• Created by: Chloe
  • Created on: 02-05-15 14:16
View mindmap
    • Is an extension of ANCOVA - uses partial correlation logic to improve internal validity in experimental studies
      • Allows the researcher to asses whether the effects of an IV on a DV remain after the influence of one or more control variables, measures at the same time (covariates)have been partialled out
        • The measurement of the covariates can be used to improve the sensitivity of the test of the IV/DV relationship by taking into account the regression of the DV onto the covariate
    • There may be several covariates, these may be scores on a pretest or other measures thought to have a direct functional relationship to the DV
      • Covariates need to be strongly correlated with the DV, not just convenient measures such as age or IQ
        • Naturally ocurring differences between groups (confounding variables) which may affect the DV cannot be controlled by randomisation in field experiments, but ANCOVA allows you to exercise statistical control
    • Main assumptions of ANCOVA: (1) that the DV & covariate should be linearly related (2) that the regression coefficient (slope) should be about the same for each subgroup


No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all ANCOVA resources »