Religion, Political Affiliation, and Traditional Gender Roles

¶ … ANOVA to the statistical assumptions of an ANOVA are as follows:

The principal assumptions of ANOVA are as follows:

There must be statistical independence of the errors, and no correlation between independent variables and error.

There must be normality of the error distribution.

Homogeneity of variances

The dependent variable is a belief that it is better for the man in a couple relationship to go to work and for the woman to stay home to care for the children and the house. The independent variables are: 1) The respondents' income; 2) the highest year of school completed; and 3) the tendency "to think of self as liberal or conservative."

Ho = A belief in the traditional gender roles in a marriage is unrelated to income, education level, or political self-identification as conservative or liberal.

Ha = A belief in the traditional gender roles in a marriage is related to income, education level, or political self-identification as conservative or liberal.

For the ANOVA shows significant main effects for education level, or political self-identification as conservative or liberal (p < .001, C.I. [confidence interval] = 95.0%). The confidence interval is a range of variables for which one can be 95% certain contains the true mean of the population. The null hypothesis is rejected, as the main effects do indicate a relationship between the traditional gender roles in a marriage for education level, or political self-identification as conservative or liberal. Since the three factors are listed in the hypotheses as a relationship based on "or" not "and," the null hypothesis can be rejected.

ONEWAY POLVIEWS EDUC RINCOME BY FEFAM

/MISSING ANALYSIS

/POSTHOC=WALLER (100) ALPHA (0.05).

Oneway

Notes

Output Created

03-AUG-2015 13:19:34

Comments

Input

Data

/Users/gigi/Downloads/gss04student_corrrected-5.sav

Active Dataset

DataSet1

Filter

one>

Weight

one>

Split File

one>

N of Rows in Working Data File

Missing Value Handling

Definition of Missing

User-defined missing values are treated as missing.

Cases Used

Statistics for each analysis are based on cases with no missing data for any variable in the analysis.

Syntax

ONEWAY POLVIEWS EDUC RINCOME BY FEFAM

/MISSING ANALYSIS

/POSTHOC=WALLER (100) ALPHA (0.05).

Resources

Processor Time

00:00:00.02

Elapsed Time

00:00:00.00

[DataSet1] / Users/Downloads/gss04student_corrrected-5.sav

Sum of Squares

df

Mean Square

F

Sig.

THINK OF SELF AS LIBERAL OR CONSERVATIVE

Between Groups

68.386

3

22.795

12.774

.000

Within Groups

1.785

Total

HIGHEST YEAR OF SCHOOL COMPLETED

Between Groups

3

72.889

9.150

.000

Within Groups

7.966

Total

RESPONDENTS INCOME

Between Groups

9.049

3

3.016

.389

.761

Within Groups

2017.856

7.761

Total

2026.905

Post Hoc Tests

Homogeneous Subsets

THINK OF SELF AS LIBERAL OR CONSERVATIVE

Waller-Duncana, b, c

BETTER FOR MAN TO WORK, WOMAN TEND HOME

N

Subset for alpha = 0.05

1

2

3

STRONGLY DISAGREE

84

3.52

DISAGREE

4.15

AGREE

4.59

STRONGLY AGREE

41

4.68

Means for groups in homogeneous subsets are displayed.

a. Uses Harmonic Mean Sample Size = 81.412.

b. The group sizes are unequal. The harmonic mean of the group sizes is used. Type I error levels are not guaranteed.

c. Type 1/Type 2 Error Seriousness Ratio = 100.

HIGHEST YEAR OF SCHOOL COMPLETED

Waller-Duncana, b, c

BETTER FOR MAN TO WORK, WOMAN TEND HOME

N

Subset for alpha = 0.05