Values of dataframe in psid_data_setup.py

[prrathi]

The final dataframe panel_liof psid_data_setup.py seems to have issues with the values in its columns. Couple that I noticed while working with some (through the pickle file that the code stores) were:

• many of the values in column num_children were null
• the values in column spouse_age were 0

I suspect this is the case for several other columns as well. These columns also weren't manipulated in psid_data_setup.py, so @MaxGhenis suggested looking back at the psidR package.

[prrathi]

Just experimented more with the variables, a lot of the columns were only used for specific years or have been discontinued.
In the example I gave, num_children was only used in 2007, so it's null for the majority of instances. Another example of this is fam_smpl_wgt_core which is discontinued after 1992. Also, the spouse_age isn't an issue it's intended to be 0 sorry about that.

[jdebacker]

@prrathi Let me know if anything here is a major issue. Note that in the psid_download.R script, I'm pulling some variables we may not currently be using just in case we need them for future calibration efforts. I was also liberal in the drawing of variables related to sampling weights since I am not a PSID expert and wasn't sure what was of most use in different contexts.

[prrathi]

@jdebacker I believe '# of Individual Records' with most recent name ER52168 would be a beneficial add, has data since 1981 about total number in each household, which I don't think is part of the variables currently used.

[MaxGhenis]

In the example I gave, num_children was only used in 2007, so it's null for the majority of instances. Another example of this is fam_smpl_wgt_core which is discontinued after 1992.

So there's no associated variable for num_children in other years in the crosswalk spreadsheet?

Also, the spouse_age isn't an issue it's intended to be 0 sorry about that.

Cool, so it's zero for unmarried people and nonzero for married people?

[jdebacker]

@prrathi That probably is a useful variable to pull. Please feel free to open a PR to add that (or any other variables you think would be useful).

[prrathi]

So there's no associated variable for num_children in other years in the crosswalk spreadsheet?

That specific variable is for children 30 and less, with no associated variables, num_children_under18 is present across the years

Cool, so it's zero for unmarried people and nonzero for married people?

yes

[MaxGhenis]

Ah I'm seeing I found that about num_children in #6 (comment). Removing num_children from the data pull probably makes sense in that case.