Abstract
Two H-technique cumulative scales (based on responses of alcoholics to the Jellinek drinking history questionnaire), one of preoccupation with alcohol and one of psychological involvement, has been shown by previous research to be reliable and accurate on the basis of a test using 189 respondents and a retest using 80 respondents. The present study was concerned with the time relationships of the scale items. If alcoholics agree that the items are ordered in time in the same manner as on the scales, the earlier assumption that they measure the process of becoming an alcoholic can be verified. If alcoholics agree that the items are ordered in time, but in a manner different from the scale order, there would be evidence to support the alternate hypothesis that the scales differentiate between types of alcohol symptom constellations. The majority of the data supported the progression hypothesis. Alcoholics agree in statistically significant proportions on the order in which symptom clusters begin and are completed. While the appearance of symptoms is not in the scale order when time is taken into consideration, the completion of symptom clusters is. In addition, the median ranks of symptom completion are as expected. Alcoholics of scale type 1 on both scales tend to have the longest symptom span, of scale type 2 the next longest, and of scale type 3 the briefest. The differences in the length of symptom spans between scale types 1, 2 and 3 are statistically significant for the scale of preoccupation with alcohol and of borderline statistical significance for the scale of psychological involvement. Although the hypothesis of progression appears to be tenable, there is a need for further research on this subject. Possible uses of scales which establish time relationships between symptoms are discussed.