**Mean
Scores and ANOVA Tables**

Try the following exercise:

1. Open

keast4j.sav2. Click on

Statistics, Compare Means, Means3. Select the variables of interest that are bivariate in nature (or categorized) such as

dlowedn(education level),lowlit(low/not low literacy level),dbadro(good/ bad roofing),watsourc(good / bad water source) andnotoliet(no/yes toilet access). Place these variables in theIndependent variable box4. Select the variable

wazand place it in theDependent variable box5. Click on

Optionsand select the options forMean, Std. deviation, Number of Casesand forStatistics for the first layerclick on the box forANOVA table6. Click on

ContinueandOK.

The mean scores and ANOVA tables with P-values look like this in SPSS output. These can be sorted through and the interesting ones can be presented in a table for easy reading. Look at the output below, and a section will follow to show an interesting method for presentation. (Use the P = 0.05 to approximate a significant finding).

**ROOFING (SES)**

**WATER SOURCE**

**EDUCATION**

**TOILET (SANITATION)**

The results for all of the variables come out with the expected associations, and the differences for roofing, education and sanitation are all significant (p-value < 0.05). The one non-significant association is that for water source, mostly influenced by the small cell size for the group of districts that have safe water (n=33). The size of the difference is still large, and would be the primary interest here (-0.98 SD WAZ vs. -1.30 SD WAZ).