Chi-square or t-test?

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Chi-square or t-test?
Message posted by Jenny (via 202.84.162.147) on June 11, 2001 at 2:33 AM (ET)

I have some questionnaire responses which need analysing, and I have 2 people telling me two different things! I have agree/disagree responses to a series of statements; the responses come from 2 different groups of teachers, distinguished on the basis of where they went to secondary school. So, that's one group of 37, one of 38, and I want to know if the differences in the numbers from each group who agree with each statement are significant. One person told me t-tests based on the percentages who agree will give me the significance - but my MA supervisor wants me to do chi-squares (agree/disagree, educated here/not educated here). I am totally confused, as a newcomer to even the most basic statistics. Help!!!!!!

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Chi-square or t-test?
Message posted by JG (via 128.8.23.172) on June 11, 2001 at 10:51 AM (ET)

You should start with a Chi-squared test for independence. The results of this test will determine what you do next.

Re: Chi-square or t-test?
Message posted by suluan (via 202.184.26.249) on June 12, 2001 at 7:51 PM (ET)

Your MA superv. is right. Your data are in nominal/categorical form. So, chi square is the answer.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.