Advance Biostats SPSS (Two-Way ANOVA) Use Tukey

Part 3

Step-by-Step Guide Assignment Problem 3.3

2-Way ANOVA

Problem 3. Discuss whether the assumptions of homogeneity of variance of the groups and normality of the data on Exercise are met. Be sure to include output to support your decision on whether the assumptions have been met. (Continue with the analyses even if assumptions are not met.)

Review:

Homogeneity of variance means that the variance about the means is equal among the independent variables.

Normality means that the data for each level of the independent variables are normally distributed.

Step 1. Open the SPSS Practice data set if you have closed it.

Step 2. Go to Analyze(Descriptive Statistics(Explore

Step 3. In Explore window, place Steps per month [EXERCISE] in the Dependent List. Place Education level [EDUC] and Sex [SEX] in the Factor List. Click on Plots…

Step 4. Select None in Boxplots, check Normality plots with tests, and select Power estimation in Spread vs Level with Levene’s Test. Click Continue.

Step 5. In the Explore window, make sure Both is selected in Display then click OK.

Step 6. Locate and examine the Test of Homogeneity of Variance for SEX and EDUC in the Output:

SEX:

Test of Homogeneity of Variance
	Levene Statistic	df1	df2	Sig.
Steps per month	Based on Mean	5.328	1	98	.023
	Based on Median	3.932	1	98	.050
	Based on Median and with adjusted df	3.932	1	87.549	.051
	Based on trimmed mean	4.938	1	98	.029

EDUC

Test of Homogeneity of Variance
	Levene Statistic	df1	df2	Sig.
Steps per month	Based on Mean	2.072	2	97	.131
	Based on Median	1.748	2	97	.180
	Based on Median and with adjusted df	1.748	2	89.800	.180
	Based on trimmed mean	1.978	2	97	.144

Step 7. Locate and examine the Tests of Normality tables for both Sex and Education level:

SEX

Tests of Normality
	Sex	Kolmogorov-Smirnova	Shapiro-Wilk
		Statistic	df	Sig.	Statistic	df	Sig.
Steps per month	Male	.130	49	.038	.915	49	.002
	Female	.139	51	.015	.929	51	.005
a. Lilliefors Significance Correction

EDUC

Tests of Normality
	Education level	Kolmogorov-Smirnova	Shapiro-Wilk
		Statistic	df	Sig.	Statistic	df	Sig.
Steps per month	No or some high school	.181	16	.170	.895	16	.066
	High school diploma	.222	31	.000	.872	31	.002
	College Degree	.115	53	.076	.932	53	.005
a. Lilliefors Significance Correction

Discussion:

Your discussion must cover issues such as

1. Meaning of the test for homogeneity

2. Which variables meet the assumptions for the test

3. Meaning of the test for normality

4. Which variables are normally distributed and why