search

ropensci / essurvey

Download data from the European Social Survey
https://docs.ropensci.org/essurvey
Other
49 stars 9 forks source link

Add code to run Daniel Oesch's social class scripts? #50

Open briatte opened 3 years ago

briatte commented 3 years ago

https://people.unil.ch/danieloesch/scripts/

Available for Rounds 1--9.

See also:

https://github.com/jolyphil/oesch-class-ess-R/

… although this one is produced in Stata.

cimentadaj commented 3 years ago

Thanks @briatte. This is indeed a useful addition and something which I think would make essurvey much more popular within the social science community. I don't have enough time at the moment to implement this but I'd be happy to review a PR if you or anyone else is interested in contributing :smile:.

briatte commented 3 years ago

Alright, thanks!

Will do the following after we hear from ESS on the server issue (#51):

  1. Compute social class from Oesch's scripts for all available ESS rounds and countries.
  2. Rewrite his scripts as something that could be executed internally to the package.
  3. Write an oesch_class function that produces the class columns.
cimentadaj commented 3 years ago

This is a great. Would you mind looping me in the development? We should discuss the input/output just so that it can generalize well to all rounds. For example, perhaps it can be a funcion that accepts only a column of social classes and you need to provide a year argument to select which round transformation. This has the benefit that it can be embedded in mutate calls and/or used nicely with group_by if rounds are collapsed together. Were you thinking along these lines or something different?

briatte commented 3 years ago

Hi @cimentadaj — No problem, will let you know when I start working on this!

perhaps it can be a funcion that accepts only a column of social classes and you need to provide a year argument to select which round transformation.

I was thinking that perhaps the simplest way to get this right is to ask on import/download whether social class variables should be added to the result. This ensures that the function gets the right cntry and essround parameters, and also allows to know what's the data format (Stata/SPSS) on which it operates.

cimentadaj commented 3 years ago

Ah, I hadn't thought about this that way. Let's discuss this further when you work on this. I need to think whether it feels right to mix downloading/mutating into one call.

Jorge Cimentada https://cimentadaj.github.io/ https://cimentadaj.github.io/

On Mon, Apr 12, 2021 at 8:52 PM François Briatte @.***> wrote:

Hi @cimentadaj https://github.com/cimentadaj — No problem, will let you know when I start working on this!

perhaps it can be a funcion that accepts only a column of social classes and you need to provide a year argument to select which round transformation.

I was thinking that perhaps the simplest way to get this right is to ask on import/download whether social class variables should be added to the result. This ensures that the function gets the right cntry and essround parameters, and also allows to know what's the data format (Stata/SPSS) on which it operates.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ropensci/essurvey/issues/50#issuecomment-818048796, or unsubscribe https://github.com/notifications/unsubscribe-auth/AB6NCOKFKQLR4M37ZTUGXVDTIM6ONANCNFSM4YVVKOJQ .