Missing Data

Bummer! But, that's life. Stuff happens. Gone!

If you are lucky, the data can be considered Missing Completely At Random (MCAR), where being missing is unrelated to any of the observations' values. In that case, ignore them and soldier on as best you can. If "missingness" is related to data values--in particular, the outcome (for example, if those not losing weight are more likely to drop out of a diet study)--you're out of luck!

There are many techniques that have been used over the years to impute missing data. The problem is that they all involve assumptions. We all know the old saw about "assume" and what it does to "u" and "me". Unfortunately, it's true!

Copyright © 2008 Gerard E. Dallal