| Titre | L'analyse causale face aux effets d'interaction : réexamen de la méthode de dépendance de Raymond Boudon | |
|---|---|---|
| Auteur | Michel Loriaux | |
 |
Revue | Revue Française de Sociologie |
| Numéro | 1972, 13-4 | |
| Page | 550-563 | |
| Résumé anglais |
Michel Loriaux : Causal analysis versus interaction effects ; reexamination of Raymond Boudon's dependance method.
According to the author, the concept of generalized correlation used by R. Boudon in his dependance analysis to measure variable interaction effects structures in terms of causal influences is hardly justified and the corresponding equations used for dichotomous variables are non-valid. He shows then, that the origin of these errors lies in an inadequate representation of interaction variables and that the basic correlation coefficient is the adequate tool for dépendance analysis even in case of non-additive models. In addition, letters between R. Boudon and the author. Source : Éditeur (via Persée) |
|
| Article en ligne | http://www.persee.fr/web/revues/home/prescript/article/rfsoc_0035-2969_1972_num_13_4_2100 |
