. Q. I have collected demographic and clinical data for patients with endocarditis but due to problems with patient files, there are a number of missing values. Where may I find advice on efficient procedures for exploring missing data in SPSS?
A. Details of how to perform the relevant procedures in SPSS may be found in the handbook IBM SPSS 22 Missing Values. (Note that if the hperlink does not appear to respond on clicking, right-click and choose a suitable option, such as Open in Secure Browser.) I would particularly recommend considering the exploratory approaches on p. 1 – first half of p. 6. Taking on the demands of learning multiple imputation on top of standard statistical analysis within the context of non-specialist learning in statistics for a short-term project may prove inappropriate. If you are keen to publish your work through co-authorship, enlisting the input of a professionally trained statistician to carry out this work for you after your project is submitted would make best sense. Make sure, however, that they have been fully involved in advising on your project from the outset. Approaching them retrospectively to deal with the ‘hard stuff’ inevitably leads to frustration all round when otherwise avoidable flaws found in seemingly easier analyses require all of the analyses to be re-run at a stage that is no longer convenient for you!
Summarizing and Analyzing Missing Data by Margaret MacDougall is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
