I´m searching for a gripping data set from educational research which is

  • a simple random sample (and therefore without weights)
  • representative
  • and available as public-usefile

to use for teaching. Any suggestions?

EDIT: I`m teaching introductory statistics to undergrats, which are (student) teachers. Representative means, that the demographic covariates of the sample are equally distributed to the population.

  • 1
    Sorry, your question is unclear, please edit it. What is representative? What should be the sample size? In what format? Why gripping (says nothing)? For teaching what? Why from educational research? Please read this to improve your question.
    – user4293
    Commented Nov 7, 2017 at 14:12
  • is this an oer question?
    – albert
    Commented Nov 7, 2017 at 15:13
  • Please bear in mind that this is an international website and specify the country/education system you would like the data set to be representative of. Commented Dec 13, 2017 at 10:09

1 Answer 1


Since you go with statistical analysis, I recommend the R-Datasets. They are usually small and statistically interesting: https://stat.ethz.ch/R-manual/R-devel/library/datasets/html/00Index.html

They are not necessarily representative in a bigger context, but they represent one given experiment/survey.

Famous datasets my lecturers used for statistics are: Iris, Titanic, MTCars, Anscombe

EDIT similar question: Data sets for teaching

  • 1
    Thank's for this answer. I'm already aware of these datasets and think it is great that they are implemented into R. Nevertheless I understand my students who say "I'm not interested in the distribution of petals length, I'm interested in the distribution of student's literacy and it's relation to ....".
    – sammerk
    Commented Nov 9, 2017 at 10:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.