consultation with a statistician: looking for a second opinion on a model involving multiple imputation

Looking for a second opinion on a model involving multiple imputation of pre and post survey data. survey questions Likert scale 1-5, asked at baseline, then again at the end of the program. The post data has up to 50% missing responses. I want confirmation that my assumptions are accurate and then accurately report the results. I am using paird-wilcoxin test to compare results pre and post. SPSS doesn't calculate Pooled estimates for non-parametric test on imputation databases.