Psy 40 Midterm Part (Version A) In lab (50 points total)
A researcher wants to know if memory is improved by repetition (Duh!). So he shows a group of five participants a list of 0 words at for different time points. At each time point the participants are given a free recall task. The number of words recalled at each trial is recorded below for each participant. Memory over time study Within-Subjects Factors Measure: MEASURE_ TIME 4 Dependent Variable TIME TIME TIME TIME4 Time Time Time Time 4 S 0 8 0 6 S 6 4 6 S 0 8 9 8 S 4 6 4 4 0 S 5 9 Descriptive Statistics TIME TIME TIME TIME4 Mean Std. Deviation N.0000.00000 5.0000.00000 5.0000.9548 5 8.8000.954 5 Tests of Within-Subjects Effects Measure: MEASURE_ TIME Error(TIME) 68.400 09.467 76.0.000.986 68.400.704 68.875 76.0.000.986 68.400.88.407 76.0.000.986 68.400.000 68.400 76.0.000.986 9.00.758 9.00 6.84.5 9.00.5.80 9.00 4.000.75
Tests of Within-Subjects Contrasts Measure: MEASURE_ TIME Error(TIME) TIME Linear Quadratic Cubic Linear Quadratic Cubic 585.640 585.640 55.8.000.99.800.800.667.78.400 40.960 40.960 84.454.00.955 4.460 4.5.700 4.675.940 4.485 Measure: MEASURE_ Transformed Variable: Average Intercept Error Tests of Between-Subjects Effects 8988.800 8988.800.89.000.988.700 4 7.95. Based on the output is (Time and Time 4) vs. (Time and Time ) a significant comparison? How do you know (it s right in front of you so think about it)? And how much percent of variance is accounted for by this comparison? No calculations involved. ( points). Perform a comparison of (Time and Time ) vs. (Time and Time 4) by hand and test it for significance. Show all work. (6 points)
The pharmaceutical companies are trying to figure out which is the best medication to treat erectile dysfunction (ED), Viagra (Drug ) and Levitra (Drug ). An independent company conducts clinical trials to test the treatments using three trials per treatment with 0 participants. Scores indicate the number of hours that the treatment lasted. Viagra Levitra T T T T T T S 4 5 9 S 7 9 8 4 S 6 9 9 4 S 4 6 0 0 9 4 S 5 8 6 S 6 7 0 0 0 6 S 7 6 0 8 8 S 8 7 6 S 9 9 4 4 8 S 0 5 7 4 9 Output for Drug by Trial Study Within-Subjects Factors Measure: MEASURE_ DRUG TRIAL Dependent Variable VT VT VT LT LT LT Measure: MEASURE_ Within Subjects Effect DRUG TRIAL DRUG * TRIAL Mauchly's Test of Sphericity b Epsilon a Approx. Greenhous Mauchly's W Chi-Square df Sig. e-geisser.000.000 0..000.000.000.546 4.840.089.688.77.500.48 6.785.04.66.694.500 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: DRUG+TRIAL+DRUG*TRIAL
Tests of Within-Subjects Effects Measure: MEASURE_ DRUG Error(DRUG) TRIAL Error(TRIAL) DRUG * TRIAL Error(DRUG*TRIAL) 984.50 984.50 5.800.000.99 984.50.000 984.50 5.800.000.99 984.50.000 984.50 5.800.000.99 984.50.000 984.50 5.800.000.99 7.68 9.854 7.68 9.000.854 7.68 9.000.854 7.68 9.000.854 46.900.450 95..000.94 46.900.76 4.094 95..000.94 46.900.54 0.47 95..000.94 46.900.000 46.900 95..000.94 4.4 8.46 4.4.8.58 4.4.877.9 4.4 9.000.49 5.00 7.650 94.60.000.978 5.00.7 84.99 94.60.000.978 5.00.88 69.546 94.60.000.978 5.00.000 5.00 94.60.000.978 5.67 8.98 5.67.45.469 5.67.490.40 5.67 9.000.596 Tests of Within-Subjects Contrasts Measure: MEASURE_ DRUG TRIAL SS df MS F Sig. Squared TRIAL Trial vs. Trial 68.450 68.450 0.86.000.99 Trial vs. Trial 7.00 7.00.04.000.96 Error(TRIAL) Trial vs. Trial 6.050 9.67 Trial vs. Trial 5.800 9.644 Measure: MEASURE_ Transformed Variable: Average Intercept Error Tests of Between-Subjects Effects 5088.050 5088.050 95.45.000.970 55.006 9 7.
