1
$\begingroup$

So im working on a research proposal for a college class. I have a large sample of about 300-500. The objects in the sample will separated into 3 categories. I'm looking for some type of test that determines if the difference in amounts between groups is significant.

Ex. Say Group 1 ended up with 145 Group 2 ended up with 95 Group 3 with 60.

What types of test can be done to determine if the differences are significant and are there any test to determine that these are the likely results of the larger population.

$\endgroup$
7
  • $\begingroup$ Significant - compared to what? You'll need a null hypothesis (e.g. that the group sizes are equal). Is that the one you're assuming here? $\endgroup$ Apr 19 at 1:01
  • $\begingroup$ Regarding the Null hypothesis, yes. The null would be that they sizes should be equal. Regarding the significant part. Id say are sizes significantly different compared to one another and would we see that level of difference in the overall population. $\endgroup$
    – jack Hayes
    Apr 19 at 1:04
  • $\begingroup$ Neat, thanks for the extra info. A chi-squared test is pretty typical for this situation. $\endgroup$ Apr 19 at 1:05
  • $\begingroup$ Awesome thank you. $\endgroup$
    – jack Hayes
    Apr 19 at 1:07
  • $\begingroup$ Please add new info in comments as an edit to the post, so all information is in one place! $\endgroup$ Apr 19 at 1:58
1
$\begingroup$

Turning my comment into an answer:

You can use the chi-squared test here. It compares the expected frequency of events to the observed frequency. Based on your null hypothesis from our discussion in the comments, your expected counts are 100, 100, and 100. From the test, you can get a p-value.

$\endgroup$

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.