In the last decade, the area of personality measurement has been dominated by three major systems: the Eysenckian Giant Three, the Cattellian sixteen factors and the Big Five. While many of the Cattellian second-stratum factors have been shown to fit the Big Five system, can the factors measured by the Eysenck Personality Questionnaire also be described by the five-factor model? The study reported in this article was designed to determine whether the dimensions measured by a revised Chinese version of the Junior Eysenck Personality Questionnaire would, in a Hong Kong population, replicate the Giant Three or the Big Five and whether there is evidence to support the suggested dual nature of the Extraversion dimension and the Lie scale in this Cantonese-speaking group. A four-factor solution indicated that the data did not support the notion of a Giant Three model plus a Lie scale and lacks clarity. A five-factor solution produced factors that can clearly be labelled Neuroticism, Conscientiousness, Sociability, Excitement-Seeking and Agreeableness. Sociability, focusing on meeting people and Excitement-Seeking, which consists of Impulsivity and Liveliness, derive from items in Eysencks Extraversion dimension. The Openness factor of the Big Five system is absent in this population.
In personality psychology, theorists and especially researchers using personality assessment, encounter difficulties and confusion when they face the existing bewildering array of personality factors or scales. While some factors in different systems have the same name, the concepts measured are often not the same; on the other hand, some factors with different names share the same item content.
Taxonomy is always a contentious issue because the world does not come to us in neat little packages (Gould, 1981).
In the widely-used Occupational Personality Questionnaire (Saville et al, 1984), there are 30 factors, while only three factors are emphasized in the well-established Eysenckian personality questionnaires. Eysenck (1991)suggested some taxonomic paradigms were needed to co-ordinate research work, the three major systems being the Cattellian personality factors, the Big Five of Goldberg, Costa and McCrae and the Eysenckian three-factor system. Later, Eysenck remarked that only two major systems have survived the psychometric holocaust, namely the Giant Three and the Big Five (Eysenck, 1994).
The Cattellian system is a much vaunted one and is still widely used in industrial settings, but its factor structure has proven difficult to replicate (Kline and Barrett, 1983). These authors also concluded from their review that the Eysenckian three-factor model was the most adequate representation of personality structure. However, Kline and Lapham (1991)came to believe that the Big Five provided probably the best account of ratings in personality. In view of the accumulated evidence, McCrae (1992)surmised that many personality psychologists had recently adopted the Big Five model. In the lexical approach using natural language terms, the legacy of Galton, Thurstone and Cattell has crystallized in the hands of contemporary researchers into agreement on the Big Five factors with only Cattell and Eysenck holding out on the precise number (Goldberg, 1993).
Eysenck (1990), Eysenck (1992a)was not persuaded that the five-factor model best represented the basic dimensions of personality and argued that his three-factor model accounts for the more important dimensions of personality. He also stated that there was no homological network or theoretical underpinning for the Big Five. Although Costa and McCrae (1992b)concurred with Eysenck in the need for a paradigm, they reiterated their view that at the developing point of this fruitful science, what was needed was a systematic method of description which must precede, rather than follow, personality theory.
The Eysenckian system as it relates to Hong Kong high school students does not replicate the Giant Three dimensions plus the Lie scale but appears to conform to a five-factor model. Our data, resulting from the administration of a revised Chinese version of the Junior Eysenck Personality Questionnaire (JEPQ), generally support the Big Five factors Neuroticism, Agreeableness and Conscientiousness in common with many other investigators who have used the EPI or EPQ with Western adult subjects. The Extraversion factor is dual in nature, consisting of Sociability and Excitement-Seeking components. Impulsivity does not shift to Psychoticism as claimed by a number of personality theorists, but merges with Excitement-Seeking and Liveliness to form a new factor. The fifth Big Five factor, which was labelled Culture by Norman (1963), as Intellect by Goldberg (1990)and as Openness to Experience by Costa and McCrae (1985),was not generated by our data.
For this population, the so-called Lie scale can be interpreted as measuring a personality trait in its own right—Conscientiousness—rather than being an index for the detection of lying. The fifth factor, loading eight out of the original seventeen P scale items, is labelled as Agreeableness instead of Psychoticism as these eight items are concerned primarily with conformity to social norms vs alienation and cruelty. Since P is the unique opposite pole of this fifth factor and L is found to be the C, it seems that P cannot be a superfactor of the facets of A and C.
There is broad consensus in the findings in Western countries that boys are more excitement-seeking but less agreeable than girls; the same is true in Hong Kong. Moreover, Neuroticism also correlates highly with Conscientiousness and Sociability for this Chinese sample.
