7

I love and play a lot with the Baseball data package for R (Lahman), and I was wonder if there is anything like it but for more oriented to social sciences. Either surveys or admin data (PSID, HRS, ARF, NLSY, ACS) would be great.

I would appreciate any pointers.

Patrick Hoefler
  • 5,790
  • 4
  • 31
  • 47
Fernando Hoces De La Guardia
  • 421
  • 2
  • 5

4 Answers4

7

Here is a human-friendly list of all 731 datasets in R, as well as CSV download links.

Megan Squire
  • 471
  • 2
  • 7
2

Two suggestions:

(1) Check out the WebTechnologies Task View, which contains a bunch of packages linking to (potentially) open data on the web. In most cases, these packages will give you a list or data.frame structure you can work with.

(2) Take a quick glance through the list of CRAN packages, some of which are data-only or at least contain small datasets for their examples. For example, Hadley Wickham has some data only packages (fueleconomy, and hflights) but neither are strictly social science. HistData has datasets used in famous visualizations. pdfetch can retrieve economic time-series. Etc. etc.

Thomas
  • 1,114
  • 6
  • 14
1

Have a look at the mlbench package: "A collection of artificial and real-world machine learning benchmark problems, including, e.g., several data sets from the UCI repository."

vonjd
  • 136
  • 3
1

You should also check out RSocrata which lets you list and download any public dataset from governments throughout the country that leverage Socrata for their Open Data Portals.

stuart gano
  • 11
  • 1