Research on entrepreneurship is growing as the field is coming of age. This is evident in the number of publications and journals dedicated to entrepreneurship. A large part of the entrepreneurship research is based on studies based on nonexperimental research design (Schjoedt & Bird, 2014), which is not unlike the management literature (Atinc et al., 2012). Because nonexperimental research does not benefit from control of variables as in experiments (Schwab, 2005), many such studies include statistical control of nuisance variables. These are typically referred to as control variables. Control variables which are in effect independent variables, generally only receive limited attention from researchers.

Many researchers include control variables without much consideration of how the control variables may affect the principal relationship under investigation. This may be because these particular control variables were included in other studies or the author’s assumption that “good studies” look like. This blind inclusion of control variables constitutes a methodological “urban legend” according to Spector and Brannick (2011). There are many reasons for the inclusion of control variables; however, their use is often less fruitful in generating valid research findings (Stone-Romero, 2006). Because of this potential for limiting the validity of research findings, we need to examine the use of control variables in published research. Thus, the purpose of this study is to assess the use and misuse of control variables and provide recommendations for future use of control variables in entrepreneurship research.