Consensual Validity of the EPQ: Self-Reports and Spouse-Reports

Abstract

One important question in trait theory is the concept of validity, that is, whether a given test actually measures what it is intended to measure. This is easy to establish when we have a criterion against which the test can be evaluated. Unfortunately, the criterion is not always available. One way of overcoming this problem could be to validate questionnaire responses by correlating them with ratings made by external assessors who know the ratee well: This is what is known as consensual validity. This study assesses the consensual validity of the EPQ by means of self-reports and spouse-reports. Self-reports of 198 men and women with a mean age of 35 years were correlated with their spouses' reports. Correlations ranged from .43 to .70 for the three dimensions. A multitrait-multimethod analysis showed evidence of convergent and discriminant validity in the entire sample as well as in both men and women. Two factors probably contributed to the relatively higher correlations: the psychometrical adequacy of the instrument used and the choice of well-qualified raters. The results obtained favour the use of self-reports as valid and useful methods of personality measurement.