In Longitudinal study, What are "Time-variant covariates", "Time invariant covariates" and "Time-related covariates"? and what is differences between them?


Let's say that your observations are people.

Time-invariant covariates: Values for these variables will be the same no matter when they are observed. "Place of Birth" cannot change, whether the observation is from 2000 or 2014. Race and Sex are often treated as time-invariant as well.

Time-Variant Covariates: Values for these variables can (but don't necessarily) change with time. "Education" is a classic time-variant covariate. A person observed in 2008 may only have a high school degree, but by 2014 she may have a college degree. Marital status, state of residence, etc. are also possible time-variant covariates. Some people will not got married within your study's time-frame, but others will.

Time-Related Covariates: Values for these variables necessarily change with time. Think "day," "month," "year," "age," "time in current residence," and so on. In a longitudinal study, two observations based on the same respondent should never have identical values for a time-related covariate, because time keeps on moving.

| cite | improve this answer | |

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.