The use of scores anchored at the scale midpoint in reporting individuals' traits

Standardized scores including factor scores are insufficient indicators of individuals' positions on personality traits when mean scores deviate considerably from the scale origin. Anchored scores, that is, raw scores expressed as deviations from the scale midpoint and divided by the standard deviation, retain this discrepancy. We argue in favour of reporting in terms of anchored scores in addition to standard scores, and present a solution to the specific problem of finding anchored factor (i.e., principal-component) scores. Such scores are obtained through dividing factor weights by the standard deviation of the variable in question, and applying the corrected weights to the matrix of raw scores, expressed as deviations from the scale midpoint (e.g. -2, -1, 0, +1, +2). (C) 1998 John Wiley & Sons, Ltd.