Trends in research design and data analytic strategies in organizational research

Abstract

We coded all articles that were published in the Journal of Applied Psychology in the 1975-1993 period (N = 1,929) in terms of the type research design used in studies (e.g., experimental, nonexperimental), and the major strategies used in analyzing the data of the studies (e.g., analysis of variance, multiple correlation/regression, covariance structure analysis) and computed annual frequencies of use of the research designs and the data analysis strategies. These frequencies were used to compute yearly percentage use indices (PUIs) for each of the designs and data analytic strategies. Results showed that over the period considered: (I) there were fairly minor fluctuations in the PUIs for experimental and nonexperimental designs; (2) the PUI for covariance structure analysis procedures (e.g., LISREL, EQS) was very low during the 1975-1984 period but has increased dramatically thereafter; (3) the PUIs for multiple regression and several other multivariate data analytic strategies have increased somewhat over time; and (4) the PUIs for data analytic procedures that involve tests of mean differences have fluctuated over time, but there is a slight decrease in the use of ANOVA. Implications of these trends are considered.