Subjects were asked to perform a certain task, and the nature of the instructions varied. In one condition, the subject was led to feel coerced; in the other condition, they were not. The experimenter noted how many subjects complied with the request and how many refused. The resulting data are shown below. Do the two conditions differ significantly at the 95% level?

Observed frequencies

Coerced Not Coerced

Refused 79 72

Complied 66 55

I tried mann-whitney and chi squared but it was on the stats drop down on mintab...the calc drop down version wouldnt give a chi value :S

Mann-Whitney Test and CI: Coerced, Not Coerced

N Median

Coerced 2 72.50

Not Coerced 2 63.50

Point estimate for ETA1-ETA2 is 9.00

75.5 Percent CI for ETA1-ETA2 is (-6.00,24.00)

W = 6.0

Test of ETA1 = ETA2 vs ETA1 not = ETA2 is significant at 0.6985

Chi-Square Test: Coerced, Not Coerced

Expected counts are printed below observed counts

Chi-Square contributions are printed below expected counts

Not

Coerced Coerced Total

1 79 72 151

80.50 70.50

0.028 0.032

2 66 55 121

64.50 56.50

0.035 0.040

Total 145 127 272

Chi-Sq = 0.134, DF = 1, P-Value = 0.714

An investigator has data on 9 individuals who have been in psychotherapy for five years. The original score the individual received on a personality test plus his improvement in psychotherapy over the five years are tabulated.

Produce a scatter plot of the data, plus a supporting calculation of the correlation coefficient.

Fit an appropriate linear regression to your data and estimate the improvement for original scores of 60 and 93.

Personality Score Improvement

1 60 25

2 78 25

3 93 31

4 64 24

5 77 18

6 89 38

7 78 38

8 93 32

9 75 22

Thanks everyone.

H

x