In: Statistics and Probability
Problem Set 2: The Two-factor ANOVA for Independent Measures
Research Scenario: In response to media reports of violence on college campuses, a psychologist who works at a local community college decides to study students’ perceptions of campus safety. He hopes to use these results to help develop an on-campus violence prevention program. The administration has asked him additionally to look at whether perceptions of safety differ depending on students’ year in school and gender. The psychologist administers a questionnaire with possible scores ranging from 1-70, with higher scores indicating higher perceptions of safety on campus, and lower scores indicating perceptions that the campus is less safe. Based on the data collected below, do year in school and/or gender have an effect on perceptions of campus safety?
Using this table, enter the data into a new SPSS data file and run a two-way ANOVA to test whether there is a difference in patients’ depression scores among the three therapists. Create a multiple line graph to show the difference among these scores.
Male |
Freshmen |
Sophomore |
Junior |
Senior |
39 66 54 66 60 |
44 32 62 59 29 |
63 67 46 51 41 |
45 53 68 57 60 |
|
Female |
51 46 45 57 32 |
32 21 30 49 53 |
56 52 60 47 59 |
61 55 42 58 61 |
ANSWER:
Given that,
Descriptive Statistics |
||||
Dependent Variable: Score |
||||
Gender |
Year |
Mean |
Std. Deviation |
N |
Male |
Freshmen |
57.00 |
11.225 |
5 |
sophomore |
45.20 |
15.090 |
5 |
|
Junior |
53.60 |
11.082 |
5 |
|
senior |
56.60 |
8.503 |
5 |
|
Total |
53.10 |
11.801 |
20 |
|
Female |
Freshmen |
46.20 |
9.257 |
5 |
sophomore |
37.00 |
13.509 |
5 |
|
Junior |
54.80 |
5.357 |
5 |
|
senior |
55.40 |
7.893 |
5 |
|
Total |
48.35 |
11.609 |
20 |
|
Total |
Freshmen |
51.60 |
11.247 |
10 |
sophomore |
41.10 |
14.177 |
10 |
|
Junior |
54.20 |
8.230 |
10 |
|
senior |
56.00 |
7.760 |
10 |
|
Total |
50.73 |
11.802 |
40 |
Tests of Between-Subjects Effects |
|||||
Dependent Variable: Score |
|||||
Source |
Type III Sum of Squares |
df |
Mean Square |
F |
Sig. |
Corrected Model |
1799.975a |
7 |
257.139 |
2.266 |
.054 |
Intercept |
102921.025 |
1 |
102921.025 |
906.793 |
.000 |
Gender |
225.625 |
1 |
225.625 |
1.988 |
.168 |
Year |
1333.075 |
3 |
444.358 |
3.915 |
.017 |
Gender * Year |
241.275 |
3 |
80.425 |
.709 |
.554 |
Error |
3632.000 |
32 |
113.500 |
||
Total |
108353.000 |
40 |
|||
Corrected Total |
5431.975 |
39 |
|||
a. R Squared = .331 (Adjusted R Squared = .185) |
Result:
A two-way analysis of variance yielded a main effect for the gender, F(1, 32) = 1.99, p=.168, such that the average score was not significant. The main effect of year was significant, F(3,32) = 3.92, p=.017. However, the interaction effect was not significant, F(3,32) = 0.71, p = .55.