Premium
Clinical versus statistical assessment of psychotherapy: A mathematical model of the dilemma
Author(s) -
Brownsberger Carl N.
Publication year - 1965
Publication title -
behavioral science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.371
H-Index - 45
eISSN - 1099-1743
pISSN - 0005-7940
DOI - 10.1002/bs.3830100405
Subject(s) - dilemma , psychology , psychotherapist , reliability (semiconductor) , statistical analysis , perception , statistical thinking , statistical evidence , statistical model , clinical practice , computer science , management science , artificial intelligence , epistemology , statistics , mathematics , medicine , mathematics education , engineering , philosophy , power (physics) , physics , null hypothesis , family medicine , quantum mechanics , neuroscience
In principle any therapeutic technique can and should be subjected to a formal statistical trial. Those patients helped and those patients harmed should be counted. Only if the first number significantly exceeds the second should the treatment be retained. In practice, remarkable difficulties supervene. Abundant consideration has been given in the past to the problems of defining psychotherapy, of defining improvement, and of achieving reliability among judges to identify improvement. The abstracting effect of the common statistical techniques, however, has been less commonly recognized as a problem. Presented here is a mathematical model of a controlled statistical trial of psychotherapy. It illustrates how accurate clinical perceptions about individuals and valid statistics about groups may provide apparently conflicting evidence regarding the effectiveness of psychotherapy.