Week_9_10_prep_for_ANOVA

Week_9_10_prep_for_ANOVA - SAN JOS STATE UNIVERSITY College...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
SAN JOSÉ STATE UNIVERSITY College of Social Work S. W. 242 Spring 2008 Edward Cohen Week 9 March 21, 2008 Note on your calendar: I’ll be out of town Thursday April 3 (no office hours) Let’s slow down—nagging questions, unfinished business, general confusion? Discuss Final Paper (don’t slow down too much!) Prep for ANOVA Happy Spring Break! Concepts you should know: When to use ANOVA One-way ANOVA Two-way ANOVA (or Two-factor ANOVA) F statistic Within-groups variation Between-groups variation Post-hoc tests Multiple comparisons I. When to use ANOVA? A. Remember that the t -test compared means (of ratio or interval level variables) along the attributes of a dichotomous nominal variable (e.g. a variable with only two attributes such as male and female), i.e. comparing only two means at a time B. ANOVA is a distant cousin to the t -test. It compares means for categorical ( nominal or ordinal ) variables with three or more attributes , i.e. it compares three or more means, for example 1. Are there differences between the mean LCSW licensing scores for MSW graduates of various ethnicities? 2. For children residing in residential treatment does the number of behavioral outbursts vary by level of self esteem (High, Medium, Low)? 3. In an experimental study of treatment for depression, clients are randomized to the following groups: 1) medication only, 2) medication plus individual psychotherapy, and 3) psychotherapy only. Do the clients in the “medication plus psychotherapy” group show more improved outcomes (lower depression scores) than the other two groups? II. How it works 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
A. ANOVA is an “Omnibus test” 1. “Omnibus” is an adjective defined as covering many things or classes , in this case it makes a global statement about the Null hypothesis. 2. ANOVA only tests the Null hypothesis of “no difference between the means” , or “the means of the dependent variables are equal among the groups”– in other words ANOVA tells us whether or not any differences among the means might have been produced due to sampling error 3. The ANOVA statistic and associated p value does not prove or disprove your research hypothesis by singling out one of the means as “significantly different than the others” – a statistically significant finding merely says that the Null can be rejected and “the means are different”(or “the means are not equal”) 4. The ANOVA statistic is called the “ F ratio”. It has the same function as the t statistic and the Chi Square value, and it has its own distribution table (built into SPSS) so it can also be associated with a p value. Except the F ratio is easier to interpret: The F ratio reflects the variation of means between the groups divided by the variation of means within the groups. Why do we need this ratio? Because we need some way to tease out the variation
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 09/08/2010 for the course SCWK 242 at San Jose State University .

Page1 / 7

Week_9_10_prep_for_ANOVA - SAN JOS STATE UNIVERSITY College...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online