0
$\begingroup$

I have a dataset with some missing data. In this dataset, there are two variables that their missing data behaviors are same. Because these variables represent hormone measures, absence of one leads to lack of another (those are measured together). What is the missing-data mechanism in this situation? Missingness completely at random (MCAR) or Missingness at random (MAR) or another mechanisms?

$\endgroup$
0
$\begingroup$

Without further information about the data at your disposal, I'd say this is missing not at random (MNAR). I am saying this because the reason the 'other' value is missing, is another value missing. Basically, the missingness of the data is caused by/depended on some aspect of the missing data (see this reference).

However, if there is more data at your disposal which is known to be associated or unknowingly associated with the missing hormone levels (suggesting a partial missing at random (MAR) mechanism), you might still be able to make valid estimates of the missing values through multiple imputation techniques.

$\endgroup$

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.