KPerkins-MFT7110-8

.docx

School

North Central University *

*We aren’t endorsed by this school

Course

MISC

Subject

Statistics

Date

Feb 20, 2024

Type

docx

Pages

Uploaded by MagistrateFog3596

Quantitative Analysis in MFT Northcentral University Assignment 8: Conduct ANOVA Using SPSS In this assignment, you will conduct an analysis of variance (ANOVA), a statistical test typically used when there are more than two categories or levels of the independent variable, and the dependent variable is measured at the interval or ratio level. Again, you will use the SPSS Dataset.Adolescent FT.sav file, which contains 13 variables: (a) Participant.ID (Each person’s identification number), (b) Fam.ID (Family identification number), (c) Race (1. African American, 2. European American, 3. Mexican American, 4. Multiracial) (d) Fam.Pos (Family position: 1. Mother, 2. Adolescent) (e) Adol.Gender (Gender of adolescent: 1. Female, 2. Male) (f) Ther.ID (Therapist identification number) (g) Pre.Prob (The adolescent’s presenting problem: 1. Depression, 2. Oppositional Defiant Disorder, 3. Anxiety) (h) PHQ9 (Scores on the Patient Health Questionnaire-9, a brief depression screening measure; Kroenke, Spitzer, & Williams, 2001) (i) Anxiety (Scores on a measure of anxiety developed for this study) (j) FAD (Scores on the Family Assessment Device; item scores are averaged, with higher scores representing perceptions of poorer family functioning; Mansfield, Keitner, & Dealy, 2015) (k) Life.Sat (Scores on a hypothetical measure of life satisfaction) (l) Alliance (Scores on a hypothetical measure of the therapeutic alliance) (m) Drop.Out (Did the family drop out prior to the completion of the seven-session study treatment protocol?) Once again, ANOVA assumes that the observations are independent, so you can only use data from one person in the family (another solution would be to average relational observations, but that is often problematic. For example, suppose scores on a couple satisfaction assessment ranged from 0-100. If you averaged the scores for the members of each couple, a couple with scores of 10 and 90 would have same average score, 50, as a couple with scores of 45 and 55. Can you see why averaging scores is not optimal? Note that summing scores is the mathematical equivalent of averaging scores, and thus is equally problematic). For this assignment, you will evaluate whether there are differences in anxiety levels based on the adolescent’s race or gender. To complete the assignment, address the following: a. State the null and alternative hypotheses (one pair for race and one pair for gender). Assume you do not have sufficient evidence to anticipate the outcome, so factor that into your alternative hypotheses. Null Hypothesis for Gender: There is no differences in anxiety levels between adolescent genders.

Alternative Hypothesis for Gender: There is a difference in anxiety levels based on adolescent gender. Null Hypothesis for Race: There is no difference in anxiety levels based on adolescent Race. Alternative Hypothesis for Race: There is a difference in anxiety levels based on adolescent Race. b. Conduct the univariate ANOVA in SPSS (refer to the SPSS example in the Between Groups with More Than Two Levels of an IV section in Chapter 7 (p. 88) in the Schwartz et al. text). Your dependent variable is Anxiety and your fixed factors are Race and Adolescent Gender . You will not enter anything into the Random Factors, Covariate, or WLS Weight boxes. Note that you will obtain a non- significant interaction term in the output (Race * Adol Gender). Just ignore that line. c. Request a post hoc test (LSD) for Race. Then, in your response, explain why you don’t need to request a post hoc test for gender. Post hoc tests are completed when there is a significant difference between the groups as notated by the ANOVA. Due to one of our variables (Race) having multiple levels, though we are able to identify a significant difference, we are unable to identify where the difference is. Therefor, we must run a post hoc test to determine which levels of Race differ from one another. Our other variable, Gender, only has two levels and thus, when we are able to identify a significant difference in Gender by looking at the standard deviation and means of the only two existing levels and do not need a post hoc test. d. Report the results of the ANOVA (including the post hoc results for race) in APA format (model your language after the example in the Schwartz et al. text). A one-way, between-groups ANOVA was used to analyze the data. Tukeys post hoc comparisons examined differences between groups (p < .05). Race affected level of anxiety in adolescents, F(3,42)=3.25, p < .03, n^2=.19. Adolescents of the European American race (m=3.92, SD=2.92, n=18), differed from adolescents of a multiracial race (m=7.17, SD=3.28, n=14). Gender also affected levels of anxiety in adolescets, F(1,42)=7.3, p < .01, in that female adolescent anxiety (m=6.81, SD=3.50) differed from male adolescent anxiety (m=4.82, SD=2.93). e. Indicate what you learned about these variables after conducting this analysis. From these analyses I learned that gender does in fact play a role in level of anxiety in adolescents, and only certain variables of race do, such as European American, and Multiracial adolescents. Please correct me if I am wrong, but based on the means of anxiety for gender, I believe that our data shows higher levels of anxiety in female adolescents than males. This would mean that we would reject both of our Null Hypotheses, and fail to reject both alternative hypothese.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version