Estimated Marginal Means Measure: MEASURE_ DRUG. DRUG.900.985 9.67 4.9 0.000.88 8.00.998 Measure: MEASURE_ TRIAL. TRIAL 5.50.9.44 7.456 7.00.904 5.54 9.46 5.00.96.6 7.474 Measure: MEASURE_ DRUG TRIAL. DRUG * TRIAL 8.900.06 6.60.98.00.94.009 5.9.700.044.8 6.06.800.867 9.89.76.00.895 9.75.5 6.900.9 4.86 8.964 Profile Plots 4 Estimated Marginal Means of MEASURE_ 0 8 6 4 DRUG 0 8 TRIAL
. Did we meet the sphericity assumption for all variables? How do you know? Why isn t sphericity test for the Drug variable? ( points) 4. Create an ANOVA summary table that looks a little more like we re used to. Include all sources of variance. ( points) SS DF MS F Total 5. Why is there a main effect for trial (where are the differences?)? When describing it include the directionality of the comparison. (4 points). 6. Base on the line graph why is there a significant interaction (in words)? ( points)
7. Based on the significant effects what other follow-up analysis(es) would you perform (in words)? ( point) An experimenter is doing a study where variable A is a between groups variable and variables B and C are within subjects variables. Results are shown below. b b A B C study Within-Subjects Factors Measure: MEASURE_ B C Dependent Variable BC BC BC BC a a a c c c c S 5 9 4 S 9 4 6 S 7 9 S 4 5 7 0 S 5 6 8 0 S 6 4 8 0 S 7 8 5 7 S 8 7 0 4 S 9 7 4 4 Between-Subjects Factors A.00.00.00 N
Tests of Within-Subjects Effects Measure: MEASURE_ B B * A Error(B) C C * A Error(C) B * C B * C * A Error(B*C) 40.50 40.50 96.000.000.994 40.50.000 40.50 96.000.000.994 40.50.000 40.50 96.000.000.994 40.50.000 40.50 96.000.000.994.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000.500 6.50.500 6.000.50.500 6.000.50.500 6.000.50 78.08 78.08 65.5.000.965 78.08.000 78.08 65.5.000.965 78.08.000 78.08 65.5.000.965 78.08.000 78.08 65.5.000.965.889.444.94.44.9.889.000.444.94.44.9.889.000.444.94.44.9.889.000.444.94.44.9.8 6.47.8 6.000.47.8 6.000.47.8 6.000.47 6.50 6.50 75.000.000.96 6.50.000 6.50 75.000.000.96 6.50.000 6.50 75.000.000.96 6.50.000 6.50 75.000.000.96.000.000.000.008.800.000.000.000.000.008.800.000.000.000.000.008.800.000.000.000.000.008.800.500 6.08.500 6.000.08.500 6.000.08.500 6.000.08
Measure: MEASURE_ Transformed Variable: Average Intercept A Error Tests of Between-Subjects Effects 8.6 8.6 5.76.000.977 4.889 7.444 5.5.047.640 80.500 6.47 Estimated Marginal Means Measure: MEASURE_ A.00.00.00. A 0.08.057 7.496.67 7.08.057 4.496 9.67.97.057 9.9 4.504 Measure: MEASURE_ B. B 7..64 5.54 8.679.78.59 0.8.7 Measure: MEASURE_ C. C 8..66 6.690 9.754.67.66 9.659.674 Measure: MEASURE_ A.00.00.00 B 4. A * B 7.500.0 4.785 0.5.667.0 0.64 5.70 4.500.0.785 7.5 9.667.0 7.64.70 9..0 6.68.049 4.500.0.997 7.00
Measure: MEASURE_ A.00.00.00 C 5. A * C 8.500.084 5.847.5.667.067 9.055 4.78 5.500.084.847 8.5 8.667.067 6.055.78 0.667.084 8.0.0.67.067 0.555 5.778 Measure: MEASURE_ B C 6. B * C 5..667.59 6.85 9.000.68 7.48 0.56..598 9.758.686..609.844 4.8 7. A * B * C Measure: MEASURE_ A.00.00.00 B C 5.667.55.84 8.49 9..06 6.68.08..06 8.797.869 4.000.054.4 6.579.667.55 -.59 5.49 6..06.68 9.08 8..06 5.797 0.869.000.054 8.4.579 7..55 4.508 0.59..06 8.68 4.08 4.000.06.464 6.56 5.000.054.4 7.579
8. What effects are NOT significant? How do you know? ( points) 9. Compare A to A and test for significance. Show your work. (6 points) 0. Is the A vs. A comparison significant after a Tukey adjustment? Show your work. ( points)
. Based on the Tukey test you just performed (without any further computations), would the A vs. A comparison be significant after a Scheffe adjustment. How do you know? ( point). Graph the A x C interaction. ( points)