Project

General

Profile

Actions

Support #1944

open

Sample size in regressions

Added by Michael Vallely about 1 year ago. Updated 11 months ago.

Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
Data management
Start date:
07/27/2023
% Done:

100%


Description

Hi Alita,

I'm running Nested OLS regressions to examine the social class pay gap for each wave of data (I am using waves 1-9). Firstly, I proxy for respondents' social class, then control for their demographics, education, labour market features etc. in subsequent models. I am conditioning all models on respondents' being in employment. The sample size decreases each time I add in an additional control. This happens when I am running the models on an unbalanced and a balanced panel. Therefore, my question is why is the sample size not consistent across all models?

Thanks,
Michael

Actions #1

Updated by Understanding Society User Support Team about 1 year ago

  • Status changed from New to Feedback
  • Assignee deleted (Alita Nandi)
  • % Done changed from 0 to 80
  • Private changed from Yes to No

Dear Michael,

If I understand your question right, this happens because each additional variable added to your model has its own missing data, therefore with each added variable the missingness accumulates.

Best wishes,
Piotr,
UKHLS USer Support

Actions #2

Updated by Understanding Society User Support Team about 1 year ago

  • Category changed from Data analysis to Data management
  • % Done changed from 80 to 100
Actions #3

Updated by Understanding Society User Support Team 11 months ago

  • Status changed from Feedback to Resolved
Actions

Also available in: Atom PDF