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 DocumentThis 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: 1 Chapter 14 HP350 Spring 2010 Midterm #2 Question The p value associated with a statistical test indicates __________ while the associated effect size indicates __________. a. practical importance; statistical significance b. statistical significance; statistical power * c. statistical significance; practical importance d. statistical power; statistical significance 2 3 t vs. F The t test is used to compare two group means. With more than two groups, the F test is used for a comparison of several means. While the t test compares the means , the F test compares the variances. 4 ttest is used to compare two means : H : μ A = μ B H 1 : μ A ≠ μ B Ftest is used to compare more than two means : ● H : μ A = μ B = μ C =… ● H 1 : not H 0 (the means are not equal)called an Omnibus test (at least one difference between the means) Hypothesis tests for t vs. F 5 Logic of ANOVA The F test is a signaltonoise ratio that divides up variability in a procedure called analysis of variance , or ANOVA . The analysis of variance provides a comparison of the variation between (signal) the groups and the variation within (noise) the groups. In this kind of analysis, a ratio (the F ratio , or F test ) is formed of and . 2 between S 2 w ith in S 6 F Ratios F ratios usually have values close to 1.0 when the variation between groups is not different from the variation within groups (i.e., when H is true). The larger the F ratio becomes, the greater the dispersion of group means relative to the dispersion of scores within groups. 7 ANOVA Example 8 Descriptives AGE AT FIRST INJECTION 51 16.94 2.626 .368 16.20 17.68 12 23 96 16.42 3.114 .318 15.79 17.05 6 25 96 16.26 3.072 .314 15.64 16.88 8 23 243 16.47 3.000 .192 16.09 16.84 6 25 NY LA NO Total N Mean Std. Deviation Std. Error Lower Bound Upper Bound 95% Confidence Interval for Mean Minimum Maximum...
View
Full Document
 Spring '09
 LANKENAU
 Statistics, Statistical hypothesis testing, Statistical significance, mean square, INJECTION Sum

Click to edit the document details