Home > Psychology > Beyond Allport’s Paradox: The Religion and Prejudice Equation: Part Two

Beyond Allport’s Paradox: The Religion and Prejudice Equation: Part Two


In the second installment of my blog series entitled, Beyond Allport’s Paradox, I review some studies on religious prejudice and follow-up with a brief analysis and critique .  Read Part One here.

Literature Review and Meta-Analyses

Summary: Allport and Ross (1967)

In Allport and Ross’s 1967 study, they examined religiosity as a factor in prejudice using the Intrinsic/Extrinsic Orientation scale. Applying it to religion, they surmised that intrinsically motivated individuals live their religion, while extrinsically motivated people use their religion. Their sample of religious individuals consisted of six American Christian groups who were identified as church goers. One of their hypotheses was that the intrinsically oriented are much less prejudiced than the extrinsically oriented. The conclusion of their study aptly demonstrates the inconsistencies in such research. An analysis of the subsamples indicates that for at least two of the groups, the extrinsically orientated were less prejudiced than the intrinsically oriented. The researchers suggest that local social factors may have contributed to this disparity, particularly due to the fact that these groups showed higher levels of prejudice to one of the three target groups, African Americans. In addition, one of these subsamples showed the reverse trend across all three target groups compared to the other five subsamples.

Analysis: Allport and Ross (1967)

This brief example highlights two of the fundamental problems previously identified. Despite the sample seemingly identified as a homogenous group of Christian church-goers, there were differences. Closer examination shows that each subsample was denominationally different. Perhaps, this played a role in the inconsistent results. In addition, the researchers attempted to reduce religion to a sample of individuals who attended church regularly. The narrow definition of religion used in this study would not be adequate to apply to the SBNR group or other religious groups who reject institutional worship. Given the trends previously indicated, whereby more and more religious people are moving away from institutions, this research fails to hold up across time. Finally, as the authors indicated, there may have been socio-political factors contributing to these disparities. This becomes more salient when noting that this study was conducted during a period in American history that was experiencing some highly contentious racial issues, specifically directed toward African Americans.

Summary: Hasnain and Abidi (2007)

A study originating in India by Hasnain and Abidi (2007) explored the role of religiosity in prejudice and ethnocentrism. Muslims and Hindus were identified as either religious or non-religious based on the frequency of prayer offered. Researchers hypothesized that prejudice and ethnocentrism would differ between the religious and non-religious groups. The results showed that religious people – regardless of their religion – showed higher prejudice and ethnocentrism than nonreligious people, suggesting that religions do not teach intolerance, but that some other factor may be responsible.

Analysis: Hasnain and Abidi (2007)

Two items are particularly interesting to note in this study. For these researchers, the terms Muslim and Hindu may carry more of a cultural label rather than a religious label. In the review of the literature for this paper, there was no research originating from Western countries that defined these groups as anything other than religious. Thus, it would be difficult to replicate this study in countries outside of India. Given the heated nature of some Muslim-Hindu relations in India, it is also possible that results of this study were confounded by local socio-political factors. In addition, religion was defined in very narrow terms, specifically if the participants regularly offered prayer. This reductionist approach negates other factors that contribute to the richness of the religion spectrum. Further, it cannot be adequately applied across populations. Many individuals self-identify as religious regardless of the number of times they may pray. Prayer is also just one of many rituals encompassed within religious practices.

Summary: Ysseldyk, Haslam, Matheson, and Anisman (2011)

This research study examined differences in feelings of Atheists and religious individuals toward (ir)religious groups under group-based threat. Researchers hypothesized that both religious and irreligious people would report more ingroup favouritism and outgroup derogation when the ingroup was threatened. There were some particularly curious findings in this research that suggest further challenges to definitions of religious identification. The first was that self-identified Jewish participants displayed more positive feelings toward Atheists than toward all other religious groups. Atheists were consistently rated most negatively by the other religious groups. Secondly, both Jews and Muslims felt colder to each other than they did toward Christians.

