Pergamon Person. indicid. Oif/. Vol. 23. No. 4, pp. 631-637. 1997 mC* 1997 Elsevier Science Ltd. All rights reserved Printed in Great Britain PII: SO191-8869(97)00082-2 019l-8869/97 $17.00+0.00 DIMENSIONAL ANALYSIS OF THE EPQ-R LIE SCALE WITH A SPANISH SAMPLE: GENDER DIFFERENCES AND RELATIONS TO N, E, AND P Pere J. Ferrando,* Eliseo Chico and Urban0 Lorenzo Department of Psychology, ‘Rovira i Virgili’ University, Carretera de Valls. s/n. 43007 Tarragona. Spain zyxwvutsrqponmlk (Receioed 23 Norember 1996) Summary-A Spanish translation of the 21-item Lie scale included in the EPQ-R was administered to a group of 2026 university undergraduates. The plausibility of the bifactorial structure proposed by Pearson and Francis (Pearson and Francis, Personality and Individual Differences, 10. 1989, 1041-1048) instead of the originally assumed unifactorial structure was assessed using a specific factor analytic model for binary data. Factor structures, as well as total scale scores, were compared for males and females. Relations with N, E, and P dimensions were studied in the male, female and total samples. Results suggest that the unifactorial structure seems more appropriate and that this structure is essentially invariant across gender. 0 1997 Elsevier Science Ltd INTRODUCTION A number of studies carried out by the same research group, (Francis, 1991; Pearson & Francis, 1989; Francis, Brown & Pearson, 1991; Francis, Philipchalk & Pearson, 1991) have questioned the unitary nature of the lie scales belonging to the Eysenck questionnaires. In an initial research project in which 53 items from the lie scales of the J-EPI, J-EPQ and EPQ were jointly analyzed by means of item and exploratory factor analyses, two components, labelled A and B, were identified (Pearson & Francis, 1989). Item content analysis suggested that component A was mainly made up of items concerned more with the image of the well behaved, socially conforming individual, while items in component B were concerned with desirable but unlikely behaviours (or vice versa). So, it was hypothesised that responses to items in Component A might reflect, at least in part, the genuine social practices of some sorts of Ss, while responses to Component B items would be a purer measure of lying or of faking well. Further research (Francis, 1991; Francis et al., 199 1 a, 1991 b), showed certain differential relation- ships when the total scores in components A and B were used. Briefly, component A was found to correlate negatively with Extraversion and Psychoticism, while component B did not correlate at all with E and slightly negatively with P; finally, both A and B correlate negatively and moderately with N. As far as gender differences are concerned, women scored more highly on component A, but no gender differences were found for component B. Detailed information about the construction and nature of the EPQ Lie scale is reviewed in Eysenck and Eysenck (1976) in particular Chapter 11). In short, they believe that, when the motivating conditions are roughly equalized, the scale measures a single consistent personality trait of ‘social conformity’ (a term suggested by Furnham, 1986); on the other hand, in conditions such as selection, application for jobs, etc., where there is external stress for people to appear more conformist than they really are, the scale scores can reflect not only this trait, but also dissimulation (Eysenck & Eysenck, 1976). It also seems that the ways in which both variables might influence the Lie scale scores under different conditions of external pressure are, possibly, quite complex (see e.g. Furnham, 1986; Cowles, Darling & Skanes, 1992; Elliott, Lawty-Jones & Jackson, 1996) A review of the referenced works by Francis and associates, suggests that, in all cases, the tests were administered in neutral conditions, without any kind of pressure which may have encouraged dissimulation (we don’t know, however if they were administered voluntarily). Although Cowles zyxwvutsrqponm et *To whom all correspondence should be addressed. E-mail: pjfp(a astor.urv.es. 631