t test between two variables

[lone0708]

Hi all, 

i have a dataset with one oberservation for each id. I want to compare values between two variables using paired t-test. 

My problem is, that i have a lot of missings in one of the variables and when i do the proc t-test, sas only uses a small number of observations.

Is there a way, where i can compare means calculated using all the observations in each variable?

[PaigeMiller · Accepted solution]

Lots of discussion on the internet about this issue of missing values in a paired t-test.

For example, here we see

Missing values can severely impact a paired sample t-test because the entire row of data will generally be excluded. If you have a lot of missing data, one alternative would be to perform a single factor repeated measures mixed model ANOVA. This would allow for the computation of estimated marginal means to compensate for the uneven replication between groups.