Comments on Statistical Issues in November 2015

purpose of this section is to show how many observations are actually analyzed in multivariate analyses, such as multiple linear regression analysis or multiple logistic regression analysis, when there are different numbers of missing values in each explanatory variable. Let’s perform a multiple linear regression using the following hypothetical data (Table 1). In this data, explanatory variable x1 has four, x2 has two, and x3 has no missing values, respectively (denoted as a dot), and we will perform three analyzing processes using SPSS, 1) Pearson correlation analysis, 2) multiple linear regression analysis, and 3) stepwise multiple linear regression analysis.