Measurement and Statistics_Dasch_Date__040910

Measurement and Statistics_Dasch_Date__040910 - the...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Regression cont'd - Forced entry regression all variables entered into model simultaneously the results obtained depend on the variables entered into the model must have good theoretical reasons for including a particular variance - stepwise regression variables are entered into the model based on mathematical reasoning (semi-partial correlation) computer selects variables in steps step 1: SPSS looks for the predictor that can explain the most variance in the outcome variable (highest semi-partial correlation) step 2: a second one is chosen from the remaining predictors the same way step 3: this process continues until all predictors are entered Multicollinearity - generally assume there isn’t - when there is a strong correlation between 2 or more predictors (.7 or greater) variance inflation factor (VIF) o greater than 10 shows a problem with multicollinearity, need to not use one of
Background image of page 1

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

View Full DocumentRight Arrow Icon
Background image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: the variables tolerance (1/VIF) o .1 or less is a problem Regression in SPSS Analyze regression linear Outcome dependent Move independent variable you know has an effect block 1 Click next All other independent variables block 2 Click stat check: estimates, model fit, r-square change, collinearity diagnostics - r-square turn into percent - look at r-square change for variables entered into model 2 - anova table F ratios (overall fit of the model) look at the significance -coefficients table t statistic see if its significant (compare to .05) o if significant: means the variance has a significant effect on the outcome unstandardized B slope standardized B strength of effect...
View Full Document

This note was uploaded on 08/30/2010 for the course PSYC 209 taught by Professor Hoffman during the Spring '08 term at University of Delaware.

Page1 / 2

Measurement and Statistics_Dasch_Date__040910 - the...

This preview shows document pages 1 - 2. Sign up to view the full document.

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