Please double check the answer to this question for me befor

Please double check the answer to this question for me before I turn it in.

With 8 subjects in one group and 12 in the second group, what t-ratio defines the 0.01 critical region for a one-tail test? For a two-tail test?

Sample 1 df= 8-2 = 6 so 1.9432 one-tail and 2.4469 two-tail

Explain, without math, the relationship among the three t-tests; also explain when it

Solution

Degree of freedom for two samples in a conservative way = smallest -1

Hence if n1 =8 n2 =12 df = 8-1 =7 in a conservative method

Otherwise n1+n2-2 = 18.

A one sample t test is for testing a sample mean with that of population mean. Here assumption is sample mean is normally distributed.

When two independent groups are compared then the difference of their means is assumed to follow a normal distribution and values compared.

For a paired t test, two related observations, example a no of workers before and after training, a list of patients health before and after some medicine etc.

IN all tests, statistic is calculated and also p value.

If p value < alpha say 0.01, null hypothesis is rejected

Null hypothesis normally will be two means are equal.