weights and MI

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

Amira ELasra
Posts: 1
Joined: Thu Jul 30, 2015 6:10 pm

weights and MI

Postby Amira ELasra » Thu Jul 30, 2015 6:19 pm


I ham using the LSYPE longitudinal study to perform multiple imputation. it has 7 waves and each wave has its weights variable(s) and i want to impute many variables from all waves. my questions are:
1. which weight variable(s) of which wave to use when i run MI?
2. if i run MI without weights, how can i adjust for them in later analysis, such as using calibration or post-stratification?
3. if i chose not to use MI and just use the weight variable(s) in my analysis models, which weight variable(s) to use if i am mixing variables from different waves? should i use the most recent weights of the most recent wave?

Who is online

Users browsing this forum: No registered users and 1 guest