Titre :
|
Methodological problems in the retrospective computation of responsiveness to change : The lesson of cronbach. (1997)
|
Auteurs :
|
G.R. NORMAN ;
G. REGEHR ;
P. STRATFORD ;
Department of Clinical Epidemiology and Biostatistics. McMaster University. Hamilton Ontario. CAN
|
Type de document :
|
Article
|
Dans :
|
Journal of clinical epidemiology (vol. 50, n° 8, 1997)
|
Pagination :
|
869-879
|
Langues:
|
Anglais
|
Mots-clés :
|
Evaluation
;
Epidémiologie
;
Méthodologie
;
Homme
|
Résumé :
|
[BDSP. Notice produite par INIST R0xGRTzW. Diffusion soumise à autorisation]. Objective : To examine the relation between responsiveness coefficients derived directly from a calculation of average change resulting from a treatment intervention (Responsiveness-Treatment or RT) and those derived from retrospective analysis of changed and unchanged groups (Responsiveness Retrospective or RR) based on a global measure of change. Method : Two approaches were used. First, we used simulation methods to examine the analytical relationship between the RT and RR coefficients. We then located eight studies where it was possible to compute both RT and RR coefficients. As anticipated from theoretical arguments, the RR coefficients were larger than the RT coefficients (1.50 versus 0.41, p<. within study there was no predictable relationship between the two indices. across studies magnitude of rr coefficient strongly related to correlation with retrospective global scale and unrelated rt coefficient. simulated curves fit well observed data substantiated observation that relation coefficients is complex only weakly size treatment effect. conclusion : methods computing responsiveness yield little information about ability an instrument detect effects should not be used as a basis for choice applications clinical trials.>
|