5. Concluding discussion
In this study, a randomly drawn sample that represents all Swedish corporations is used as a database. Hence, the overwhelming majority of firms are non-listed. The study is unique in making a distinction between family and non-family firms in analyzing the efficiency of female leadership. This distinction is motivated by the presumption that the conditions for female leadership differ between these two types of firms. In leadership, we include both ownership and control. Stronger leadership potential is obtained if there is no separation between ownership and control. By efficiency, we mean performance in the form of profitability. Ideally, we would have liked a performance measure that discounted expected future profitability. Such measures include Tobin’s Q, as used by Terjesen et al. (2016), and marginal Q, as presented by Bjuggren (2016). However, because of the nature of our sample, i.e. the inclusion of private non-listed firms, no company stock data could be obtained; hence, no marketbased performance measures like these two could be used. Instead, an accountingbased measure of firm performance (the net profit margin) was used in the econometric analysis.