erikgahner / poldata Goto Github PK

Thank you for this great resource.

Demscore (https://demscore.se/) might be a good addition. It might fit under the subcategory Democracy or Additional Overviews of Datasets.

I would suggest to use the flat ODS spreadsheet format. This can be used in addition to .csv or .xlsx.
ODS is the OpenDocument Spreadsheet format of LibreOffice Calc; in short, .ods files are zipped XML files.
Flat ODS is the same standard; except the "compression" algorithm is flat: meaning, uncompressed.

Flat ODS is more convenient than CSV: it opens right away in a spreadsheet software; column widht and types are preserved, as well as formatting...
Flat ODS is more convenient than XLSX: the result is readable by human, it is a 100% opensource technology.
Flat ODS is more convenient than ODS: being plain text, git knows how to make diffs of them; see a sample here; this is why LibreOffice created the flat format; see "LibreOffice and Version control".

Flat ODS files can be named .fods or .ods; they will be opened by Calc and Excel.

readODS::write_fods() supports writing in flat ods.

Based on code in #14, below is a sample code to write a flat ODS file:

readODS::write_fods(spdx_licenses_2_df, 
                    path = "spdx_licenses.fods",
                    row_names = FALSE # the default anyway
                    )

Another advantage of write_fods (and of write_ods) is the support of sheets within a single spreadsheet document.
This would allow, for instance, to store the content of the README in a additional sheet, within the same document.

This might be of interest:

Cooperative Congressional Election Study
link: https://cces.gov.harvard.edu/
year_start: 2006
year_end: 2016 (ongoing)
availability: free online

Thank you for this great catalog!
It would help a lot to have a field for the licence of each dataset, using one of the normalised licence codes.
This would complement these very helpful fields :

• availability = Availability of dataset
• registration = Requirements for data access
• free= Free access to data (1 = Yes, 0 = No)

In addition, you may want to specify an open data licence for the list itself (not the R package). (If one is already specified, I failed to see it on the Readme).

The current link for The American Panel Study (https://taps.wustl.edu/data-archive) is invalid. Please consider using the updated one (https://wc.wustl.edu/american-panel-survey).

We have just released a couple of sets on Chilean cabinets that might be interesting.

I am attaching a CSV with further details.

PolData.csv

I work on the Crowd Counting Consortium, and I think the compiled version of our dataset (on GitHub, here) would make a good addition to your list. It's comparable to the European protest datasets you list now, but it covers the U.S. since January 2017. It currently contains more than 110,000 records and is updated weekly.

In any case, thanks for creating and maintaining this resource.

add variables:

• codebook
• file_csv
• file_dta
• file_sav
• file_excel
• file_r
• file_zip