Page 72 - Basic Statistics

67

95 percent confidence interval:
NA -6.305911

sample estimates:
mean of x mean of y
65.86667 73.93333

Comments:

Although the above results do computerized with t test, but its the same as

the Z-test. These results can be directly compared with the critical value of the

Z-test.

a). The conclusion of the two-tailed test; Pr = [ P ( Z  -z-actual) + P ( Z  z-
actual ] < , ie 0.0000< 0.01. If used manual testing criteria, then

 Z-actual  > Z/2, ie 7.658 > 2.57, then H0 is rejected at the 1 percent

significance level. We conclude that there is a significant difference between

the average test scores of STAD and TPS.

b). The conclusion of the two-tailed test; Pr ( Z  z actual ) <  , ie 0.0000 < 0.01,

then H0 is rejected at the 1 percent significance level. We conclude that TPS

is better than STAD.

4.2.2 MEAN DIFFERENCE TEST USING T-STUDENT TEST

When dealing with situations where variance of the two populations (1

and 2) are not known and sample sizes small (n1 < 30 atau n2 < 30 ), the tes
procedure make use of the t distribution. The procedural steps are identical to

those specified above for large independent samples, however to special case

noted below.

1. Hypothesis test to test the equality of the mean of two independent
populations (two-tailed test) is formulated as follows:

~~* CHAPTER 4 TESTING HYPOTHESIS *~~