submitted for publication
In philanthropic research, a much-neglected aspect is the methodology used for analyzing donating behaviour. This is a serious problem, as different methods lead to as many different conclusions. In this paper, we will discuss the limitations of using linear (OLS) regression analyses, as well as discuss the advantages and limitations of the two most often used parametric models for analyzing charitable donations: Tobit and Heckman Two-Stage regression analyses. Our main objectives in this paper are to explain Tobit and Heckman two-stage regression models in a non-technical way and to clarify the consequences for the findings derived with these models.