Can I drop obs whose EDGRADE and EDGRDLY values are both NIU for MISH values of 4 and 8? October supplement of CPl


#1

I want to find the effect on graduate enrollment of business cycles for respondents aged 21-35 who have completed a bachelor’s degree from the October supplement data of CPS (1994-2015). I understand that if the MISH values are 4 or 8, they form part of outgoing rotation groups and are not interviewed the following time. So can I drop observations for the above two educational attainment variables whose codes are both NIU, for MISH values of 4 and 8? Or should I drop all observations for all MISH values if both EDGRADE and EDFRDLY are NIU?

Thank you!


#2

I’m not 100% certain of what you are aiming to do, since the outgoing rotation group does not directly relate to the October Education Supplement variables. In general, it is reasonable to drop all observations that are not in the universe when analyzing a specific variable. So, from what I can understand, there is no reason to drop only observations with NIU for MISH==4 | MISH==8. Of course, feel free to follow up with more details if I haven’t adequately addressed your question.