View Full Version : t-test report


Humphrey
05-09-2006, 06:04 AM
Dear list members,
I have done t-test with SPSS and have the table. it has a lot of info. which i think is redundant.

Levene's Test for Equality of Variances t-test for Equality of Means
F Sig. t df Sig. (2-tailed) Mean Difference Std. Error Difference 95% Confidence Interval of the Difference
Lower Upper
READING Equal variances assumed 5,946 ,016 -18,921 158 ,000 -6,01679 ,31800 -6,64486 -5,38872
Equal variances not assumed -20,153 153,275 ,000 -6,01679 ,29856 -6,60660 -5,42697




Do i have to report all of it?
i'm writing a papre. what should i report as the result of the t-test?
should I report the table as it stands here?
what usually gets reported in the case of a t-test?
Thanks

JohnM
05-09-2006, 09:13 AM
Put the whole thing in an Appendix, but in the main report, you should talk about the null and alternative hypotheses, your alpha level (significance level) for the test, the t-statistic, and its p-value, and whether you had enough evidence to reject Ho.