Hello, I am having trouble solving this correlated groups t-test word problem.

Hello, I am having trouble solving this correlated groups t-test word problem. Can someone please help me, my APA conclusion doesn't seem to match my results, and I would like to know why.

1. A measure of social phobia was administered to ten individuals who had come to an anxiety clinic for treatment. The scores range from 0 to 100 (with higher scores indicating greater phobia). Each individual was provided with a ten-week clinical intervention effort and were re-evaluated one month after the intervention. Is the ten-week clinical intervention effective at reducing social phobia? Here are the data:

Individual                  Pretest            Posttest

  1                             97                   70

  2                              90                   72

  3                              92                   70

  4                              92                   76

  5                              96                   82

  6                              94                   85

  7                              82                   79

  8                              91                   80

  9                              95                   79

10                             89                   80

a.     State the test to be conducted and why:

Correlated groups t test; comparing same people in group twice (repeated measures) and this is a within groups design (2 levels: Posttest and Pretest)

b.    State the null and alternative hypotheses:

Ho: mPretest = mPosttest

H1: mPretest < mPosttest

c.     Find the CV and state the decision rule. Sketch the curve.

           CV: 1.833  Decision Rule: Reject Ho if tobs ≤ 1.883             tobs =   6.599           Fail to reject the Ho!

Eta2 = 6.5992/6.5992 + 9                       43.55 / 43.55 + 9 = 0.83

 Mean 1 (Pretest) = 91.8Mean 2 (Posttest) = 77.3

a.     Write an APA conclusion based on your results:      

A correlated groups t test was performed to determine if individuals who had attended an anxiety clinic for a ten-week clinical intervention would have lower anxiety scores on the posttest than they did on the pretest prior to attending the intervention. Anxiety scores for the pretest (M = 91.8) were not significantly higher than posttest scores (M = 77.3), t (9) = 6.599, p > 0.05, eta2 = 0.83. The effect of the ten-week clinical intervention was strong, accounting for 83% of variability in anxiety scores, after controlling for individual differences. This suggests that the intervention did not result in lower anxiety scores on the posttest.