PSYC 355 SPSS HOMEWORK 3 INSTRUCTIONS: ONE-WAY ANOVA Part 1: Green & Salkind: Lesson 25 - Exercises 1-3 The following helpful tips are numbered to correspond with the exercise number to which they refer: 1. Use Tukey's test as the post hoc test for ANOVAs in this course. Be sure to check this box when you run analyses. For letters a-d, instead of identifying these values on your output, as the text states, please write them into your Word file as written answers for #1 a, b, c and d. (2 pts for output; 2 pts each for a-d) 2. The effect size is found in the SPSS output. Review the section on effect size statistics in this lesson. (3 pts) 3. Remember to put your dependent variable in the \"variable\" box and your independent, or grouping, variables in the \"category axis\" box. (3 pts) Part 2: 1. Twenty-four adults who have been diagnosed with Borderline Personality Disorder (BPD) are randomly assigned to 1 of 3 group therapy conditions in order to improve their interpersonal skills: dialectical behavior therapy (DBT), general cognitive-behavioral therapy (CBT), and talk therapy (T). Following 3 months of therapy, the participants are assessed on a standardized measure of interpersonal skills. On this instrument, scores range from 0 to 60, and higher scores indicate better or improved interpersonal skills, while lower scores indicate interpersonal skills that need improvement. These scores are shown in the table below. Conduct a one-way ANOVA to determine how effective these 3 therapy conditions are at improving interpersonal skills in patients with BPD. The steps will be the same as the ones you have been practicing in Part 1 of the assignmentthe only difference is that you are now responsible for creating the data file as well. Remember to name and define your variables under the \"Variable View,\" then return to the \"Data View\" to enter the data. (3 pts) DBT 30, 22, 26, 33, 33, 37, 19, 28 CBT 19, 21, 28, 16, 20, 27, 29, 30 T 15, 17, 24, 16, 20, 28, 27, 24 2. What is the F ratio for the therapy group main effect? (3 pts) Page 1 of 2 PSYC 355 3. What is the effect size for the overall effect of therapy type on interpersonal skills scores? According to general conventions, is this effect small, medium, or large? (See the \"Effect Size Statistics\" section in Lesson 25 for conventions.) (3 pts) 4. Write a current APA-style Results section based on your analysis. All homework Results sections must follow the example given in the SPSS tutorials and the Course Content document \"Writing Results of Statistical Tests in Current APA Format\" (note: you do not have to refer to a figure). Remember to include a decision about the null hypothesis. (3 pts) Part 3: Cumulative Homework 1. A health psychologist in a northern climate wants to evaluate the claim that UV lamps help lower depressive symptoms in middle-aged women. She recruits volunteers who meet the criteria for clinical depression and assigns them to 2 groups: one group receives a standard treatment for depression and undergoes a half hour of UV lamp therapy each day; the other group receives the same standard treatment for depression but without UV lamp therapy. At the end of 2 months, she administers a depression inventory where lower scores indicate fewer depressive symptoms (lower levels of depression). Assume all other variables are controlled for in the study. Choose the correct test to evaluate the claim that depression treatment plus the UV lamp results in lower depression scores than depression treatment alone. Depression Treatment + UV 34 14 29 31 43 17 31 22 29 19 25 27 12 32 Depression Treatment Only 39 31 29 25 13 14 41 24 26 37 23 41 47 42 a) Paste the appropriate SPSS output. (4 pts) b) Paste the appropriate SPSS graph. (4 pts) c) Write a current APA-style Results section based on your analyses. All homework Results sections must follow the example given in the SPSS