I have a presumably easy question for you, but I am a real newbie in statistics, so please be patient. I have a data set containing $N$ values that more or less follows a normal distribution. I need to select $N/10$ values among them, creating a subset with the same distribution. How can I do that? Do you have softwares to recommend? I usually use PSPP.

  • $\begingroup$ Why do you need to choose N/10? $\endgroup$ – rolando2 Apr 23 '18 at 16:00
  • $\begingroup$ Because I have a huge dataset and I cannot afford to continue my analysis and work with all of the values $\endgroup$ – Elisa Pieri Apr 24 '18 at 9:34

A sufficiently large random sample taken from your normally distributed population would give you a normal distribution. The larger the sample, the closer it will be to the original distribution. Choice of sample size will depend on how close you want to get to your population distribution.
I've never used PSPP, but it should have a random sampling feature.
You could use the sample() function in R.

| cite | improve this answer | |
  • $\begingroup$ Thank you very much! How do I know if my random sample is large enough? Is there some test to check it? $\endgroup$ – Elisa Pieri Apr 24 '18 at 9:09
  • $\begingroup$ There is a Kolmogorov-Smirnov test to see whether two samples come from the same population distribution. Probably under Nonparametric Tests...Independent Samples. Now, in your case you'll know that they do, but even so the result might reassure you that the features of the distribution are being preserved in the smaller sub-sample. $\endgroup$ – rolando2 Apr 24 '18 at 12:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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