The Big Five and the Big Six
These are factors of personality and social behaviour that may be considered to give a ‘personality portrait’ of an individual. They have emerged over years of research to be relatively stable and somewhat independent of each other (Digman, 1990).
The Big Five are openness, conscientiousness, extraversion, agreeableness and neuroticism (acronym OCEAN). See Goldberg (1990) for a description of these factors and how they were derived.
The Big Six (HEXACO) were developed following more extensive international testing using a wider range of vocabulary (Ashton et al., 2004). The Big Six factors each have four facets: ‘honesty-humility’ (H): sincerity, fairness, greed avoidance, modesty; ‘emotionality’ (E): fearfulness, anxiety, dependence, sentimentality; ‘extraversion’ (X): social self-esteem, social boldness, sociability, liveliness; ‘agreeableness’ (A): forgivingness, gentleness, flexibility, patience; ‘conscientiousness’ (C): organisation, diligence, perfectionism, prudence; ‘openness to experience’ (O): aesthetic appreciation, inquisitiveness, creativity, unconventionality (acronym HEXACO).
Although there is some overlap with the names of the Big Five, HEXACO’s ‘emotionality’ is not identical to the Big Six’s ‘neuroticism’, the factor ‘agreeableness’ is not identical in the two sets, and the factor ‘honestyhumility’ is not separately or totally covered in the Big Five. However, ‘extraversion’, ‘conscientiousness’ and ‘openness’ correspond in the two sets of measurements. Researchers will opt for the Big Five or the Big Six depending on their needs for the assessment of emotionality/neuroticism, agreeableness and honesty/humility.
Measures of social attitudes and cognitions
Authoritarianism: California F (Adorno, Frenkel-Brunswick, Levinson, & Sanford, 1950): This is the justly famous traditional measure of authoritarianism. The ‘F’ stands for ‘Potentiality for Fascism’. It is one of several scales developed in the late 1940s by a team of social scientists in the United States, refugees from Nazi Germany. The F-Scale is as much a measure of beliefs, personality, or both as it is of attitudes. It is supposed to assess such traits as concern with power in social relationships, antiintraception and superstition, and it relates well to ‘purer’ measures of social attitudes assessing anti-Semitism, ethnocentrism (belief in the superiority of one’s own cultural group) and political and economic conservatism (anti-welfare and anti-equal rights). The F-Scale has attracted much criticism, including concern about its applicability to the study of left-wing authoritarianism, and concern about the fact that none of the items are ‘negatively worded (reverse-meaning)’. The scale is hardly a contemporary one, but most of the items still sound relevant and intelligible.
Wilson-Patterson C (Conservatism) Scale (Wilson & Patterson, 1968): This is a more modern measure of conservatism in social attitudes and beliefs than the California F Scale. Some of the items may be found puzzling or dated by you or by testees, such as fluoridation, pyjama parties, birching and learning Latin, but there is no more modern measure. This measure has good psychometric properties, it is delightfully easy to administer and to score, and in spite of the momentary bewilderment that may be caused by some items, people appear to enjoy completing this scale, which is quick and quite interesting. Psychometric properties are good. Even the bewildering items cause amusement and interest, rather than the anger or fear that may be provoked by some items on some scales.
Burns Life Styles for Women (Burns, 1974): This measure assesses the extent to which the person feels that it is appropriate for women to pursue a career regardless of other commitments, or whether women should prioritise marriage, homemaking and childcare. It was developed at a time when ‘feminism’ was a hot topic of debate, and it still taps concerns that are important to many women, especially the more highly educated. The scale would not be appropriate for use in less-privileged groups. The scale is not appropriate for use on men either, although the author suggests that it can be administered to men by asking them to say how their ideal woman should think. The men tested by Burns certainly had a more ‘traditional’ view of women’s lifestyles than did the women, but this may have been because of the way men were asked to complete the scale. It is quick to do, easy to administer (when testing those for whom it is appropriate, i.e. middle-class women), easy to score, quite interesting and apparently not too threatening. Reliability is good. The Bern Sex Role Inventory (Bern, 1974) assesses psychological masculinity, femininity and androgyny. People are asked to say whether they feel stereotypically masculine or feminine personality characteristics (adjectives) apply to them. The test has good psychometric properties and has been very popular.
