Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.
Sign up to join this community
So I have big dataset with many independent variables. Because there are data gaps in many variables (which some don't overlap between variables) the final model is made of much smaller sample size than the whole dataset. I am trying to find a way to extract only the cases that were used in the regression model so I can create some graphs based on it. However, I can't find how to do it in SPSS, neither googling is giving results. Any help is appreciated
Zorin
Software-specific questions are off-topic here, but your question has some on-topic statistical content.
Because there are data gaps in many variables (which some don't overlap between variables) the final model is made of much smaller sample size than the whole dataset.
A better way to proceed in this type of situation is imputation to get multiple copies of a full data set with probabilistically estimated values for the missing data. That removes the problem of having a "much smaller sample size than the whole data set" while allowing you to take both the inherent modeling error and the error in multiple imputation into account. See the on-line book Flexible Imputation of Missing Data. I don't use SPSS, but I suspect that it will have tools to let you implement that approach.
