# T-test,Z-Test & F-Test & Chi-Square

Discussion in 'Scientific Statistics Math' started by Judge, Jul 13, 2008.

1. ### JudgeGuest

All,

T-Test & Z-Test are used to check means of two groups. T-test is used
when groups are small & variance is not know otherwise Z-test is used
to check regarding mean diff. in two groups.

F-Test & Chi-Square are used to check variance of two groups. When
each of these is used.

I am not a statistics student but am a professional in OLAP reporting;
am trying to learn Data Mining(Stats) on my own. Your help in removing
my stat concepts confusions can save a lot of time for me. Thanks.
Judge

Judge, Jul 13, 2008

2. ### Jack TomskyGuest

All,

Generally, you're right except that chi-square is used to test for a single variance. Suppose that the underlying distribution is normal with the mean and variance being unknown. We wish to test the null hypothesis H0: sig^2 = sig0^2, where sig^2 is the true population variance and sig0^2 is the specified population variance.

From the sample of size N, we calculate the sample standard deviation s. Then when H is true,
(N-1)s^2/sig0^2 has a chi-square distribution with N-1 degrees of freedom. We use this fact, along with whether the specified alternative hypothesis H1 is one-sided or two-sided to decide whether accept H0 or H1.

Jack

Jack Tomsky, Jul 14, 2008