Law - Research Methods in CJ

SPSS Weekly Assignment

ANOVA One-Factor Independent Measures

Karen Raymond

Keiser University

MACJ590

Dr. Arthur Jones

04/14/2013

Explanation

In some situations, researchers will want to compare at least three means to each other. Analysis of Variance (ANOVA) is a procedure that enables the researcher to make such comparisons. The one way or one factor ANOVA is titled as such because it involves a single independent variable. In this particular instance, the research question sought to determine whether students who work a certain number of hours outside of school have a lower GPA. Of the tests depicted in the below charts, the Kolmogorov-Smirnov test evaluated whether a significant departure from normality existed within the three student categories (i.e., working no hours, working some hours, or working many hours). The depicted P values were higher than 0.05, and therefore, the null hypothesis was not rejected and these measures were deemed statistically significant, meaning these data did not violate the normality assumption. Levene’s test for homogeneity of variances found that the null hypothesis was again rejected, meaning the data did not violate the homogeneity of variance assumption. Lastly, having satisfied the independent measures of ANOVA, the analysis found that the depicted means do differ in a statistically significant manner. That is, students who worked some hours had higher GPAs than students who didn’t work. The displayed post-hoc tests confirmed this inference.

Oneway

Descriptives
Current GPA
	N	Mean	Std. Deviation	Std. Error	95% Confidence Interval for Mean	Minimum	Maximum
					Lower Bound	Upper Bound
none (0 hrs)	99	2.8800	.45845	.04608	2.7886	2.9714	1.90	4.00
some (1-19 hrs)	77	3.1664	.42697	.04866	3.0695	3.2633	1.98	3.94
many (20-99 hrs)	36	3.0197	.46263	.07710	2.8632	3.1763	2.10	3.90
Total	212	3.0077	.46437	.03189	2.9449	3.0706	1.90	4.00

Test of Homogeneity of Variances
Current GPA
Levene Statistic	df1	df2	Sig.
.414	2	209	.662

Post Hoc Tests

Multiple Comparisons
Dependent Variable: Current GPA
Tukey HSD
(I) work category	(J) work category	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Lower Bound	Upper Bound
none (0 hrs)	some (1-19 hrs)	-.28636*	.06807	.000	-.4470	-.1257
	many (20-99 hrs)	-.13972	.08719	.247	-.3455	.0661
some (1-19 hrs)	none (0 hrs)	.28636*	.06807	.000	.1257	.4470
	many (20-99 hrs)	.14664	.09045	.239	-.0669	.3601
many (20-99 hrs)	none (0 hrs)	.13972	.08719	.247	-.0661	.3455
	some (1-19 hrs)	-.14664	.09045	.239	-.3601	.0669
*. The mean difference is significant at the 0.05 level.

Current GPA
Tukey HSDa,b
work category	N	Subset for alpha = 0.05
		1	2
none (0 hrs)	99	2.8800
many (20-99 hrs)	36	3.0197	3.0197
some (1-19 hrs)	77		3.1664
Sig.		.210	.180
Means for groups in homogeneous subsets are displayed.
a. Uses Harmonic Mean Sample Size = 58.979.
b. The group sizes are unequal. The harmonic mean of the group sizes is used. Type I error levels are not guaranteed.