Learning From Divestitures to Complete Acquisitions: A Study Using Secondary Data and Logistic Regression Models to Explore the Determinants of Mergers and Acquisitions Completion Likelihood

This case highlights the major steps of conducting our research, in which we examined whether firms could learn from prior divestitures, that is, selling activities, to complete subsequent acquisitions. We drew arguments from the organizational learning theory to develop our research model and used secondary data to test the proposed hypotheses. As the research topic was quite new and not many studies had explored it, we developed our theoretical model and collected the data at the same time. We wanted to make sure that we would have appropriate data and enough observations for the empirical analysis. In addition, exploring the data from the start would give us some more insights on practical issues of mergers and acquisitions, which are not discussed much in academic papers. The data collection and organization were time-consuming because mergers and acquisitions are complicated activities and each transaction can contain a lot of information. Therefore, to avoid mistakes, we made a data diary, which saved every single step that we took to handle the data, including the reasons why we did so and other remarks. Regarding the empirical analysis, we chose a logit regression model as our dependent variable was binary. This implied that interpretation of the results was not straightforward because coefficients in non-linear models cannot be interpreted in the same way as they are in linear models. We discussed this issue in the “Result Interpretation” section before concluding the case with some practical lessons that we gained from our project.