Ψ Psy Midterm Part In lab (5 points total) Your professor decides that he wants to find out how much impact amount of study time has on the first midterm. He randomly assigns students to study for hours, 8 hours, 6 hours, hours and hours; recording each student s midterm grade. Results are shown below. hours 8 hours 6 hours hours hours 8 57 67 67 7 7 88 69 6 97 68 55 7 7 7 6 5 89 67 6 9 9 7 8 65 8 6 66 6 9 86 59 7 Mean 8.8 6. 6...7 SD.87 7.8.98 8.5 8. SPSS output for midterm study UNIANOVA score BY stdytime /CONTRAST (stdytime)=special ( - - - -) /METHOD = SSTYPE() /INTERCEPT = INCLUDE /PRINT = ETASQ HOMOGENEITY /CRITERIA = ALPHA(.5) /DESIGN = stdytime. Between-Subjects Factors STDYTIME 5 Value Label N hours 7 8 hours 7 6 hours 7 hours 7 hours 7 Lev ene's Test of Equality of Error Variances a Dependent Variable: SCORE F df df.. Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept+STDYTIME
Ψ Dependent Variable: SCORE Source Corrected Model Intercept STDYTIME Error Corrected Tests of Between-Subjects Effects Type III Sum Partial Eta of Squares df Mean Square F Squared 875.686 a 688.7 5.76..878 899. 899. 97.85..97 875.686 688.7 5.76..878 66. 87. 688. 5 68.686 a. R Squared =.878 (Adjusted R Squared =.86) Custom Hypothesis Tests Contrast Results (K Matrix) STDYTIME Special Contrast Comp Comp Comp Comp Contrast Estimate Hypothesized Value Difference (Estimate - Hypothesized) Std. Error 95% Confidence Interval for Difference Contrast Estimate Hypothesized Value Lower Bound Upper Bound Difference (Estimate - Hypothesized) Std. Error 95% Confidence Interval for Difference Contrast Estimate Hypothesized Value Lower Bound Upper Bound Difference (Estimate - Hypothesized) Std. Error 95% Confidence Interval for Difference Contrast Estimate Hypothesized Value Lower Bound Upper Bound Difference (Estimate - Hypothesized) Dependent Variable SCORE 6.857 6.857.99. 6.66 7.5 8.57 8.57.99. 8.78 8.765 5.57 5.57.99.7 -.6 5.765.57.57 Std. Error 95% Confidence Interval for Difference Lower Bound Upper Bound.99.7.78.765
Ψ Questions referring to the Midterm Experiment. Do the five groups meet the homogeneity of variance assumption? How do you know? ( points). Does amount of study time affect midterm scores? How do you know? ( points). Are the comparisons orthogonal? Show how you came to your conclusion. ( points). As a planned comparison, does studying for hours improve your score when compared to only hours? Explain your answer. ( point) 5. Is hours of study significantly different than hours of study after a Tukey adjustment? Show your work. ( points)
Ψ A research is interest in whether different stats courses offered at CSUN aversely affect quality of life for students enrolled. The researcher randomly selected 5 students from each of the following courses: Psy, Psy 5 and Psy 5. Results and layout for a regression analysis are listed below, scores are on a scale of to with meaning better quality of life. 5 5 y x x 9 - - 8 - - 8 - - 8 - - 7 - - 6-7 - 7-8 - 6 - Output for Stat Class Study Variables Entered/Remov ed b Model Variables Variables Entered Removed Method X, X a. Enter a. All requested variables entered. b. Dependent Variable: Y Model Summary Model Std. Error of R R Square Adjusted R Square the Estimate.96 a.9.98.68 a. Predictors: (Constant), X, X Model Regression Residual a. Predictors: (Constant), X, X b. Dependent Variable: Y ANOVA b Sum of Squares df Mean Square F 7. 7.67 79.9. a 5.6.67 79.7
Ψ Coefficients a Model (Constant) X X a. Dependent Variable: Y Unstandardized Coefficients Standardized Coefficients B Std. Error Beta t 5.867.76.6. -.5.5 -.9 -.9. -.6.6 -. -.777.7 Questions related to the Stat Course Study 6. Does quality of life differ for the three statistics courses? Explain. ( points) 7. What is the η value for the effect of statistics classes? ( points) 8. What is the predicted score for the first subject in the course? Show how you got the answer. ( points) 9. Is quality of life statistically worse for students in the 5 course when compared to the course? How do you know? ( points). How do you interpret the B for X (-.5)? The constant (5.867)? ( points)
PRT Autism DTT FT PRT Pervasive DD DTT FT PRT Aspergers DTT FT Name Ψ. You are an experimenter trying to test the effect of different disorders (Aspergers, Pervasive Developmental Disorder, Autism) and different types of behavioral therapy (Floor time, Discrete Trials, Pivotal Response Training) on length of eye contact of each child (measured in seconds). 9 children with each disorder were randomly assigned to one of the three treatments (7 subjects total). Set up the chart below to do an ANOVA through regression for this data; just set it up, do not proceed to the analysis ( points) A B Y Sum 59 Sum Sq 6
Output for the Disorders by Treatment study Name Ψ Between-Subjects Factors DISORDER TREATMNT...... Value Label N Aspergers 9 Pervasive Development al Disorder Autism 9 Floortime 9 Discrete Trial Training Pivotal Response Training 9 9 9 Descriptiv e Statistics Dependent Variable: Y DISORDER Aspergers Pervasive Developmental Disorder Autism TREATMNT Floortime Discrete Trial Training Pivotal Response Training Floortime Discrete Trial Training Pivotal Response Training Floortime Discrete Trial Training Pivotal Response Training Floortime Discrete Trial Training Pivotal Response Training Mean Std. Deviation N.6667.5775..5775.6667.5775..978 9.6667.5775...6667.5775.7778.99 9.6667.5775.....5556.9 9.. 9..69 9..575 9.85.79 7 Lev ene's Test of Equality of Error Variances a Dependent Variable: Y F df df. 8 8.59 Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept+DISORDER+TREATMNT+DISORDER * TREATMNT
Dependent Variable: Y Source Corrected Model Intercept DISORDER TREATMNT DISORDER * TREATMNT Error Corrected Tests of Between-Subjects Effects Name Ψ Type III Sum Partial Eta of Squares df Mean Square F Squared 8.7 a 8.59.58..8 8.96 8.96 86.778..956 6.7.7...59.96 6.8 9...68 8.7.9 6.78..58 6. 8. 6. 7.7 6 a. R Squared =.8 (Adjusted R Squared =.76) Questions related to the Disorders by Treatment Study. There is a significant interaction, draw a graph (using the grid below) that illustrates the nature of the interaction above (5 points). The effect size for treatment is.68, how did the computer calculate that number? ( points). Given the significant effects, what type of follow up comparisons should be performed (no computations, just tell me what it/they should be) ( points)