Consider this study on social stress conducted by Kim Mooney (1990) at the University of New Hampshire. SBP was measured at six points in time using a Finapres noninvasive blood pressure monitor:

Time 1: When participants arrived at the laboratory and before they signed an informed consent form

Time 2: After participants read and signed an informed consent form that told them that they would be doing a stressful social role play (about being accused of shoplifting) that would be videotaped (they were told that they had 5 minutes to think about what they would say in this situation)

Time 3: After a 5-min period (during which they prepared what to say)

Time 4: During the stressful social role play

Time 5: Ten minutes after the role play was completed

Time 6: Twenty minutes after the role play was completed

At each point in time, for each of 65 participants, SBP was measured; the variables sys1, sys2, ..., sys6 correspond to the measurements of blood pressure made at the six points in time just described. (The blood pressure monitor used in this study looked at peripheral arterial pressure in the middle finger, and so the blood pressure readings tended to be systematically higher than the readings usually obtained by more conventional blood pressure measurement methods.)

2. TIME Measure: MEASURE 1 95% Confidence Interval TIME Mean Std. Error Lower Bound Upper Bound 125.875 2.433 121.013 130.737 139.188 2.537 134.1 18 144.257 145.922 2.384 141.159 150.685 OUT A W N - 164.391 2.744 158.907 169.874 149.000 2.467 144.070 153.930 139.375 2.374 134.630 144.120 Mauchly's Test of Sphericity Measure: MEASURE_1 Epsilon Approx. Greenhous Within Subjects Effect Mauchly's W Chi-Square df Sig e-Geisser Huynh-Feldt Lower-bound TIME 253 84.022 14 000 666 .707 200 Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to an identity matrix. a. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table. b. Design: Intercept Within Subjects Design: TIMETests of Within-Subjects Effects Measure: MEASURE_1 Type Ill Sum Partial Eta Source of Squares df Mean Square F Sig Squared TIME Sphericity Assumed 52321.740 5 10464.348 105.561 .000 626 Greenhouse-Geisser 52321.740 3.329 157 15.768 105.561 000 .626 Huynh-Feldt 52321.740 3.537 14791.522 105.561 000 626 Lower-bound 52321.740 1.000 52321.740 105.561 000 626 Error(TIME) Sphericity Assumed 31226.260 315 99. 131 Greenhouse-Geisser 31226.260 209.743 148.879 Huynh-Feldt 31226.260 222.849 140.123 Lower-bound 31226.260 63.000 495.655 Tests of Within-Subjects Contrasts Measure: MEASURE_1 Type Ill Sum Partial Eta Source TIME of Squares df Mean Square F Sig. Squared TIME Level 2 vs. Level 1 11342.250 11342.250 82.649 000 567 Level 3 vs. Level 1 25720.141 25720.141 118.250 .000 652 Level 4 vs. Level 1 94941.016 94941.016 277.657 000 815 Level 5 vs. Level 1 34225.000 34225.000 148.242 000 702 Level 6 vs. Level 1 11664.000 11664.000 59.568 000 486 Error(TIME) Level 2 vs. Level 1 8645.750 63 137.234 Level 3 vs. Level 1 13702.859 63 217.506 Level 4 vs. Level 1 21541.984 63 341.936 Level 5 vs. Level 1 14545.000 63 230.873 Level 6 vs. Level 1 12336.000 63 195.810