Within Subjects DesignsTopics in WS designsWithin Subjects?Slide 4Slide 5Slide 6Types of WS designsSlide 8Issues and AssumptionsSlide 10Slide 11Slide 12Slide 13Slide 14Slide 15Slide 16Slide 17Slide 18Slide 19Slide 20Analysis – 1-way WSSlide 22Sums of SquaresSlide 24Slide 25Slide 26Slide 27Slide 28Slide 29Slide 30Slide 31Slide 32Slide 33Slide 34Slide 35Slide 36Slide 37Slide 38Within Subjects DesignsPsy 420Andrew AinsworthTopics in WS designsTypes of Repeated Measures DesignsIssues and AssumptionsAnalysisTraditional One-wayRegression One-wayWithin Subjects?Each participant is measured more than onceSubjects cross the levels of the IVLevels can be ordered like time or distanceOr levels can be un-ordered (e.g. cases take three different types of depression inventories)Within Subjects?WS designs are often called repeated measures Like any other analysis of variance, a WS design can have a single IV or multiple factorial IVsE.g. Three different depression inventories at three different collection timesWithin Subjects?Repeated measures designs require less subjects (are more efficient) than BG designsA 1-way BG design with 3 levels that requires 30 subjects•The same design as a WS design would require 10 subjectsSubjects often require considerable time and money, it’s more efficient to use them more than onceWithin Subjects?WS designs are often more powerfulSince subjects are measured more than once we can better pin-point the individual differences and remove them from the analysisIn ANOVA anything measured more than once can be analyzed, with WS subjects are measured more than onceIndividual differences are subtracted from the error term, therefore WS designs often have substantially smaller error termsTypes of WS designsTime as a variableOften time or trials is used as a variableThe same group of subjects are measured on the same variable repeatedly as a way of measuring changeTime has inherent order and lends itself to trend analysisBy the nature of the design, independence of errors (BG) is guaranteed to be violatedTypes of WS designsMatched Randomized Blocks1. Measure all subjects on a variable or variables2. Create “blocks” of subjects so that there is one subject in each level of the IV and they are all equivalent based on step 13. Randomly assign each subject in each block to one level of the IVIssues and AssumptionsBig issue in WS designsCarryover effects•Are subjects changed simple by being measured?•Does one level of the IV cause people to change on the next level without manipulation?•Safeguards need to be implemented in order to protect against this (e.g. counterbalancing, etc.)Issues and AssumptionsNormality of Sampling DistributionIn factorial WS designs we will be creating a number of different error terms, may not meet +20 DFThan you need to address the distribution of the sample itself and make any transformations, etc.You need to keep track of where the test for normality should be conducted (often on combinations of levels)ExampleIssues and AssumptionsIndependence of ErrorsThis assumption is automatically violated in a WS designA subject’s score in one level of the IV is automatically correlated with other levels, the close the levels are (e.g. in time) the more correlated the scores will be.Any consistency in individual differences is removed from what would normally be the error term in a BG designIssues and AssumptionsSphericityThe assumption of Independence of errors is replaced by the assumption of Sphericity when there are more than two levelsSphericity is similar to an assumption of homogeneity of covariance (but a little different)The variances of difference scores between levels should be equal for all pairs of levelsIssues and AssumptionsSphericityThe assumption is most likely to be violated when the IV is time•As time increases levels closer in time will have higher correlations than levels farther apart•The variance of difference scores between levels increase as the levels get farther apartIssues and AssumptionsAdditivityThis assumption basically states that subjects and levels don’t interact with one anotherWe are going to be using the A x S variance as error so we are assuming it is just randomIf A and S really interact than the error term is distorted because it also includes systematic variance in addition to the random varianceIssues and AssumptionsAdditivityThe assumption is literally that difference scores are equal for all casesThis assumes that the variance of the difference scores between pairs of levels is zeroSo, if additivity is met than sphericity is met as wellAdditivity is the most restrictive assumption but not likely metIssues and AssumptionsCompound SymmetryThis includes Homogeneity of Variance and Homogeneity of CovarianceHomogeneity of Variance is the same as before (but you need to search for it a little differently)Homogeneity of Covariance is simple the covariances (correlations) are equal for all pairs of levels.Issues and AssumptionsIf you have additivity or compound symmetry than sphericity is met as wellIn additivity the variance are 0, therefore equalIn compound symmetry, variances are equal and covariances are equalBut you can have sphericity even when additivity or compound symmetry is violated (don’t worry about the details)The main assumption to be concerned with is sphericityIssues and AssumptionsSphericity is usually tested by a combination of testing homogeneity of variance and Mauchly’s test of sphericity (SPSS)If violated (Mauchly’s), first check distribution of scores and transform if non-normal; then recheck.If still violated…Issues and AssumptionsIf sphericity is violated:1. Use specific comparisons instead of the omnibus ANOVA2. Use an adjusted F-test (SPSS)•Calculate degree of violation ()•Then adjust the DFs downward (multiplies the DFs by a number smaller than one) so that the F-test is more conservative•Greenhouse-Geisser, Huynh-Feldt are two approaches to adjusted F (H-F preferred, more conservative)Issues and AssumptionsIf sphericity is violated:3. Use a multivariate approach to repeated measures (take Psy 524 with me next semester)4. Use a maximum likelihood method that allows you to specify that the variance-covariance matrix is other than compound symmetric (don’t
View Full Document