lynusrynu Posted January 28, 2014 Report Share Posted January 28, 2014 Hello! I'm doing my IA on correlation between goals scored and footballer's BMI on the sample of 50 footballers.I know that Chi-squared test must have expected value bigger than 5 and I have divided goals scored in 3 groups and BMI also in 3 groups to perform the test, however I get expected value of 4.8 in one case. Therefore my question is: should I narrow the groups to 2x2 or should I leave it this way and state it as limitation/give a comment on test's valadility?There are these example IA's on the CD added to Oxford's book for Studies where once they said the whole calculations are bad due to performing Chi-squared with smaller values than 5 but once when someone added comments to that there was no negative comment from the moderator... thanks in advancecheers Reply Link to post Share on other sites More sharing options...
sameera95 Posted January 28, 2014 Report Share Posted January 28, 2014 Maybe this is because your sample is too small? We were told that we need a sample of at least 80, and you only have 50. Would you be able to get more data? Reply Link to post Share on other sites More sharing options...
lynusrynu Posted January 28, 2014 Author Report Share Posted January 28, 2014 yes I am, but we were told the minimum for IA is 30 so I thought 50 would be suitable Reply Link to post Share on other sites More sharing options...
sameera95 Posted January 28, 2014 Report Share Posted January 28, 2014 will it take you too long to get more data? have you done any other calculations so far? because if you have then your results will change if you get more data Reply Link to post Share on other sites More sharing options...
lynusrynu Posted January 29, 2014 Author Report Share Posted January 29, 2014 it won't but yes I have done all other calculations I could and one set of my data is not normally distributed so I had some problems with it too.. well I guess I will just add data and calculate everything again :/ Reply Link to post Share on other sites More sharing options...
sameera95 Posted January 29, 2014 Report Share Posted January 29, 2014 oh I'm sorry good luck! Reply Link to post Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.