The reason for asking is that I have to teach a fi of course in excel (the students don't have access the Stata). I'm trying t to keep it simple and a binomial test in excel is more complicated.
Using a chi-squared test to approximate a binomial test is only valid with a very large number of samples (see e.g. http://www.talkstats.com/showthread.php/13396-Fair-Coin...-Chi-square-or-binomial)
If you want something with a direct implementation of the binomial test and don't have access to Stata, you could consider using R. But in any case, the Excel function looks pretty straightforward to me too:
http://homepages.wmich.edu/~bwagner/StatReview/Binomial/binomia%20in%20excel.htm
http://www.real-statistics.com/binomial-and-related-distributions/hypothesis-testing-binomial-distribution/
http://www.real-statistics.com/binomial-and-related-distributions/binomial-distribution/
Those are helpful. But I think I can use a Chi-squared "goodness of fit" test. It's a little complicated, but I've already been teaching Chi-squared tests by brute force because it's an important test. The binomial test requires a number of steps. I think I'll stick to Chi-squared test for this application, since it's fewer things for me to walk my students through (in a one-week crash course on statistics).
- Badges
- Science topic