Western astrology, Chinese astrology also incorporates two opposing
but complementary forces of yin (masculine) and yang (feminine), which, it is claimed, influence all other
aspects of the zodiac (see Giles & Diagram Group, 2000; White, 1993). These differences are assumed to be of
little relevance to the present study.
3. The Western sample comprised mainly Caucasian respondents (92.1%), with the Chinese sample consist-
ing of Chinese nationals studying English as a Foreign Language. To avoid ethnic ambiguities, respondents who
either (a) were in the Western sample but who indicated having a Chinese or other Asian (e.g., Indian, Pakistani)
ethnicity, (b) were in the Chinese group but who indicated they were of non-Chinese origin, or (c) failed to state
a specific ethnic origin were all removed from the data set. As such, the term Westerner hereafter refers to
Caucasian respondents of predominantly British origin. Finally, any respondents who stated they had taken part
in previous studies of astrology were also removed. In all, 24 of the 365 individuals initially sampled were
omitted for these reasons.
4. Virtually all questionnaire and profile terms were back-translated into their original English format, with
only three original profile terms (affable, wary, and reserved) translated differently (as courteous, suspicious,
and conservative, respectively). Given the conceptual similarity of these terms, Chinese translations were
deemed acceptable.
5. This is justified on two counts: first, because fewer Western (n = 83) and Chinese (n = 82) respondents
returned useable Belief in Astrology Questionnaires (BIAQ) at Stage 2, and second, because respondents were
new to the study at Stage 1 and thus were deemed less likely to give socially desirable responses.
6. Respondents’ levels of psychological awareness was a significant and positive covariate for Stage 1 BIAQ,
F(1, 222) = 7.27, p = .008, eta2 = .03, but not Stage 2 BIAQ ratings. No other significant covariates were found.
References
Bayne, R. (1980). Interpretations and uses of research on “Barnum” personality statements.
British Journal of Guidance and Counselling, 8(2), 233-236.
Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (2002). Cross-cultural psychology: Research and
applications (2nd ed.). Cambridge, UK: Cambridge University Press.
Rogers, Soule / Cross-Cultural Difference in the Barnum Effect 17
McClenon, J. (1990). Chinese and American anomalous experiences: The role of religiosity. Sociological
Analysis, 51(1), 53-67.
McClenon, J. (1993). Surveys of anomalous experience in Chinese, Japanese and American samples. Sociology
of Religion, 54(3), 295-302.
McClenon, J. (1994). Surveys of anomalous experience: A cross-cultural analysis. Journal of the American
Society for Psychical Research, 88, 117-135.
Meehl, P. E. (1956). Wanted: A good cook book. American Psychologist, 11, 262-272.
Munro, G. D., & Munro, J. E. (2000). Using daily horoscopes to demonstrate expectancy confirmation.
Teaching of Psychology, 27(2), 114-116.
Nelson, L. J., Badger, S., & Wu, B. (2004). The influence of culture in emerging adulthood: Perspectives of
Chinese college students. International Journal of Behavioural Development, 28(1), 26-36.
Piper-Terry, M. L., & Downey, J. L. (1998). Sex, gullibility and the Barnum effect. Psychological Reports, 82,
571-576.
Pulido Rull, M. A. (2000). El efecto Barnum estudiantes universitarios y profesionales de al psicologia en
México [The Barnum effect in university students and psychology professional in Mexico] [English
abstract]. Revista Intercontinental deo Psicologia y Educación, 2(2), 59-66.
Randall, T. M. (1997). Paranormal Short Inventory. Perceptual and Motor Skills, 84, 1265-1266.
Roe, C. A. (1995). Pseudopsychics and the Barnum effect. European Journal of Parapsychology, 11, 76-91.
Rosen, G. M. (1975). Effects of source prestige on subjects’ acceptance of the Barnum effect: Psychologist
versus astrologer. Journal of Consulting and Clinical Psychology, 43, 95. [AQ: 6]
Rotter, J. (1966). Generalized expectancies for internal vs. external control of reinforcement. Psychological
Monographs, 80, 1-28.
Shiraev, E., & Levy, D. (2007). Cross-cultural psychology: Critical thinking and contemporary implications
(3rd ed.). Boston: Pearson.
Snyder, C. R. (1974a). Acceptance of personality interpretations as a function of assessment procedure. Journal
of Counselling and Clinical Psychology, 42, 150.
