This preview shows pages 1–4. Sign up to view the full content.
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 6 Part 1 Correlation and Regression • Up until now we have examined relationships between • A nominal variable and an ordinal variable • TwoSample ttest • ANOVA • Nominal and ordinal variables • Independence • PRE Measures • NonPre Measures Correlation and Regression • Now we are interested in evaluating relationships between two interval or ratio level variables • For these techniques, we have a more dif¡cult time “cheating” and using ordinal level data Correlation and Regression • We are going to be interested in three things pertaining to relationship between two interval/ratio level variables • Form of the relationship • Strength of the relationship • Signi¡cance of the relationship Correlation and Regression • We are going to be interested in three things pertaining to relationship between two interval/ratio level variables • Form of the relationship  Regression • Strength of the relationship  correlation • Signi¡cance of the relationship  Ftest Correlation and Regression • For the form of the relationship between X and Y, we need to conduct regression analysis • A regression in general linear model for which ANOVA is a special case • Like ANOVA, we are interested in variation; however, now we are also interested in covariation • With regression analysis, it is important to remember that the single best predictor of a interval/ratio level variable is the mean of that variable Correlation and Regression One way to conceptualize what we are doing with regression is to look at the data using a scatterplot We are looking at Medium income and Part I crimes Correlation and Regression With a regression, we are trying to construct a single line which best summarizes the relationship between our IV and DV Correlation and Regression • With regression analysis, we are trying to predict a linear relationship • It is possible to have a nonlinear relationship • As long as the data appear to be related in a linear manner we can use regression analysis Correlation and Regression • As a result of this linear assumption in regression, we express the equation for a regression much in the same way a line is expressed in algebra Y = a + bX Y= the Score of the DV a=the Y intercept b=the slope of the regression line X=Score of IV Correlation and Regression In order to create the regression line, we need to Frst calculate the slope b = ! ( X " X...
View
Full
Document
This note was uploaded on 10/23/2011 for the course SOC 10 taught by Professor Dunn during the Spring '10 term at UC Riverside.
 Spring '10
 dunn
 Sociology

Click to edit the document details