Types and variables: towards a congenial procedure for handling personality data
Standard procedures for processing and interpreting data in personality assessment run the risk of losing their audience. Most notably, relative scaling of data, whether through interindividual or intra‐individual comparison, leads to losing either the persons or the variables from view. I set out an alternative, more congenial procedure for handling personality data, consisting of (i) translating assessments to a bipolar bounded scale running from − 1 to + 1, (ii) adopting the uncorrected average cross‐product (ACP) as the index of association or correspondence between variables and between individuals, and (iii) applying raw‐scores principal component analysis to find factors and types. The ACP index appears eminently fit for handling individual (N = 1) cases. Adoption of the congenial procedure would imply a substantive correction of one's views of individual differences in personality and their structure. Copyright © 2002 John Wiley & Sons, Ltd.