Analysis: Ysseldyk, Haslam, Matheson, and Anisman (2011)

The aforementioned results give ample demonstration of how cultural influences of a given time can influence religious labels. Closer examination of the religious group that defied expected results may suggest that re-evaluations of defining such a group is in order. Jewish participants were drawn from a Canadian sample. Most often these individuals identify with the Reform tradition of Judaism, a very liberal, Americanized denomination of the faith. Thus, this branch of Judaism may be viewed as more modern and progressive than traditional forms of the faith. It may also carry more of a cultural symbolism, rather than religious. This socio-cultural phenomenon may help to explain the differences. In addition, ongoing socio-political hostilities in traditional Middle Eastern homelands of both Muslims and Jews may also contribute to explaining the negative feelings shared between these two groups. Again, the role of culture and politics appears to be confounding research results and limiting application of the examination to time and place.

Summary: Hewstone, Newheiser, and Voci (2011)

This investigation explored the strength of religious attitudes, social dominance orientation (SDO), intergroup contact, and mortality salience on attitudes towards Muslims. It revealed some conflicting conclusions pertaining to the role of religious strength in prejudice. The researchers intended to show that religiosity was a greater predictor of prejudice. In this study, the researchers statistically controlled for individual differences in order to assess the role of religious strength in prejudice. Self-report methods assessing religiosity, SDO, and intergroup contact were completed by both groups. Religious individuals were identified by those indicating a religious affiliation, while non-religious participants were those indicating Atheist, Agnostic, secular, or no response. A reformulation of the data on religious attitude responses yielded three factors (certainty, personal relevance, ambivalence). Paradoxically, the results of this study suggest that individuals who indicated high certainty – or ambivalence – toward religion showed decreased negativity.

Analysis: Hewstone, Newheiser, and Voci (2011)

The authors’ own discussion indicates the potential confounding influences to the contradictory conclusions in this research. They suggest that strength of social and political ideologies may be related to the religion-prejudice relationship. This seems particularly relevant given that the target group of this study was Muslim, and the fact that this study was conducted post 9/11. In addition, the participants were exclusively from a Western society. Also, the manner in which the researchers identified religious and non-religious individuals is not particularly useful for assessing what it is about religion that may influence prejudicial attitudes. Self-identified religious affiliation is a label that can carry subjective meanings. As previously indicated, within a faith tradition there can be vast differences between individuals. This study is also a good example of how incorporating numerous variables and religious measurement scales does not necessarily lead to more consistent research findings. Perhaps then, the issue of finding an appropriate definition of religiousness first, remains a concern that ought to be seriously addressed.

Summary: Rowatt, LaBouff, Johnson, Froese, and Tsang (2009)

This research examined associations between general religiousness and tolerance toward disadvantaged social groups. Three separate hypotheses were tested, and included: 1) general religiousness promotes acceptance, 2) mainstream religion is associated with selective prejudice and, 3) extraneous (or confounding) variables explain associations between religiousness and prejudice. Examining data from the 2007 Baylor Religion Survey (BRS), a mixed mode method (telephone and self-administered mail-in), a sample of 1588 U.S. citizens was used in which responses were converted to z-scores. Using correlational and multiple regression analyses, the results indicated that general religiousness was positively associated with intolerance toward homosexuals and slightly negatively associated with racial prejudice when controlling for other variables. This demonstrated support for an association between general religiousness and selective intolerance, while virtually no corroboration was found for the extraneous variable hypothesis. Some results indicated that certain demographics appear to be associated with attitudes and prejudice.

