The construct of affect intensity is widely used as a measure of the strength with which individuals experience emotion. This construct is inferred from the presence of a large positive correlation between affect intensity measures computed from scales measuring positive and negative affect when these are administered over several days. A serious statistical artifact is reported in the procedure used to compute affect intensity. This is believed to comPletely invalidate the affect intensity construct. A Monte Carlo study reveals a large correlation between positive and negative affect intensity from data which vary randomly from day to day. Several features of the score distribution are shown to affect the magnitude of these correlations, and it is proposed that affect intensity can be more parsimoniously explained as a mixture of trait neuroticism ana extraversion. The affect intensity literature is shown to be consistent with such an explanation.
|Journal||Personality and Individual Differences|
|Publication status||Published - Jan 1993|