Is intrinsic religious orientation a culturally specific American Protestant concept? The fusion of intrinsic and extrinsic religious orientation among non-Protestants

The relationship between intrinsic and personal extrinsic religious orientation as suggested by Gorsuch and McPherson is studied within four denominational samples of university students in four different cultural environments. Results show that intrinsic and extrinsic personal religious orientation form two separate dimensions only within the American Protestant sample. In three different European religious environments (one Eastern Orthodox, one Islamic, and one Roman Catholic), all extrinsic personal and intrinsic items can be combined into a single dimension. It is speculated that the intrinsic orientation may be culturally tied to Protestantism. Copyright # 2007 John Wiley & Sons, Ltd. The multifaceted and complex nature of religiousness has been analytically patterned in many ways that make it more comprehensible. One of these was the classification into components by Glock and Stark (1965). Behind the behaviors, deeds, actions, and institutional affiliations, there are also motives, which are often inseparable from substantial religiousness itself. This was the approach taken by G. Allport (1950): namely, to find the ‘religious orientation’, the motives behind the actions in order to comprehend the variety of deeds, including the seemingly contradictory prejudice on the part of many Christian believers. Starting from the notions of immature and mature religiousness, he arrived at a typology involving intrinsic and extrinsic religiousness (Allport & Ross, 1967). Extrinsic religiousness was considered to be instrumental in nature, described as immature and utilitarian, whereby a person uses his/her religiousness to achieve extra-religious (psychological and social) ends. In intrinsic religiousness, the motive for religiousness would be autonomous and ‘over-reaching’. Allport summarized this distinction thus: ‘the extrinsically motivated individual uses his religion, whereas the intrinsically motivated lives his’ (Allport & Ross, 1967, p. 434). While extrinsic religious orientation was found to be negatively correlated to mental health, intrinsic religious orientation was generally positively correlated with mental health (Baker & Gorsuch, 1982; Bergin, Masters, & Richards, 1987). partment of Sociology, University of Maribor, Koroska 160, 2000 Maribor, s, Ltd. Received 13 September 2006 Accepted 9 March 2007 Sergej Flere and Miran Lavrič However, additional studies have proven that results were not uniform in confirming that intrinsic religiousness would always be psychologically ‘healthy’. It has been shown that it may also be linked to prejudice (Griffin, Gorsuch, & Davis, 1987; McFarland, 1989), to authoritarianism (Kahoe, 1977), to closed mindedness, and dogmatism and might even be contaminated by social desirability (Batson & Ventis, 1982). Religious orientation has become the dominant paradigm in the study of religious motivation and of the psychological study of religiousness in general. This is so despite the many often-repeated reservations: thus, Hill and Hood (1999) in a review of literature and instruments in the area, write that ’Allport’s uncanny popularity among psychologists of religion may have more to do with his apostolic reputation than with the conceptual soundness of his religious orientation framework or the instruments used to assess it’ (Hill & Hood, 1999, p. 120). The necessity for many improvements to as well as modifications and expansions of the notions and instruments in the area need to be mentioned, starting from Allport’s (1966, p. 6) need to reinterpret the analytical framework of their data findings, because of the appearance of indisicriminate pro-religious orientation, of ‘muddleheads who refuse to conform to our neat religious logic’, thus creating a need to analyze the relationship within an orthogonal framework and not as a continuum. More recently, it has been suggested that the Religious Orientation Scale (ROS) is best described as measuring three factors (Gorsuch & McPherson, 1989; Kirkpatrick, 1989; Leong & Zachar, 1990; Maltby, 1999). While supporting intrinsic orientation (I) as a unified construct, these studies suggested that extrinsic religious orientation consists of two distinct components: a social and a personal one. Social extrinsic orientation (Es) deals with attainment of social benefits, while personal extrinsic orientation (Ep) deals with overcoming and controlling psychological troubles and distress. As religious orientation became the dominant paradigm in the psychological research of religion and as it was based on general psychological theory (Allport, 1950), it was studied in non-Protestant and non-American environments as well: among others, Hovemyr (1998) studied Swedish Protestants; Socha (1999), Brewczynski and MacDonald (2006) studied the Catholic Polish environment; Tapanya, Nicki and Jarusawad (1997) studied Thai Buddhists; Iranian Muslims were studied by Ghorbani, Watson, Ghramaleki, Morris, and Hood (2000) and Watson, Ghorbani, Davison, Bing, Hood and Ghramaleki (2002); Kaldestad (1992) and Kaldestad and Stifoss-Hanssen (1993) studied Norwegian Protestants; Elbedour, ten Bensel and Maruyama (1993) studied the Palestinians. In an important paper, Cohen and associates contest that intrinsic orientation is the only authentic motivation across all religious traditions, suggesting particularly that social extrinsic orientation, including, that is, many types of ecclesiastical and community religiousness, may be an even more theologically authentic orientation in some Abrahamic religions (Cohen, Hall, Koenig, & Meador, 2005). Nevertheless, they do not address the issue of personal intrinsic orientation. While the cross-cultural validity of the distinction between personal and social motivation in religion does not seem to be disputable, the question of dividing personal orientation into intrinsic orientation and personal extrinsicness could prove more challenging. The issue of intrinsic–extrinsic (I-E) dimensionality has already been raised by Kirkpatrick and Hood (1990), who had noticed the intriguing results from a Gallup survey in which single-item measures of I, Ep, and Es were included. In these data, Ep and Es were mutually uncorrelated (r1⁄4 .01), whereas I and Ep were significantly positively correlated (r1⁄4 .40, p< .01). They also report finding a similar pattern in some of their own unpublished data. These findings are recognized by them as extremely problematic: ‘. . . both for the prevailing two-factor theory (according to which Ep and Es should be highly inter-correlated, as they together comprise a unitary factor) and for Allport and Ross (1967, p. 453) original bipolar conceptualization (according to which I and Ep should be negatively correlated).’ A closer look at recent studies dealing with the three-factor structure of the ROS (e.g., Brewczynski & MacDonald, 2006; Genia, 1993; Khan, Watson, & Habib, 2005; Maltby, 1999; Maltby, 2002; Watson Copyright # 2007 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. (in press)