tutorials and the Course Content document \"Writing Results of Statistical Tests in Current APA Format\" (note: you do not have to refer to a figure). Remember to include a decision about the null hypothesis. (4 pts) This assignment is due by 11:59 p.m. (ET) on Monday of Module/Week 3. Page 2 of 2 PSYC 355 SPSS HOMEWORK 3 INSTRUCTIONS: ONE-WAY ANOVA Part 1: Green & Salkind: Lesson 25 - Exercises 1-3 The following helpful tips are numbered to correspond with the exercise number to which they refer: 1. Use Tukey's test as the post hoc test for ANOVAs in this course. Be sure to check this box when you run analyses. For letters a-d, instead of identifying these values on your output, as the text states, please write them into your Word file as written answers for #1 a, b, c and d. (2 pts for output; 2 pts each for a-d) 2. The effect size is found in the SPSS output. Review the section on effect size statistics in this lesson. (3 pts) 3. Remember to put your dependent variable in the \"variable\" box and your independent, or grouping, variables in the \"category axis\" box. (3 pts) #1. a) SPSS output: Descriptives Social Extroversion 95% Confidence Interval for Mean Mea N n Std. Deviation Std. Error Lower Bound Upper Bound Minimum Maximum Blond 6 5.17 2.787 1.138 2.24 8.09 2 10 Brunet 6 3.67 1.211 .494 2.40 4.94 2 5 Redhea 6 2.33 1.033 .422 1.25 3.42 1 4 18 3.72 2.109 .497 2.67 4.77 1 10 d Total ANOVA Social Extroversion Sum of Squares df Mean Square Between Groups 24.111 2 12.056 Within Groups 51.500 15 75.611 3.511 Sig. 3.433 Total F .056 17 Page 1 of 9 PSYC 355 Multiple Comparisons Social Extroversion Tukey HSD 95% Confidence Interval Lower (I) Hair Color (J) Hair Color Mean Difference (I-J) Std. Error Sig. Blond Bound Upper Bound 1.500 1.070 .365 -1.28 4.28 Redhead 2.833* 1.070 .045 .05 5.61 Blond -1.500 1.070 .365 -4.28 1.28 1.333 1.070 .446 -1.45 4.11 Blond -2.833* 1.070 .045 -5.61 -.05 Brunet Brunet Brunet -1.333 1.070 .446 -4.11 1.45 Redhead Redhead *. The mean difference is significant at the 0.05 level. b) F-ratio: ANOVA Social Extroversion Sum of Squares df Mean Square Between Groups 24.111 2 12.056 Within Groups 51.500 15 75.611 3.511 Sig. 3.433 Total F .056 17 The F ratio from the table is 3.511. C) Effect size: Effect size is = 24.11/75.611 = 0.318 This is above 0.2 , so the effect size is medium according to cohen's guidelines for social sciences. Page 2 of 9 PSYC 355 d) The oneway anova showed the averages for three hair colors Blonde, Brunet, Redhead are 5.17, 3.67, 2.33 respectively. The significant main effect of harir color on social extreoversion is F(2,15)=3.511, 'p' value is 0.056. since 'p' value is greater than 0.05 we conclude that there is no significant of hari color. The post-hoc Tukey's test showed that the p-value of the difference of scores between Blonde and Redhead is 0.045. This means the difference in scores between these two groups are significant. And for the remaining groups, the significant value is greater than 0.05. Hence we conclude that the difference in scores are not significant. SPSS graph: c) Part 2: 1. Twenty-four adults who have been diagnosed with Borderline Personality Disorder (BPD) are randomly assigned to 1 of 3 group therapy conditions in order to improve their interpersonal skills: dialectical behavior therapy (DBT), general cognitive-behavioral therapy (CBT), and talk therapy (T). Following 3 months of therapy, the participants are Page 3 of 9 PSYC 355 assessed on a standardized measure of interpersonal skills. On this instrument, scores range from 0 to 60, and higher scores indicate better or improved interpersonal skills, while lower scores indicate interpersonal skills that need improvement. These scores are shown in the table below. Conduct a one-way ANOVA to determine how effective these 3 therapy conditions are at improving interpersonal skills in patients with BPD. The