Hello all! I am trying to conduct a study analyzing voting patterns using SOEP panel data. I am specifically looking to track respondents turnout in multiple waves of elections. I am using the most up-to-data .dta files from the panel versin of the SOEP (soep-is.2022_stata_de/pl.dta). Unfortunately, when I access the data, I only seem to have the variables for tunrout in 2009 (plh0006) and 2013 (plh0333), but I cannot find any other variables for turnout/voting in other years. I see that ostensibly these turnout/party voting variables are measured in other years (see, for example https://paneldata.org/soep-core/instruments/soep-core-2022-pe2-m3456/197), however I cannot find any variables and none of the other variables appear to correspond. Is it possible to obtain SOEP voting history for other years? Is this a mistake?
Many thanks!
Edit: more specific details on the data issues below.
Dear John,
Thank you for using our online forum.
First I have a clarification question:
In your text it looks like you mix up SOEP-Core data and SOEP-IS data.
These are two completely different studies with completely different questionnaires and different samples/participants.
The first part of your text indicates that you are using the SOEP-IS data, while the link you shared below is pointing to the SOEP-Core data.
You can find the questionnaires used for SOEP-IS here:
For researching the voting behaviour and party preference, it might be more useful to use the SOEP-Core data. Or is there a specific reason why you want to use the SOEP-IS data?
Best regards and enjoy the weekend
Philipp
1 „Gefällt mir“
Dear Philipp,
Thank you very much for responding and for your suggestions! Yes, I think I’m a bit unsure of which of the two datasets to use. Really what I would like to do is match any available voting/turnout data from the SOEP to their meaasures of night work. I know the latter have a number of different variables, which I believe are in the SOEP-Core data. For example, in the .dta file, I find plb0206_v2 to measure night work in years 2000-2022. I also find on paneldata previous years of data (see here, for example) with the night work questions. So I am actually agnostic about using the SOEP-Core or the SOEP-IS, but I would just like to maximize the coverage of voting records with night shift work, which I would like to match by respondent. Would you have any suggestions on how to retrieve these sets of variables and where to look in the data?
Thank you so much and have a wonderful weekend yourself!
Best,
John
Dear John,
I’ve spoken with a colleague of the SOEP-IS team and he strongly agrees to recommend to use the SOEP-Core data instead of SOEP-IS for this.
You should find all this information in the longitudinal „pl“ dataset of SOEP-Core:
Variable for party preference:
plh0012_h (https://paneldata.org/soep-core/datasets/pl/plh0012_h)
Variable for which party a person voted for in the last Bundestag-election:
plh0333 (https://paneldata.org/soep-core/datasets/pl/plh0333)
Variables about night work (unfortunately it looks like there is no harmonized version of the variable yet, so there are versions for different survey years):
plb0206_v1 (https://paneldata.org/soep-core/datasets/pl/plb0206_v1)
plb0206_v2 (https://paneldata.org/soep-core/datasets/pl/plb0206_v2)
plb0206_v3 (https://paneldata.org/soep-core/datasets/pl/plb0206_v3)
plb0206_v4 (https://paneldata.org/soep-core/datasets/pl/plb0206_v4)
plb0206_v5 (https://paneldata.org/soep-core/datasets/pl/plb0206_v5)
plb0206_v6 (https://paneldata.org/soep-core/datasets/pl/plb0206_v6)
Hope this helps and that you are having a great week so far!
Best,
Philipp
1 „Gefällt mir“
Dear Philipp,
Thank you so much, this is very helpful and I truly appreciate your support. If I may follow up, I have two questions:
- Looking at the pl data, it seems that only the plb0206_v2 variable is available, but it looks like it may be harmonized because it’s available for multiple years:
I checked and the values of the variable vary within pid across waves, so it doesn’t seem like it’s the same year’s values being repeated for multiple waves of the same respondent. Would this interpretation be accurate and I can assume it to be harmonized?
- Thank you veyr much for sharing the party preference/vote variables. For the voted variable, it seems that this is only available for 1 singular wave, the most recent wave of the SOEP (2022).
But when I follow the online documentation you sent, it suggests that there are previous versions of the question available, for example here: link
but such variables are also ostensibly labelled under plh0333 in the longtiduinal data. Is there a chance this is a glitch with the latest data, or is it not possible to retrieve pl0333 from previous years?
Thank you so much for your patience. I truly appreciate your receptiveness to my inquiry and I hope you have a great week as well!
Best,
John
Sorry, I cannot seem to post that link, so it should be
soep-core/instruments/soep-core-2018-pe/175
after https://paneldata.org/
Dear John,
I have forwarded the question about the harmonization of plb0206 to my responsible colleagues and look forward that they can clarify this soon.
About your second question regarding plh0333, it should be filled for 2014 and 2018, too. When I look into the dataset, there are values for 2014, 2018 and 2022.
Please have in mind that the Bundestag-election is only every 4 years.
Best,
Philipp
1 „Gefällt mir“
Dear Philipp,
Thank you so much for following up. On the second question, I am thinking that there might be something wrong with the dataset I have. I’m using the pl dta file, which I have downloaded into R, and it seems there is only the 2022 value:
Is there a chance that something is wrong with the dataset, or should I look to another file instead?
Many thanks for your patience with all of my questions.
Best,
John
Dear John,
regarding plb0206*: we have not harmonized the variables concerning night work so you would have to combine them yourself according to your needs. It is noteworthy that night work is not filled continuously - especially in 2014, 2018 and 2022 this was not asked and therefore you will only find value -8 for those years:
(excerpt from SOEP.v40 (internal version) loaded into Stata)
So I would suggest to use party preference instead of preference for Bundestag-election if it fits your research question.
I have looked into the data of SOEP.v39 and SOEP.v40 and plh0333 is filled with vaild cases for 2014, 2018 and 2022.
I can’t really tell why it only shows year 2022 for you, maybe you can give more details about what you did to get to the outcome in your screenshots or maybe @SOEP_P.Kaminsky has any further ideas.
