I am currently trying to perform some statistical analysis on some data to see if there is any meaningful conclusion for a research project I am working on; however, I have come across a problem. There is a lot of null responses in my data set (because the person could not answer the question).

Here is the link to the data if you would like to look at it: https://docs.google.com/spreadsheets/d/1QXUVGP7mRJJERUFkv10O-3EjhSNJCq79EOC6F_Gtm30/edit?usp=sharing (It is the top block)

This data records the amount of time it took responders to answer various questions I posed, but the problem is the responders were not always able to answer the questions. I am comparing the times of the people in the L category to the people in the R category (last column)

Is there a generally accepted way for a method to include those null responses into the results? I have thought about penalizing them with number such as two times the mean of the other results, but that would ruin $\sigma$ (it also affects $\mu$).


Here are a few links to some of the huge literature on dealing with missing data:

Wikipedia, Missing data

Wikipedia, Expectation–maximization algorithm

Little--Rubin, Statistical Analysis with Missing Data, 2nd Edition

  • $\begingroup$ Thank you for these resources. I discovered what I was looking for was regression imputation. $\endgroup$ – Stephen Fratamico Mar 10 at 23:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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