This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: SOCIOLOGY 005 Lecture 4 Testing Hypotheses About Two Means • Up until this point, we have focused on testing the difference between a sample mean and a population mean identiFed in a null hypothesis • We can also test for a difference between two or more sample means • Two techniques are available for this kind of bivariate analysis • TwoSample ttest • ANOVA TwoSample ttest • Twosample ttest • Can only be used with 2 groups • Requires us to know the mean, std. deviation, and N for each t = X 1 ! X 2 s 1 2 N 1 + s 2 2 N 2 TwoSample ttest InClass Example Anova • Analysis of Variance (ANOVA) is a statistical technique designed to handle situations in which two or more groups are being used to predict the value of a dependent variable • It utilizes a Ftest to determine if two or more groups have different means • The twosample ttest is a special case of an Ftest Anova • With ANOVA, we can also calculate: • Effects • A measure of association Anova • Effects • Within ANOVA an effect is simply the impact of the classi¡cation variable on the dependent variable • In order to calculate the effects, we need to know: • The grand mean • The group mean Anova • Effects • To calculate effects, we use the following equation: ! K = μ K " μ Anova • Measure of Association • In order to calculate EtaSquared, our measure of association, it is helpful to wait and calculate this until after we calculate the necessary statistics for our ftest Anova • Since ANOVA is related to variability, we’ll be comparing variability between groups and variability within groups • This is called between variance and within variance Anova Let’s imagine an income distribution in which 3 groups (HS, BA, BA+) had identical means and variances Income In such an example, the within variance of a group would be no different that the within variation of other groups or the...
View
Full Document
 Spring '10
 dunn
 Sociology, Variance, yk

Click to edit the document details