Statistical guidelines for studies of the theory of reasoned action and the theory of planned behaviour

We identify potential problems in the statistical analysis of social cognit ion model data, with special emphasis on the theories of reasoned action (T RA) and planned behaviour (TPB). Some statistical guidelines are presented for empirical studies of the TRA and the TPB based upon multiple linear reg ression and structural equation modelling (SEM). If the model is tested usi ng multiple regression, the assumptions of this technique must be considere d and variables transformed if necessary. Adjusted R-2 (not R-2) should be used as a measure of explained variance and semipartial correlations are us eful in assessing each component's unique contribution to explained varianc e. R-2 is not an indicator of model adequacy and residuals should be examin ed. Expectancy-value variables that are the product of expectancy and value measures represent the interaction term in a multiple regression and shoul d not be used. SEM approaches make explicit the assumptions of unidimension ality of constructs in the TRA/TPB, assumptions that might usefully be chal lenged by competing models with multidimensional constructs. Finally, stati stical power and sample size should be considered for both approaches. Inat tention to any of these aspects of analysis threatens the validity of TRA/T PB research.