Problems of meaning in psychiatric screening scales: a critique and possible solutions

It is argued that there are problems in the way that psychiatric screening scales are constructed. The normal procedure is to collect a group of symptom items and sum them creating a single score. While researchers generally assign an equal weight to each symptom, this is not warranted. Using data from two psychiatric screening scales from large representative samples, it is shown that symptoms vary tremendously in the degree to which they are experienced in the population; the more rare symptoms appear to measure depression while the more common symptoms sound much like stress. Summing these items and assigning them equal weights makes little sense, given these differences. Examining case profiles shows that individuals with quite different symptom patterns receive identical scores. Several alternative methods of creating psychiatric screening scales are suggested, including eliminating some items, and devising systems of symptom weights; factor analysis is ruled out, however.