steps will be the same as the ones you have been practicing in Part 1 of the assignmentthe only difference is that you are now responsible for creating the data file as well. Remember to name and define your variables under the \"Variable View,\" then return to the \"Data View\" to enter the data. (3 pts) DBT 30, 22, 26, 33, 33, 37, 19, 28 CBT 19, 21, 28, 16, 20, 27, 29, 30 T 15, 17, 24, 16, 20, 28, 27, 24 1) SPSS output: Descriptives Scores_Interpersonal_skills 95% Confidence Interval for Mean N Mean Std. Deviation Std. Error Lower Bound Upper Bound Minimum Maximum DBT 8 28.5000 6.02376 2.12972 23.4640 33.5360 19.00 37.00 CBT 8 23.7500 5.33854 1.88746 19.2869 28.2131 16.00 30.00 T 8 21.3750 5.06916 1.79222 17.1371 25.6129 15.00 28.00 Total 24 24.5417 6.05754 1.23649 21.9838 27.0995 15.00 37.00 Test of Homogeneity of Variances Scores_Interpersonal_skills Levene Statistic .066 df1 df2 Sig. 2 21 .936 Page 4 of 9 PSYC 355 ANOVA Scores_Interpersonal_skills Mean Sum of Squares df Between Square 210.583 2 F Sig. 105.292 3.49 .049 Groups 1 Within Groups 633.375 21 Total 30.161 843.958 23 Multiple Comparisons Scores_Interpersonal_skills Tukey HSD 95% Confidence Interval (I) Group (J) Group Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound DBT 4.75000 2.74594 .218 -2.1713 11.6713 T 7.12500* 2.74594 .043 .2037 14.0463 DBT -4.75000 2.74594 .218 -11.6713 2.1713 2.37500 2.74594 .668 -4.5463 9.2963 DBT -7.12500* 2.74594 .043 -14.0463 -.2037 CBT CBT CBT -2.37500 -9.2963 4.5463 T T 2.74594 .668 *. The mean difference is significant at the 0.05 level. 2. What is the F ratio for the therapy group main effect? (3 pts) Page 5 of 9 PSYC 355 ANOVA Scores_Interpersonal_skills Mean Sum of Squares df Between 210.583 2 Square F Sig. 105.292 3.49 .049 Groups 1 Within Groups 633.375 21 Total 30.161 843.958 23 The F ratio from the table is 3.491. 3. What is the effect size for the overall effect of therapy type on interpersonal skills scores? According to general conventions, is this effect small, medium, or large? (See the \"Effect Size Statistics\" section in Lesson 25 for conventions.) (3 pts) Effect size is = 210.583/843.958 = 0.2495 This is above 0.2 , so the effect size is medium according to cohen's guidelines for social sciences. 4. Write a current APA-style Results section based on your analysis. All homework Results sections must follow the example given in the SPSS tutorials and the Course Content document \"Writing Results of Statistical Tests in Current APA Format\" (note: you do not have to refer to a figure). Remember to include a decision about the null hypothesis. (3 pts) The oneway anova showed the averages for three groups DBT, CBT and T are 28.5, 23.75, 21.375 respectively. The significant main effect for therapy type on interpersonal skills score of 24 adults is F(2,21)=3.491, 'p' value is 0.049. since 'p' value is less than 0.05 we conclude that the scores are significant. The post-hoc Tukey's test showed that the p-value of the difference of scores between DBT and T is 0.043. This means the difference in scores between these two groups are significant. And for the remaining groups, the significant value is greater than 0.05. Hence we conclude that the difference in scores are not significant. Part 3: Cumulative Homework Page 6 of 9 PSYC 355 1. A health psychologist in a northern climate wants to evaluate the claim that UV lamps help lower depressive symptoms in middle-aged women. She recruits volunteers who meet the criteria for clinical depression and assigns them to 2 groups: one group receives a standard treatment for depression and undergoes a half hour of UV lamp therapy each day; the other group receives the same standard treatment for depression but without UV lamp therapy. At the end of 2 months, she administers a depression inventory where lower scores indicate fewer depressive symptoms (lower levels of depression). Assume