Assignment 5: ANOVA TESTS. Two parts. Psychology Essay

Assignment 5: ANOVA TESTS
Statistical Methods
July 10th, 2018
ASSIGNMENT 5: ANOVA TESTS
The assignment is in two parts. The first part involves performing a one-way ANOVA test to establish the effect of marital status on the level of life satisfaction. A two-way ANOVA test was conducted in the second part to determine the effect of alcohol on mate selection among all participants. The two ANOVA tests were based on two distinct datasets where the One-Way ANOVA test was based on the Divorce.sav dataset, while the two-way ANOVA was based on Goggles, .sav data dataset. The test output results were then used to answer the assignment questions. The one-way and two-way ANOVA tests were performed using SPSS version 21. The test output results were imported
Part I: One-Way ANOVA test
The divorce.sav dataset was analyzed to determine the independent and dependent variables. The One-way ANOVA test variables were determined as follows:
One-way ANOVA Test Variables
Dependent variable:

Life satisfaction

Independent variable:

Marital status

Null Hypothesis
The null hypothesis was determined as follows:
Null Hypothesis, [H0]: There is no difference in the life satisfaction means of all marital status groups (married, separated, divorced, widowed, cohabit).
Alternate Hypothesis
Alternate Hypothesis, [Ha]: Life satisfaction means of all marital status means are not equal.
One-way ANOVA test was performed on the Divorce. sav dataset using a confidence level of 95%. The output table below presents the sample size data for the five marital status groups.
Table 1: One-Way ANOVA test –Life satisfaction vs. Marital status
1781175173990
The above output result suggests that the widowed group only comprised of 3 participants. The small sample size for the group could be linked to an increase in the sampling error probability for the widowed group. As a result, inferential statistical challenges could arise given that the sample is non-representative of the entire population, which could adversely impact on the generalizability of results (Creswell, p. 146). One-way ANOVA test analysis further provided the means and standard deviations for the different marital status groups. The test output results are depicted in Table 2 as shown below.
Table 2: One-Way ANOVA Means and standard deviation results
105727513970
First and foremost, the one-way ANOVA test focused on determining the means and standard deviation for the different marital status levels as depicted in the tabulated output results. The mean was calculated based on a mean range of 4.53-4.94 even though the distinct group was determined to have the lowest life satisfaction score whereas the married group depicted the highest life satisfaction score. Furthermore, a Levene’s test of Equality of Error Variances was performed. The test output results indicated that p=. 068 thereby suggesting that the null hypothesis should not be rejected.
According to Levene’s Equality of Error Test, there are equal variances among all the dependent variable groups, and hence the null hypothesis should not be rejected.
Further analysis of the One-way ANOVA results to determine the F statistic where F=1.22; The output results also suggested a degree of freedom (df=4,224). The output also determined a p-value, p=.31. The SPSS results also indicated an eta squared value of η2 = .02. The SPSS output results are depicted in Table 3.
Table 3: Outputs from a One-Way ANOVA test on life satisfaction versus marital status (F statistic, degrees of freedom, p-value, and eta squared (η2))
209550160020
Analysis of the Output Results
Based on the output results, the p-value, p=.31 suggested that there was no significant difference between them (p>.05). Green and Salkind (2008), was used to determine the effect size index, η2. The results indicated an effect size index of η2=.02 indicating an insignificant difference between life satisfaction and marital status means. Green and Salkind (2008) assert that the effect size index, η2, largely depends on the specific area of research. Thus the η2 result of .01 should be interpreted as small, whereas that of .06 and .14 should be interpreted as medium and large respectively (Green and Salkind, 2008, p.185). The one-way NOVA test results indicated non-statistical significance which conventionally eliminated the need to perform Post Hoc tests on the dataset. Furthermore, the test results indicated that the life satisfaction level across all the sampled groups lacked a variability in statistical significance. This result suggested that the null hypothesis, H0 should be accepted rather than rejected. Based on the findings of Morgan, Reichert, and Harrison (2002), the study hypothesis was used to analyze the ANOVA test based on a 95% confidence level. The results were then used to make test output projections. The output results suggested that there was no significant difference between the life satisfaction and marital group means. This lack of significant difference suggested that, F(4, 224) = 1.22, p > .05, eta2 = .02.
Two-way ANOVA test
Two types of variables were determined: dependent and independent variables. The independent variables for the test were two, alcohol consumption (3 levels) and gender (2 levels). Dependent variable-Attractiveness of the mate. The null hypothesis – the amount of alcohol consumed by the participants irrespective of gender had no impact on the subjective perception of the mate with whom the participant engaged with during the study. Conversely, the Null hypothesis, H0: Alcohol consumption for all participants irrespective of their gender had no im...