This paper investigates whether there is an endogeneity between companies which receive non-audit services and other companies which do not receive non-audit services from their auditors. After controlling the endogeneity, this paper also examines the relationship between non-audit services and audit quality, audit fee, and audit hours using pre-post comparative approach. In addition, to test empirical results, two statistical methods, propensity score matching model and Heckman selection model were used to control the endogeneity in different way. The main findings are as follows; First, when comparing absolute value of discretionary accruals, real activities manipulation, audit fee, and audit hours between companies which receive non-audit services and other companies which do not receive non-audit services, it was found that the endogeneity makes significant differences. This finding suggests that results from previous studies might be biased because of the endogeneity. Second, when comparing two periods of the same companies using pre-post comparative approach, the period that companies receive non-audit services and the previous period that the companies did not receive non-audit services, no significant differences were found in audit fee and audit hours between the two periods. However, with a propensity score matching model and Heckman selection model, there were significant differences, showing that companies which receive non-audit services have larger audit fee and higher audit hours. Therefore, it is critical to pay attention to control the endogeneity when examining empirical research. Our findings provide an important implications to not only academic researchers but also auditors, investors, practitioners, and regulators who are interested in audit quality and Non-audit Services. Also, academics can also apply the discussion in this paper for related researches.