Snyder, C. R. (1974b). Why horoscopes are true: The effects of specificity on acceptance of astrological inter-
pretations. Journal of Clinical Psychology, 30, 577-580.
Snyder, C. R., & Larsen, G. R. (1972). A further look at student acceptance of general personality interpreta-
tions. Journal of Counselling and Clinical Psychology, 38, 384-388.
Snyder, C. R., Larsen, G. R., & Bloom, L. J. (1976). Acceptance of general personality interpretations prior to
and after receipt of diagnostic feedback supposedly based on psychological, graphological and astrological
assessment procedures. Journal of Clinical Psychology, 32, 251-265.
Snyder, C. R., & Shenkel, R. J. (1976). Effects of favourability, modality and relevance on acceptance of general
personality interpretations prior to and after receiving diagnostic feedback. Journal of Consulting and
Clinical Psychology, 44, 43-41.
Snyder, C. R., Shenkel, R. J., & Lowery, C. R. (1977). Acceptance of personality interpretations: The “Barnum
effect” and beyond. Journal of Consulting and Clinical Psychology, 45, 104-114.
Sosis, R. H., Strickland, B. R., & Haley, W. E. (1980). Perceived locus of control and beliefs about astrology.
Journal of Social Psychology, 110, 65-71.
Stachnik, T., & Stachnik, B. (1980). Acceptance of non-specific astrological personality descriptions: An
empirical demonstration. Psychological Reports, 47, 537-538.
Standing, L., & Keays, G. (1985). Do the Barnum effect and paranormal belief involve a general gullibility
factor? Psychological Reports, 61, 435-438.
Stipek, D. (1998). Differences between Americans and Chinese in the circumstances evoking pride, shame and
guilt. Journal of Cross-Cultural Psychology, 29(5), 616-629.
Thiriart, P. (1991). Acceptance of personality test results. Skeptical Inquirer, 15(2), 161-165.
Tobacyk, J. J. (1988). A revised paranormal belief scale. Unpublished manuscript, Louisiana Technical
University, Ruston.
Tobacyk, J. J., Milford, G., Springer, T., & Tobacyk, Z. (1988). Paranormal beliefs and the Barnum effect.
Journal of Personality Assessment, 52(4), 737-739.
Triandis, H. C. (1989). The self and social behaviour in differing cultural contexts. Psychological Review, 96(3),
506-520.
Rogers, Soule / Cross-Cultural Difference in the Barnum Effect 19
Tyson, G.A. (1982). Why people perceive horoscopes as being true: A review. Bulletin of the British
Psychological Society, 35, 186-188.
Van de Vijver, F., & Tanzer, N. K. (2004). Bias and equivalence in cross-cultural assessment: An overview.
European Review of Applied Psychology, 54(2), 119-135.
Van Rooij, J. J. F. (1999). Self-concept in terms of astrological sun-sign traits. Psychological Reports, 84, 541-546.
White, S. (1993). The new Chinese astrology. London: MacMillan.
Wiseman, R., & Smith, M. D. (2002). Assessing the role of cognitive and motivational biases in belief in the
paranormal. Journal of the Society for Psychical Research, 63(868), 157-166.
Yip, P. S. F., Lee, J., & Cheung, Y. B. (2002). The influence of the Chinese zodiac on fertility in Hong Kong
SAR. Social Sciences and Medicine, 55, 1803-1812.
Zhang, C., Oian, G., & Yuling, S. (2003). Development of the Chinese version TEG. Chinese Mental Health
Journal, 17(8), 560-562.
Paul Rogers has completed a BA (Hons) in economics and social policy and administration (University of
Canterbury, 1985 to 1988), an MSc in experimental psychology (University of Sussex, 1991 to 1992), an MSc
in psychological research methods (University of Exeter, 1994 to 1996), and a PhD in psychology (University
of Hertfordshire, 1997 to 2001). He has worked at the University of Central Lancashire since 2002, first as a
lecturer, and since 2007, as a senior lecturer in psychology. His research interests include the psychology of
belief in the paranormal, the Barnum effect, self-perceived intuitiveness, psychological factors underlying lot-
tery play, and child sexual abuse blame attributions.
Janice Soule graduated with a BSc in psychology (University of Central Lancashire, 2003 to 2006). In that
time, she worked as a volunteer research assistant on several psychology studies and is currently undertaking
an MA in social work at the University of Lancaster.