In this chapter, two important pitfalls in (multiple) regression analysis are discussed. Collinearity occurs when two independent variables are highly correlated with each other. This can lead to huge standard errors and sometimes to a so-called sign switch. Mediation analysis is an important tool to investigate causal mechanisms. Within mediation analysis, a distinction is made between the total effect, the direct effect and the indirect effect. The problem with mediation analysis is that it is basically the same as the analysis of confounding. The only difference is the interpretation of the results of the analysis. It is also shown, that there is a pitfall within the pitfall of mediation analysis, i.e. the estimation of the standard error of the indirect effect.