Hello,
I am currently writing my bachelor thesis on the topic of the gender pay gap and am conducting an empirical study on this topic. I am using Stata 17 and a SOEP dataset (more precisely the SOEP core v36eu $pequiv) for the analysis.
Since I am only calculating a GPG for the year 2018, I first filtered out only the 2018 data.
Now I want to use the income variable to calculate the effect of gender using a simple bivariate linear regression for now (the unadjusted GPG).
The dataset has only annual income (both individual and household income) as variables. My question now is how to convert this to an hourly wage?
I then want to logarithmize the variable of 'hourly wage' and use it as dependent variable.
The codebook for the dataset is here: https://www.diw.de/de/diw_01.c.80965..._the_soep.html
The income variables I want to use are called i11110 and i11103
Also, I had wondered how to weight correctly afterwards. Namely, in the dataset there are a lot of weighting variables and now I am a bit overwhelmed which ones to use and how (The weighting variables are from W11101 to W11111).
I think the code might look something like this: reg ln_individual income i.gender [iweight = w11101], r
What do you guys think?
Can you guys help me?
Best regards,
Dario Nitschke
I am currently writing my bachelor thesis on the topic of the gender pay gap and am conducting an empirical study on this topic. I am using Stata 17 and a SOEP dataset (more precisely the SOEP core v36eu $pequiv) for the analysis.
Since I am only calculating a GPG for the year 2018, I first filtered out only the 2018 data.
Now I want to use the income variable to calculate the effect of gender using a simple bivariate linear regression for now (the unadjusted GPG).
The dataset has only annual income (both individual and household income) as variables. My question now is how to convert this to an hourly wage?
I then want to logarithmize the variable of 'hourly wage' and use it as dependent variable.
The codebook for the dataset is here: https://www.diw.de/de/diw_01.c.80965..._the_soep.html
The income variables I want to use are called i11110 and i11103
Also, I had wondered how to weight correctly afterwards. Namely, in the dataset there are a lot of weighting variables and now I am a bit overwhelmed which ones to use and how (The weighting variables are from W11101 to W11111).
I think the code might look something like this: reg ln_individual income i.gender [iweight = w11101], r
What do you guys think?
Can you guys help me?
Best regards,
Dario Nitschke
