Premium
Problems with change in R 2 as applied to theory of reasoned action research
Author(s) -
Trafimow David
Publication year - 2004
Publication title -
british journal of social psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.855
H-Index - 98
eISSN - 2044-8309
pISSN - 0144-6665
DOI - 10.1348/0144666042565344
Subject(s) - theory of reasoned action , variance (accounting) , variables , psychology , variable (mathematics) , argument (complex analysis) , contrast (vision) , social psychology , action (physics) , regression analysis , field (mathematics) , epistemology , econometrics , statistics , mathematics , computer science , philosophy , business , mathematical analysis , biochemistry , chemistry , physics , accounting , quantum mechanics , artificial intelligence , pure mathematics
The paradigm of choice for theory of reasoned action research seems to depend largely on the notion of change in variance accounted for (ΔR 2 ) as new independent variables are added to a multiple regression equation. If adding a particular independent variable of interest increases the variance in the dependent variable that can be accounted for by the list of independent variables, then the research is deemed to be ‘successful’, and the researcher is considered to have made a convincing argument about the importance of the new variable. In contrast to this trend, I present arguments that suggest serious problems with the paradigm, and conclude that studies on attitude‐behaviour relations would advance the field of psychology to a far greater extent if researchers abandoned it.