all other variables are controlled for in the study. Choose the correct test to evaluate the claim that depression treatment plus the UV lamp results in lower depression scores than depression treatment alone. Depression Treatment + UV 34 14 29 31 43 17 31 22 29 19 25 27 12 32 Depression Treatment Only 39 31 29 25 13 14 41 24 26 37 23 41 47 42 a) Paste the appropriate SPSS output. (4 pts) Group Statistics Depression_tr eatement Depression_treatment_score UV without UV N Mean Std. Deviation Std. Error Mean 14 26.0714 8.52566 2.27858 14 30.8571 10.62549 2.83978 Page 7 of 9 PSYC 355 Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. (2F Depression_treatment_score Equal variances Sig. 1.174 .288 t df - Mean Std. Error tailed) Difference Difference 26 .200 -4.78571 3.64092 1.314 Difference Lower Upper - 2.69830 12.26973 assumed Equal variances - 24.834 1.314 .201 -4.78571 3.64092 - 2.71544 12.28687 not assumed b) Paste the appropriate SPSS graph. (4 pts) c) Write a current APA-style Results section based on your analyses. All homework Results sections must follow the example given in the SPSS tutorials and the Course Content document \"Writing Results of Statistical Tests in Current APA Page 8 of 9 PSYC 355 Format\" (note: you do not have to refer to a figure). Remember to include a decision about the null hypothesis. (4 pts) The two sample independent groups t-test is conducted to test whether the UV lamp results in lower depression scores than normal lamp. The results indicate that UV lamp results average score lower than the depression treatment alone. (M1=26.0714, M2=30.8571 S1=8.52566,S2=10.62549). The standard deviation of UV lamp scores is less than the treatment alone. The differenence in average scores is (-4.78571) t(26)=1.314,p=.2. This is one left tailed test so the p-value is 0.2/2=0.1 Since p-value (0.1) is greater than 0.05 we accept the null hypothesis at 0.05 level. So we conclude that that there is no signifacant difference in mean scores. This assignment is due by 11:59 p.m. (ET) on Monday of Module/Week 3. Page 9 of 9 PSYC 355 SPSS HOMEWORK 3 INSTRUCTIONS: ONE-WAY ANOVA Part 1: Green & Salkind: Lesson 25 - Exercises 1-3 The following helpful tips are numbered to correspond with the exercise number to which they refer: 1. Use Tukey's test as the post hoc test for ANOVAs in this course. Be sure to check this box when you run analyses. For letters a-d, instead of identifying these values on your output, as the text states, please write them into your Word file as written answers for #1 a, b, c and d. (2 pts for output; 2 pts each for a-d) 2. The effect size is found in the SPSS output. Review the section on effect size statistics in this lesson. (3 pts) 3. Remember to put your dependent variable in the \"variable\" box and your independent, or grouping, variables in the \"category axis\" box. (3 pts) #1. a) SPSS output: Descriptives Social Extroversion 95% Confidence Interval for Mean N Mean Std. Deviation Std. Error Lower Bound Upper Bound Minimum Maximum Blond 6 5.17 2.787 1.138 2.24 8.09 2 10 Brunet 6 3.67 1.211 .494 2.40 4.94 2 5 Redhead 6 2.33 1.033 .422 1.25 3.42 1 4 Total 2.109 .497 2.67 4.77 1 10 18 3.72 ANOVA Social Extroversion Sum of Squares df Mean Square Between Groups 24.111 2 12.056 Within Groups 51.500 15 75.611 3.511 Sig. 3.433 Total F .056 17 Page 1 of 8 PSYC 355 Multiple Comparisons Social Extroversion Tukey HSD 95% Confidence Interval (I) Hair Color (J) Hair Color Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound Blond 1.500 1.070 .365 -1.28 4.28 Redhead 2.833 * 1.070 .045 .05 5.61 Blond -1.500 1.070 .365 -4.28 1.28 1.333 1.070 .446 -1.45 4.11 Blond -2.833 * 1.070 .045 -5.61 -.05 Brunet Brunet Brunet -1.333 1.070 .446 -4.11 1.45 Redhead Redhead *. The mean difference is significant at the 0.05 level. b) F-ratio: ANOVA Social Extroversion Sum of Squares df Mean Square Between Groups 24.111 