Personality in Proportion: A Bipolar Proportional Scale for Personality Assessments and Its Consequences for Trait Structure

Abstract

Trait structures resulting from personality assessments on Likert scales are affected by the additive and multiplicative transformations implied in interval scaling and correlational analysis. The effect comes into view on selecting a plausible alternative scale. To this end, we propose a bipolar bounded scale ranging from -1 to +1 representing an underlying process in which the assessor would review and discount positive and negative behavioral instances of a trait. As an appropriate index of likeness between variables X and Y, we propose LXY = SigmaXY/N, the average of the raw scores cross products. Using this index, we carried out a raw scores principal component analysis on data consisting of 133 participants who had each been rated by 5 assessors including self on 914 items. Contrary to the Big-Five structure that was found in these data on standard analysis, the results showed a relatively large first principal component F1 and 2 very small ones, F2 and F3. The sizes LFF=SigmaF2/N, the averages of the squared component scores, were modest to small. It thus appears that the scale, bipolar proportional versus standard, has a profound impact on the size and structure of personality assessments. The dissimilarity remains on analyzing self-ratings rather than averaged (over the 5 assessors) ratings.

Personality and the Prediction of Consequential Outcomes

“…But these topics do not directly address what we understand to be the source of abiding interest in individual differences in character and temperament since antiquity. Personality matters, not just in ways that interest the differential psychologist or those attached to a "romantic" conception of human nature (Hofstee & Ten Berge 2004), but also in ways that matter to most people and policy makers.…”

Promoting intersubjectivity: a recursive-betting model of evaluative judgements

“…For the validating and weighting of an individual judgement against the panel or committee average, an appropriate coefficient of correspondence is needed. The simplest index of the correspondence between [−1…+1] variables is their mean product, named Likeness coefficient L by Hofstee and Ten Berge (2004); other coefficients of association for scales with a fixed midpoint take the same shape for the [+1, −1] scale (see Appendix 1). In the single case, L is simply the product of the two values; here, an individual judgement and the committee average.…”

“…With more cases taken together, the iterative procedure would lead to taking the first principal component of the raw-scores matrix, with the sum of the absolute component weights set at unity (see, Hofstee & Ten Berge, 2004). However, for reasons of simplicity and because judgements may not form a matrix (as different sets of judges may operate on different cases), one may wish to profit from the fact that L and related coefficients for scales with a fixed midpoint are defined in the single case.…”

Structuring Assessments of Psychopathology

“…

Hofstee and Ten Berge (2004a) have proposed a new look at personality assessment data, based on a bipolar proportional (-1, . .

…”

“…Details are given in the Appendix. Hofstee and Ten Berge (2004a) applied this procedure to data provided by Hendriks (1997), consisting of ratings of 133 target persons on 915 items used in the construction of the Five-Factor Personality Inventory (FFPI). The means of such items tend to deviate considerably from the midpoint of the scale, in the direction of the socially desirable pole.…”