Personality, Christian belief and paranormal belief 1 Running head: Personality, Christian belief and paranormal belief Personality, conventional Christian belief and unconventional paranormal belief: A study among teenagers

A sample of 10,851 pupils (5,493 males and 5,358 females) attending year-nine classes (13to 14-year-olds) and a sample of 9,494 pupils (4,787 males and 4,707 females) attending yearten classes (14to 15-year-olds) in non-denominational state-maintained secondary schools in England and Wales completed questions concerned with conventional Christian belief and unconventional paranormal belief, alongside the short-form Junior Eysenck Personality Questionnaire. The data demonstrated that conventional Christian belief and unconventional paranormal belief occupy different locations in relation to the Eysenckian model of personality in respect of the psychoticism scale and the lie scale. While conventional Christian belief is associated with lower psychoticism scores and higher lie scale scores (greater social conformity), unconventional paranormal belief is associated with higher psychoticism scores and lower lie scale scores (lower social conformity). Personality, Christian belief and paranormal belief 3 Personality, conventional Christian belief and unconventional paranormal belief: A study among teenagers Eysenck’s dimensional model of personality proposes that individual differences in personality can be most economically and adequately summarised in terms of three orthogonal higher order factors (Eysenck & Eysenck, 1975, 1991). The first factor is expressed on the continuum from introversion, through ambiversion, to extraversion. Those who score high on the extraversion scale can be characterised as sociable, lively, active, carefree, dominant and assertive. The second factor is expressed on the continuum from emotional stability, through emotional lability, to neurotic disorder. Those who score high on the neuroticism scale can be characterised as anxious, depressed, tense, emotional, irrational and often have low self-esteem. The third factor is expressed on the continuum from tendermindedness, through toughmindedness, to psychotic disorder. Those who score high on the psychoticism scale can be characterised as cold, aggressive, toughminded, antisocial and impersonal. In order to guard against dissimilation, the Eysenckian family of personality measures also generally includes what has been defined (somewhat unfortunately) as a lie scale. Those who score high on this scale can often be characterised as displaying high levels of social conformity rather than intentional or unintentional dissimilation. The relationship between personality and individual differences in conventional religiosity has been of longstanding interest within the psychology of religion. Reviewing empirical evidence on the relationship in the mid-1970s, Argyle and Beit-Hallahmi (1975) concluded that the findings were insufficient to support any consistent or stable linkage between these constructs. Reviewing the empirical evidence two decades later, Beit-Hallahmi and Argyle (1997) revised their conclusion in light of the way in which a series of independent studies had demonstrated an inverse relationship between indices of conventional Personality, Christian belief and paranormal belief 4 religiosity and Eysenck’s dimension of psychoticism. A somewhat closer reading of the evidence would have convinced these reviewers of a second stable and consistent relationship, namely a positive relationship between indices of conventional religiosity and the Eysenckian lie scale. Evidence for these conclusions is provided, for example, in a series of studies reported across a range of samples in Australia and Canada (Francis, Lewis, Brown, Philipchalk, & Lester, 1995), France (Lewis & Francis, 2000), Germany (Francis & Kwiran, 1999), Greece (Youtika, Joseph, & Diduca, 1999), Hong Kong (Francis, Lewis, & Ng, 2003), Northern Ireland (Lewis, 1999, 2000, 2001; Lewis & Joseph, 1994), Republic of Ireland (Maltby, 1997; Maltby & Lewis, 1997), South Africa (Francis & Kerr, 2003), the United Kingdom (Bourke & Francis, 2000; Carter, Kay, & Francis, 1996; Francis, 1991, 1992, 1999; Francis & Bennett, 1992), and the United States of America (Lewis & Maltby, 1995; Roman & Lester, 1999). At the same time these studies generally failed to find any significant correlations between indices of conventional religiosity and either neuroticism scores or extraversion scores. The psychological mechanisms underpinning these two stable findings have been variously expressed. Two accounts have been advanced to explain the inverse relationship between indices of conventional religiosity and psychoticism scores. The first account, as rehearsed by Francis (1992), builds on Eysenck’s classic theory of social learning according to which low psychoticism scores are associated with greater conditioning into tenderminded social attitudes (Frcka & Martin, 1987) and according to which conventional