Reply
Sun 28 Jul, 2013 08:34 am
I'm not sure at all how to test this hypothesis.
I have made a piece of code that distinguishes males and females, by calculating their fundamental frequency. My program has done this with a success rate of 87%
null hypothesis: The code can distinguish males and females with a success rate of 75%
alt. hypothesis: the code can not distinguish males and females with a success rate of 75%
I am very confused as to how I should test this hypothesis, i.e. what tests to use etc.
@jbeirh,
i don't understand. Can you explain clearly. what the hypothesis use to?
@jbeirh,
IMO the null hypothesis is that "the code" success rate above 75% is a chance event. The alternative hypthesis is that it is not a chance event. The binomial test may be appropriate, (but it is many years since I used it).
http://www.elderlab.yorku.ca/~aaron/Stats2022/BinomialTest.htm
@jbeirh,
Hey guys, thanks for the answer. I ended up figuring it out earlier
@jbeirh,
This page shows how to perform a number of statistical tests using Stata. Each section gives a brief description of the aim of the statistical test, when it is used, an example showing the Stata commands and Stata output with a brief interpretation of the output.