2 12.056 Within Groups 51.500 15 75.611 3.511 Sig. 3.433 Total F .056 17 The F ratio from the table is 3.511. C) Effect size: Effect size is = 24.11/75.611 = 0.318 This is above 0.2 , so the effect size is medium according to cohen's guidelines for social sciences. Page 2 of 8 PSYC 355 d) The oneway anova showed the averages for three hair colors Blonde, Brunet, Redhead are 5.17, 3.67, 2.33 respectively. The significant main effect of harir color on social extreoversion is F(2,15)=3.511, 'p' value is 0.056. since 'p' value is greater than 0.05 we conclude that there is no significant of hari color. The post-hoc Tukey's test showed that the p-value of the difference of scores between Blonde and Redhead is 0.045. This means the difference in scores between these two groups are significant. And for the remaining groups, the significant value is greater than 0.05. Hence we conclude that the difference in scores are not significant. SPSS graph: c) Part 2: 1. Twenty-four adults who have been diagnosed with Borderline Personality Disorder (BPD) are randomly assigned to 1 of 3 group therapy conditions in order to improve their interpersonal skills: dialectical behavior therapy (DBT), general cognitive-behavioral therapy (CBT), and talk therapy (T). Following 3 months of therapy, the participants are assessed on a standardized measure of interpersonal skills. On this instrument, scores Page 3 of 8 PSYC 355 range from 0 to 60, and higher scores indicate better or improved interpersonal skills, while lower scores indicate interpersonal skills that need improvement. These scores are shown in the table below. Conduct a one-way ANOVA to determine how effective these 3 therapy conditions are at improving interpersonal skills in patients with BPD. The steps will be the same as the ones you have been practicing in Part 1 of the assignmentthe only difference is that you are now responsible for creating the data file as well. Remember to name and define your variables under the \"Variable View,\" then return to the \"Data View\" to enter the data. (3 pts) DBT 30, 22, 26, 33, 33, 37, 19, 28 CBT 19, 21, 28, 16, 20, 27, 29, 30 T 15, 17, 24, 16, 20, 28, 27, 24 1) SPSS output: Descriptives Scores_Interpersonal_skills 95% Confidence Interval for Mean N Mean Std. Deviation Std. Error Lower Bound Upper Bound Minimum Maximum DBT 8 28.5000 6.02376 2.12972 23.4640 33.5360 19.00 37.00 CBT 8 23.7500 5.33854 1.88746 19.2869 28.2131 16.00 30.00 T 8 21.3750 5.06916 1.79222 17.1371 25.6129 15.00 28.00 Total 24 24.5417 6.05754 1.23649 21.9838 27.0995 15.00 37.00 Test of Homogeneity of Variances Scores_Interpersonal_skills Levene Statistic df1 df2 Sig. .066 2 21 .936 ANOVA Scores_Interpersonal_skills Page 4 of 8 PSYC 355 Sum of Squares df Mean Square Between Groups 210.583 2 Within Groups 633.375 21 Total F Sig. 105.292 3.491 .049 843.958 23 30.161 Multiple Comparisons Scores_Interpersonal_skills Tukey HSD 95% Confidence Interval (I) Group (J) Group Mean Difference (I-J) Std. Error Sig. Lower Bound Upper Bound DBT 4.75000 2.74594 .218 T CBT CBT 7.12500 DBT T T -2.1713 11.6713 2.74594 .043 .2037 14.0463 -4.75000 2.74594 .218 -11.6713 2.1713 2.37500 2.74594 .668 -4.5463 9.2963 * * DBT -7.12500 2.74594 .043 -14.0463 -.2037 CBT -2.37500 2.74594 .668 -9.2963 4.5463 *. The mean difference is significant at the 0.05 level. 2. What is the F ratio for the therapy group main effect? (3 pts) ANOVA Scores_Interpersonal_skills Sum of Squares df Mean Square Between Groups 210.583 2 Within Groups 633.375 21 Total F Sig. 105.292 3.491 .049 843.958 23 30.161 The F ratio from the table is 3.491. Page 5 of 8 PSYC 355 3. What is the effect size for the overall effect of therapy type on interpersonal skills scores? According to general conventions, is this effect small, medium, or large? (See the \"Effect Size Statistics\" section in Lesson 25 for conventions.) (3 pts) Effect size is = 210.583/843.958 = 0.2495 This is above 0.2 , so the effect size is medium according to cohen's guidelines for social sciences. 