Additional file 1 of The virtue of optimistic realism - expectation fulfillment predicts patient-rated global effectiveness of total hip arthroplasty

Additional file 1: Supplementary Figure 1. Missing value analysis of the primary outcome measure and the 14 predictor variables included sequentially into the multiple regression analysis. A. The table displays the descriptives (mean and standard deviation), the percentage of missing values per variable, the number of imputed values (missing values x number of imputations) and the type of model used for multiple imputation. For the dependent variable “global effectiveness of total hip arthroplasty” the data set was complete. More than 10% missing values were identified for the rating instrument measuring hip function and mobility (WOMAC), its symptom change score (quotient of postoperative and preoperative WOMAC score) and the calculated expectations-actuality discrepancy scores (fulfillment of expectations) of physical exercise and social interactions. B. The variable chart shows that for 93% of the predictors included at step 4 of the regression model at least 1 value is missing. The cases chart shows that 40% (N = 36) of the study participants has at least one missing value on a variable. The values chart shows that 6% of the 1260 values (cases × variables) are missing. C. The bar graph displays the percentage of study participants having none to six missing predictors. Subjects with more than three missing predictors were excluded from the analysis on the imputed data set.