Premium
Use of T scores on the Revised Child Behavior Profile with inpatients
Author(s) -
Harris F. N.,
Jenson William R.,
McMahon William M.
Publication year - 1987
Publication title -
journal of clinical psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.124
H-Index - 119
eISSN - 1097-4679
pISSN - 0021-9762
DOI - 10.1002/1097-4679(198707)43:4<376::aid-jclp2270430406>3.0.co;2-l
Subject(s) - raw score , comparability , psychology , standard deviation , standard score , scale (ratio) , statistics , clinical psychology , contrast (vision) , psychometrics , raw data , mathematics , artificial intelligence , cartography , geography , combinatorics , computer science
The assignment of T values to raw scores on the behavior problem scales of the Revised Child Behavior Profile (RCBP) was examined. RCBP T scores below 70T are normalized standard scores. Therefore, the T scores below 70T are comparable within each scale and from one scale to another. In contrast, RCBP T scores above 70T are not standard scores. Therefore, they cannot be compared appropriately to other scores. The lack of intrascale and interscale comparability of RCBP T scores above 70 is especially problematic for the study of inpatients whose scores tend to fall in the unscaled region above 70T. Therefore, it is recommended that for studies of inpatients, raw scores on the behavior problem scales should be transformed to standard scores by use of the mean and standard deviation of an appropriate comparison group. Then, for purposes of interpretation, scores that fall in the upper ranges of the inventory can be compared appropriately to other scores on the same scale or to scores on other scales of the RCBP.