Profile AnalysisProfile AnalysisSlide 3Repeated Measures DataSlide 5Commensurate DataQuestions asked by profile analysisQuestionsSlide 9Slide 10Slide 11Slide 12Slide 13Slide 14Slide 15Slide 16Slide 17Slide 18Slide 19Slide 20Slide 21Slide 22LimitationsSlide 24Slide 25Slide 26AssumptionsSlide 28Slide 29Slide 30Slide 31Slide 32Profile AnalysisProfile AnalysisIntro and AssumptionsIntro and AssumptionsPsy 524Psy 524Andrew AinsworthAndrew AinsworthProfile Analysis Profile Analysis •Profile analysis is the repeated measures Profile analysis is the repeated measures extension of MANOVA where a set of DVs extension of MANOVA where a set of DVs are commensurate (on the same scale).are commensurate (on the same scale).Profile AnalysisProfile Analysis•The common use is where a set of DVs represent The common use is where a set of DVs represent the same DV measured at multiple time pointsthe same DV measured at multiple time points•used in this way it is the multivariate alternative to used in this way it is the multivariate alternative to repeated measures or mixed ANOVArepeated measures or mixed ANOVA•The choice often depends on the number of The choice often depends on the number of subjects, power and whether the assumptions subjects, power and whether the assumptions associated with within subjects ANOVA can be associated with within subjects ANOVA can be met (e.g. sphericity)met (e.g. sphericity)Repeated Measures DataRepeated Measures Data1 2 3 412 T T T TGDV DV DV DVGDV DV DV DVProfile AnalysisProfile Analysis•The less common use is to compare groups on The less common use is to compare groups on multiple DVs that are commensurate (e.g. multiple DVs that are commensurate (e.g. subscales of the same inventory)subscales of the same inventory)•Current stat packages can be used to perform more Current stat packages can be used to perform more complex analyses where there are multiple complex analyses where there are multiple factorial between subjects effectsfactorial between subjects effectsCommensurate DataCommensurate Data1 2 3 412DV DV DV DVGGQuestions asked by profile analysisQuestions asked by profile analysis•There is one major question asked by There is one major question asked by profile analysis; Do groups have similar profile analysis; Do groups have similar profiles on a set of DVs?profiles on a set of DVs?QuestionsQuestions•Usually in application of profile analysis a Usually in application of profile analysis a researcher is trying to show that groups are researcher is trying to show that groups are not different, that is why most tests are not different, that is why most tests are named after the “null” case.named after the “null” case.QuestionsQuestions•Segments – difference scores (or other Segments – difference scores (or other linear combinations) between adjacent DV linear combinations) between adjacent DV scores that are used in two of the major tests scores that are used in two of the major tests of profile analysisof profile analysisQuestionsQuestions•Between Subjects – (univariate) – “Equal Between Subjects – (univariate) – “Equal Levels”Levels”•On average does one group score higher than On average does one group score higher than the otherthe other•Averaging across DVs are the groups differentAveraging across DVs are the groups different•This would be the between-groups main effect This would be the between-groups main effect in mixed ANOVAin mixed ANOVAQuestionsQuestions•BS – (univariate) – “Equal Levels”BS – (univariate) – “Equal Levels”•It is called the equal levels hypothesis in profile It is called the equal levels hypothesis in profile analysisanalysis•Groups are different when the equal levels Groups are different when the equal levels hypothesis is rejectedhypothesis is rejectedQuestionsQuestions•Within Subjects (multivariate) – “Flatness”Within Subjects (multivariate) – “Flatness”•This is equivalent to the within subjects main This is equivalent to the within subjects main effect in repeated measures ANOVAeffect in repeated measures ANOVA•In profile analysis terms this is a test for the In profile analysis terms this is a test for the flatness of the profilesflatness of the profiles•““Do all DVs elicit the same average response?”Do all DVs elicit the same average response?”Flatness - no WS main effect012345678910T1 T2 T3 T4TimeG1G2QuestionsQuestions•WS (multivariate) – “Flatness”WS (multivariate) – “Flatness”•If flatness is rejected than there is a main effect If flatness is rejected than there is a main effect across the DVsacross the DVs•This is usually only tested if the test for parallel This is usually only tested if the test for parallel profiles is not rejected (we’ll talk about this in a profiles is not rejected (we’ll talk about this in a second)second)QuestionsQuestions•Interaction (multivariate) – Parallel ProfilesInteraction (multivariate) – Parallel Profiles•Are the profiles for the two groups the same?Are the profiles for the two groups the same?•This is a test for the interaction in repeated This is a test for the interaction in repeated measures ANOVAmeasures ANOVA•This is usually the main test of interest in This is usually the main test of interest in profile analysisprofile analysis•An interaction occurs when the profiles are not An interaction occurs when the profiles are not parallelparallelQuestionsQuestions•If any of the hypotheses tested by profile If any of the hypotheses tested by profile analysis are significant than they need to be analysis are significant than they need to be followed by contrasts.followed by contrasts.•Contrasts (on the main effects, with no interaction)Contrasts (on the main effects, with no interaction)•Simple effectsSimple effects•Simple contrastsSimple contrasts•Interaction contrasts (done when the interaction and Interaction contrasts (done when the interaction and both main effects are significant)both main effects are significant)•More on this laterMore on this laterInteraction and possibly one (but not both) main effectQuestionsQuestions•Estimating parametersEstimating parameters•Usually done through plots of the actual Usually done through plots of the actual profilesprofiles•If the flatness hypothesis is rejected than you If the flatness hypothesis is rejected than you would plot the average DV scores averaged would