Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have a dataset of 500 people and am trying to fit a prediction model using both quantitative and categorical variables. I have adjusted the dataset as much as possible, but still have one variable (which is important in the analysis) with 19 missing cases. What would you say is the best approach/program to use to replace the missing values/instead of deletion? Would the EM estimator from SPSS provide good enough replacement? Seeing that I have important categorical predictors that will be used in the analysis, I understand that I can't use them in Amos. What other programs could be suited for this type of model?

Thank you!

share|improve this question
up vote 3 down vote accepted

There is not one best approach. All missing data methods come with assumptions, which may or may not be realistic for your data.

However, the impact of the chosen missing data method on the result highly depends on the amount of missing data. You have 500 cases and one variable with just 19 missing values. In that case, the proportion of incomplete records is so low (<4%) that it probably does not matter what you do. So unless the 19 missing cases are very special, the simplest is to delete the 19 incomplete records. This is called complete-case analysis, and the default in many software packages. For higher proportions of missing data, you need to be more careful though.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.