4. Write a current APA-style Results section based on your analysis. All homework Results sections must follow the example given in the SPSS tutorials and the Course Content document \"Writing Results of Statistical Tests in Current APA Format\" (note: you do not have to refer to a figure). Remember to include a decision about the null hypothesis. (3 pts) The oneway anova showed the averages for three groups DBT, CBT and T are 28.5, 23.75, 21.375 respectively. The significant main effect for therapy type on interpersonal skills score of 24 adults is F(2,21)=3.491, 'p' value is 0.049. since 'p' value is less than 0.05 we conclude that the scores are significant. The post-hoc Tukey's test showed that the p-value of the difference of scores between DBT and T is 0.043. This means the difference in scores between these two groups are significant. And for the remaining groups, the significant value is greater than 0.05. Hence we conclude that the difference in scores are not significant. Part 3: Cumulative Homework 1. A health psychologist in a northern climate wants to evaluate the claim that UV lamps help lower depressive symptoms in middle-aged women. She recruits volunteers who meet the criteria for clinical depression and assigns them to 2 groups: one group receives a standard treatment for depression and undergoes a half hour of UV lamp therapy each day; the other group receives the same standard treatment for depression but without UV lamp therapy. At the end of 2 months, she administers a depression inventory where lower scores indicate fewer depressive symptoms (lower levels of depression). Assume all other variables are controlled for in the study. Choose the correct test to evaluate the claim that depression treatment plus the UV lamp results in lower depression scores than depression treatment alone. Depression Treatment + UV 34 14 29 31 Depression Treatment Only 39 31 29 25 Page 6 of 8 PSYC 355 43 31 29 25 12 17 22 19 27 32 13 41 26 23 47 14 24 37 41 42 a) Paste the appropriate SPSS output. (4 pts) Group Statistics Depression_tr eatement N Mean Depression_treatment_score UV Std. Deviation Std. Error Mean 14 8.52566 2.27858 14 without UV 26.0714 30.8571 10.62549 2.83978 Independent Samples Test Levene's Test for Equality of Variances t-test for Equality of Means 95% Confidence Interval of the Sig. (2F Depression_treatment_score Equal variances Sig. 1.174 .288 t df - Mean Std. Error tailed) Difference Difference 26 .200 -4.78571 3.64092 1.314 Difference Lower Upper - 2.69830 12.26973 assumed Equal variances - 24.834 1.314 .201 -4.78571 3.64092 - 2.71544 12.28687 not assumed b) Paste the appropriate SPSS graph. (4 pts) Page 7 of 8 PSYC 355 c) Write a current APA-style Results section based on your analyses. All homework Results sections must follow the example given in the SPSS tutorials and the Course Content document \"Writing Results of Statistical Tests in Current APA Format\" (note: you do not have to refer to a figure). Remember to include a decision about the null hypothesis. (4 pts) The two sample independent groups t-test is conducted to test whether the UV lamp results in lower depression scores than normal lamp. The results indicate that UV lamp results average score lower than the depression treatment alone. (M1=26.0714, M2=30.8571 S1=8.52566,S2=10.62549). The standard deviation of UV lamp scores is less than the treatment alone. The differenence in average scores is (-4.78571) t(26)=1.314,p=.2. This is one left tailed test so the p-value is 0.2/2=0.1 Since p-value (0.1) is greater than 0.05 we accept the null hypothesis at 0.05 level. So we conclude that that there is no signifacant difference in mean scores. This assignment is due by 11:59 p.m. (ET) on Monday of Module/Week 3. Page 8 of 8