This chapter starts with a discussion of empirical testing based on structural versus reduced models in quantitative studies. Structural models consist of formulas that represent the relation of every dependent variable to its independent variables on various levels, whereas reduced models exhibit the net or overall relation between the dependent variable and the ultimate independent variables. Many quantitative studies published in management journal, especially those that use archival databases, belong to the reduced model category and thus seldom directly test the mechanisms in question. Another popular practice by quantitative researchers is post hoc hypothesis development where they develop hypotheses after they have obtained the results of data analysis. In the process, they may fudge their arguments to fit the results. A replication avoids all the shortcomings of post hoc hypothesis development because its hypotheses, which are the hypotheses of the original study, pre-exist data collection and analysis. Moreover, a replication helps to identify errors in the original study. A multi-method approach enables researchers to study a phenomenon more rigorously and may reveal unanticipated phenomena.