Analysis: Rowatt, LaBouff, Johnson, Froese, and Tsang (2009)

Religiousness was measured by a four-item assessment that gaged degree of religiosity and commitment to certain activities such as prayer and attendance at services. In addition, approximately 75% of participants self-identified with a Judeo-Christian affiliation. While such activities may fit well within the practices of many of these participants, they are not representative of traditions outside the mainstream Western establishment. Further, given increases in the SBNR group, it is doubtful that this method of defining religiousness would be particularly relevant for this religious group. Finally, the link between prejudice toward homosexuals and religiousness in this study may be confounded by socio-political factors, given the antagonistic climate toward same-sex marriage occurring in the United States at the time this research was undertaken. The authors’ concede that the findings in this study may not be applicable across cultures and times. The role of culture, politics, and, other social factors seems particularly poignant in this investigation. In addition, the methods used to define religion also provide more evidence for the issues of applying such definitions across religious identifications.

Summary: Gervais (2011)

This paper explored the relationship between perceived prevalence of Atheists and prejudice. Contrary to other studies that suggest outgroup size is positively correlated with increased prejudice, the main hypothesis was that decreases in prejudice would occur where increases in presence of Atheism was found. Four studies were implemented to examine the researcher’s hypothesis. Study 1 explored the global association between prejudice toward Atheists and Atheist demographic representation by examining data from the World Values Survey (WVS). Study 2 examined 104 individuals’ responses from a web-based questionnaire assessing anti-Atheist prejudice, perceived Atheist prevalence, belief in God, and belief in a dangerous world. Study 3 and 4 extended the findings in the previous studies by experimentally manipulating perceived prevalence of Atheists and assessing the relationship with anti-Atheist prejudice. Generally, beliefs about the prevalence of Atheists decreased prejudice towards them. In Study 3, however, distrust of Atheists was reduced, but not general prejudice.

Analysis: Gervais (2011)

The results of this study are mixed and may be confounded by the method in which the researcher chose to define Atheism. This problem is particularly salient in Study 1, which captured Atheist prevalence with a single item that asked whether participants believed in God. This implicitly implies a monotheistic, patriarchal worldview of the divine – or a very Judeo-Christian centric worldview. Given that this item was extracted from a worldwide survey, it is particularly problematic as it does not assess the religiosity of groups that may hold alternate theological worldviews, including polytheistic or feminine-centered views of the divine. Further, it does not account for religions that may be considered atheistic, such as Buddhism. By the author’s own admission, a trust-biased single item was used to assess anti-Atheist prejudice in a political context. This cannot be adequately applied to more general measures of prejudice, particularly given the traditionally strong ties between politics and religion in many countries around the world. These limitations suggest, again, that researchers must reassess their definitions of religion. They must be mindful of any inherent biases they may bring into their study, particularly as it pertains to the superimposition of their own religious worldviews onto a global context.

*** Stay tuned for Part Three of Beyond Allport’s Paradox ***

M. xo


Allport, G. W., & Ross, J. M. (1967). Personal religious orientation and prejudice. Journal of Personality and Social Psychology, 5(4), 432-443. doi:10.1037/h0021212

Gervais, W. M. (2011). Finding the faithless: Perceived atheist prevalence reduces anti-atheist prejudice. Personality and Social Psychology Bulletin, 37(4), 543-556. doi:10.1177/0146167211399583

Hasnain, N., & Abidi, K. (2007). Does religiousness promote prejudice and ethnocentrism? Psychological Studies, 52(2), 123-125.

Hewstone, M., Clare, A., Newheiser, A-K., & Voci, A. (2011). Individual and situational predictors of religious prejudice: Impact of religion, social dominance orientation, intergroup contact, and mortality salience. Testing, Psychometrics, Methodology in Applied Psychology, 18(3), 143-155.

Rowatt, W. C., LaBouff, J., Johnson, M., Forese, P., & Tsang, J. (2009). Associations among religiousness, social attitudes, and prejudice in a national random sample of American adults. Psychology of Religion and Spirituality, 1(1), 14-24. doi:10.1037/a0014989

Ysseldyk, R., Haslam, S. A., Matheson, K., & Anisman, H. (2011). Love thine enemy? Evidence that (ir)religious identification can promote outgroup tolerance under threat. Group Processes & Intergroup Relations, 15(1), 105-117.  doi:10.1177/1368430211410996

  1. No comments yet.