Popular in Advanced Statistics
Popular in Psychlogy
This 2 page Class Notes was uploaded by Alejandra D. Rocha on Wednesday October 7, 2015. The Class Notes belongs to PSYC 4317 at University of Texas at El Paso taught by Anthony Blum in Summer 2015. Since its upload, it has received 27 views. For similar materials see Advanced Statistics in Psychlogy at University of Texas at El Paso.
Reviews for ANCOVA
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 10/07/15
Analysis of Covariance ANCOVA Use ANCOVA for the same type of design you would use ANOVA Between Subjects one factor multiple level design Typically ANCOVA is a more powerful statistical test there is two main reasons to use ANCOVA 1 It is used to increase power of our test with a covariate that is related to dependent variable but not with independent variable 2 It statistically equalize different groups for the levels of independent variable Increases power of ANCOVA ANOVADF ratio independent variable effect chance variability chance variability If we could do something that decreased chance variability with decreasing independent variable then we could increase F more power F 102020302015 I F10551553 ANCOVA changes the way variance is computed ZYMquot2 To do ANCOVA factor with levels eg Temperature hot warm cold dependent variable score for each subject quantity lrelated to the dependent variable something not related to your independent variable I covariate info obtained before the experimentElQuantity not category For ANCOVA we are going to compute regression line between Covariate and Dependent Variable Covariate X axis dependent variable Y Y dependent bX covariate a Variance is computed by subtracting YYhat Yhat point of line corresponding to that Xpoint of the line predicted for the X that corresponds to Y YYhats YM which means bigger F value ANCOVA gives the group differences if all subjects had the same covariate score For ANCOVA to use the equalizing procedure the assumption of homogeneity of regression slopes must hold This means to compute a separate covariate dependent variable regression line for each group the assumption holds if the regression lines of the levels of the independent variable are parallel same slope Equalizing procedure I frequently used used for preexisting groups such as gender religion age homogeneity is usually violated homogeneity of regression lines Using SPSS for ANCOVA Use a column for the dependent variable other for the covariate and other for the independent variable using numbers for the different levels Go to analyzel Click on general linear model univariatel then click on the dependent variable and the fixed factor and covariate length Then look at the table at the main effect which is F of independent variable if it is a big number and if Sig is less than 05 we reject null hypothesis ANCOVA could lose power if you pick a covariate that is related to your independent variable
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'