religiosity belongs to the domain of tenderminded social attitudes (Eysenck, 1975, 1976). The second account, as rehearsed by Francis (2005), builds on the association between high levels of conventional religiosity and femininity (Thompson, 1991) and the association between femininity and low psychoticism scores (Eysenck & Eysenck, 1976). Three accounts have been advanced to explain the positive relationship between indices of conventional religiosity Personality, Christian belief and paranormal belief 5 and lie scale scores, all three of which were well rehearsed by Pearson and Francis (1989). The first account accepts the social conformity interpretation of lie scale scores (Massey, 1980) and argues that conventional religiosity is associated with higher levels of social conformity. The second account accepts the immaturity interpretation of the lie scale scores (Crookes & Buckley, 1976) and argues that conventional religiosity is associated with immaturity. The third account accepts the paradoxical interpretation that high lie scale scores reflect the truthful account of high ethical standards (Loo, 1980) and argues that conventional religiosity is associated with truthfulness. Although as yet much less well developed, a second research tradition has begun to map the association between Eysenck’s dimensional model of personality and paranormal beliefs. A very different pattern of relationship emerges from these studies. In a study among 51 undergraduates (mean age, 20.7 years) in the United States of America, Lester (1993) found no significant correlations between either lie scale scores or psychoticism scores and belief in astrology, faith healing, mediums, or extra-sensory perception. Similarly, in a study among 94 adolescents (mean age, 15.9 years) in the United States of America, Willging and Lester (1997) found no significant correlations between psychoticism scores and a measure concerned with belief in psychic phenomena, but this study failed to report on the lie scale. In a third study concerning paranormal beliefs and personality among black South Africans, drawing on 260 grade-eleven secondary school students (mean age, 19.2 years) and 244 firstyear social science university students (mean age, 21.9 years), Peltzer (2002) found no significant association between either lie scale scores or psychoticism scores and scores recorded on the full 25-item Paranormal Belief Scale proposed by Tobacyk and Milford (1983). In a fourth study among 279 13to 16-year-old pupils in South Wales, Williams, Francis, and Robbins (2007) found no significant correlations between either lie scale scores or psychoticism scores and a six-item measure of paranormal belief. Personality, Christian belief and paranormal belief 6 These four primary studies reported by Lester (1993), Willging and Lester (1997), Peltzer (2002), and Williams, Francis and Robbins (2007) are, nonetheless, limited by three factors: the sample sizes are relatively small; while concentrating on paranormal belief, no comparison is offered with conventional religiosity; and paranormal belief was operationalised in ways that overlap with conventional religiosity. The aim of the present study is to extend the previous research in three ways: by drawing on a much larger sample of young people; by including two discrete indicators of unconventional paranormal belief; and by including in the same study two discrete indicators of conventional Christian belief. The clear contrast between conventional Christian belief and unconventional paranormal belief removes overlap between the two constructs as represented, for example, by the notion of life after death. Method Sample The Religion and Values Today Survey, described in detail by Francis (2001), was completed by 33,982 pupils attending year-nine and year-ten classes throughout England and Wales. This database was constructed to be thoroughly representative of young people in this age group (13to 15-year-olds) being educated within both the state-maintained schools sector and the independent schools sector. Data were provided from 163 schools, stretching from Pembrokeshire in the west to Norfolk in the east, from Cornwall in the south to Northumberland in the north. A proper mix of rural and urban schools was included, as was a proper mix of independent and state-maintained schools. Procedure Participating schools were asked to follow a standard procedure. The questionnaires Personality, Christian belief and paranormal belief 7 were administered in normal class groups to all year-nine and year-ten pupils throughout the school. Pupils were asked not to write their name on the booklet and to complete the inventory under examination-like conditions. Although pupils were given the choice not to participate, very few decided not to take part in the survey. They were assured of confidentiality and anonymity. They were informed that their responses would not be read by anyone in the school, and that the questionnaires would be despatched to the University of