Religion: A number of aspects of religious belief, experience and feeling have been assessed. Reported reliabilities are sometimes a bit low. The most useful measures, in terms of their relations to other factors, are probably as follows:
The Religious Life Inventory (Batson, Schoenrade, & Ventis, 1993): This assesses extrinsic, intrinsic and quest religious orientation. This scale is based on Batson’s (1976) factor-analytic work using items developed by earlier workers.
Spiritual Support Efficacy Scale (Maton, 1989): Spiritual support is suggested to be a stress-buffering factor. The scale is delightfully brief but nevertheless has a very high reported reliability, and the validity work is impressive.
The Mysticism Scale (Hood, 1975): This measure has several subscales, which survived factor analysis satisfactorily. Construct validity is satisfactory. The scale measures reported mystical-type experiences.
Religious affiliation, religious practice and attendance and other religious factors: These should be assessed in questionnaire measures developed to suit the groups investigated. It is normally very difficult to use religion measures on groups other than those for which they were developed: for example Moslems, Jews and other non-Christians get very uptight when asked
Finding and using existing tests and scales 39 about church attendance, or even ‘attendance’ at ‘place of worship’ (in some religions, ‘worship’, especially by women, is done mainly at home). Professing agnostics are another group that need special attention when assessing religion. You will therefore probably have to develop your own set of questions, or use a questionnaire developed by someone else who has worked on the same group.
Values: The Allport- Vernon Study of Values (Allport & Vernon, 1960) uses a forced-choice format and assesses people on self-reported interest in several general areas of activity (social, religious, political, etc.), while Rokeach’s (1969) checklist involves more specific values (such as salvation, forgiving) grouped under two general headings (Means, or instrumental, and Ends, or terminal). Both measures seem to be non-threatening, and quite thought provoking. The forced-choice format of the Allport-Vernon does not reveal if a person has a generally high (or generally low) interest in all the areas examined. The measure just reveals comparative levels of self-reported interest in different areas, for example whether the person regards ‘Art’ as more important than, say, their social life. Some participants may find it annoying to have to be forced to choose between two equally desirable or undesirable alternatives. This is a general problem with the forced-choice format whenever it is used to assess opinions, values or other non-factual constructs.
Social support is an important area of investigation. The Quality of Relationships Inventory (QRI) (Pierce, Sarason, & Sarason, 1991) and the Significant Others Scale (SOS) (Power, Champion, & Aris, 1988) are both usable measures, with good psychometric properties. The QRI is available on application to the authors, while the SOS is in the original published article by Power et al (1988). The QRI uses a Likert format and assesses support (emotional), conflict and depth in a given relationship. The SOS assesses emotional and practical support, both perceived and ideal, in a given relationship, and has the advantage of a delightfully brief form (four items) as well as a longer form.
Locus of control (causal expectancy: the extent to which the person feels that events are caused by internal or external factors): Rotter’s (1966) and Levinson’s (1973) generalised scales are both still useful measures of generalised locus of control. Levinson’s measure is probably more useful in that it distinguishes between two types of external locus of control: luck and chance, and powerful others. Some investigators however prefer to use measures that are specific to the area of behaviour under investigation.
Need achievement: In the classical research, achievement-related motives and needs (need for achievement, fear of failure, fear of success) were assessed by projective tests, which often have reliability problems, and which require training. Attempts have been made to develop scales assessingneed achievement, for example Smith (1973). Our experience of using Smith’s scale has been that it is easy to deal with but unfortunately does not give very good discrimination between testees in groups of students (students get a moderately high score). It may give a wider range of scores in the general population. Smith reported a significant difference in scores between volunteer testees and men listed in ‘Who’s Who’. Smith’s reliability analysis shows only item-total correlations - rather low ones - and a rather unimpressive split-half reliability of 0.56. The literature on need achievement has reflected only low associations between need achievement and actual achievement in men, and no consistent associations for women. It is therefore unlikely that anyone would be considering assessing need achievement for assessment purposes when there is no measure with good predictive validity. For research - and assessment - purposes, your best bet might be to develop an achievement-need scale specific to the domain of achievement in which you are interested. Smith’s test could be useful in circumstances where a quick and friendly measure of general need-achievement is needed, provided the success or failure of your study does not hinge solely on this test. Another possibility is the measure of the Protestant Work Ethic (Furnham, 1984).