![[PDF]](/pb-assets/Icons/adobe-pdf-icon-1498649896243.png){kind=icon}
Abstract
Dominance analysis (DA) is a method used to compare the relative importance of predictors in multiple regression. DA determines the dominance of one predictor over another by comparing their additional R2 contributions across all subset models. In this article DA is extended to multivariate models by identifying a minimal set of criteria for an appropriate generalization of R2 to the case of multiple response variables. The DA results obtained by univariate regression (with each criterion separately) are analytically compared with results obtained by multivariate DA and illustrated with an example. It is shown that univariate dominance does not necessarily imply multivariate dominance (and vice versa), and it is recommended that researchers who wish to account for the correlation among the response variables use multivariate DA to determine the relative importance of predictors.
References
| Anderson-Sprecher, R Model comparisons andR2The American Statistician199448113117 Google Scholar | |
| Azen, R Inference for predictor comparisons: Dominance analysis and the distribution of R2 differences2000Unpublished doctoral dissertation, University of Illinois, Urbana-Champaign Google Scholar | |
| Azen, R, Budescu, DV The dominance analysis approach for comparing predictors in multiple regressionPsychological Methods20038129148 Google Scholar | Crossref | Medline | |
| Azen, R, Budescu, DV, Reiser, B Criticality of predictors in multiple regressionBritish Journal of Mathematical and Statistical Psychology200154201225 Google Scholar | Crossref | Medline | |
| Barber, BK Parental psychological control: Revisiting a neglected constructChild Development19966732963319 Google Scholar | Crossref | Medline | |
| Bring, J How to standardize regression coefficientsThe American Statistician199448209213 Google Scholar | |
| Budescu, DV Dominance analysis: A new approach to the problem of relative importance of predictors in multiple regressionPsychological Bulletin1993114542551 Google Scholar | Crossref | |
| Courville, T, Thompson, B Use of structure coefficients in published multiple regression articles: β?is not enoughEducational and Psychological Measurement200161229248 Google Scholar | SAGE Journals | |
| Cramer, EM, Nicewander, WA Some symmetric, invariant measures of multivariate associationPsychometrika1979444354 Google Scholar | Crossref | |
| Darlington, RB Multiple regression in psychological research and practicePsychological Bulletin196869161182 Google Scholar | Crossref | Medline | |
| Dunlap, WP, Landis, RS Interpretations of multiple regression borrowed from factor analysis and canonical correlationJournal of General Psychology1998125397407 Google Scholar | Crossref | |
| Efron, B Bootstrap methods: Another look at the jackknifeThe Annals of Statistics19797126 Google Scholar | Crossref | |
| Efron, B, Tibshirani, R The problem of regionsThe Annals of Statistics19982616871718 Google Scholar | Crossref | |
| Gibson, WA Orthogonal predictors: A possible resolution of the Hoffman-Ward controversyPsychological Reports1962113234 Google Scholar | SAGE Journals | |
| Green, PE, Carroll, D, DeSarbo, W A new measure of predictor variable importance in multiple regressionJournal of Marketing Research197815356360 Google Scholar | SAGE Journals | |
| Hays, WL Statistics19945th edFort Worth, TXHarcourt Brace Google Scholar | |
| Hedges, LV, Olkin, I The asymptotic distribution of commonality componentsPsychometrika198146331336 Google Scholar | Crossref | |
| Johnson, JW A heuristic method for estimating the relative weight of predictor variables in multiple regressionMultivariate Behavioral Research200035119 Google Scholar | Crossref | Medline | |
| Johnson, JW, LeBreton, JM History and use of relative importance indices in organizational researchOrganizational Research Methods20047238257 Google Scholar | SAGE Journals | |
| Johnson, RA, Wichern, DW Applied multivariate statistical analysis19984th edEnglewood Cliffs, NJPrentice Hall Google Scholar | |
| Kruskal, W Relative importance by averaging over orderingsThe American Statistician198741610 Google Scholar | |
| Kvålseth, TO Cautionary notes aboutR2The American Statistician198539279285 Google Scholar | |
| Lindeman, RH, Merenda, PF, Gold, RZ Introduction to bivariate and multivariate analysis1980Glenview, ILScott, Foresman and Company Google Scholar | |
| Mayeske, GW, Wisler, CE, Beaton, AE, Weinfeld, FD, Cohen, WM, Okada, T A study of our nation's schools1969Washington, DCUS. Department of Health, Education, and Welfare, Office of Education Google Scholar | |
| Mood, AM Macro-analysis of the American educational systemOperations Research196917770784 Google Scholar | Crossref | |
| Mood, AM Partitioning variance in multiple regression analyses as a tool for developing learning modelsAmerican Educational Research Journal19718191202 Google Scholar | SAGE Journals | |
| Mosteller, F, Tukey, JW Data analysis and regression: A second course in statistics1977Reading, MAAddison-Wesley Google Scholar | |
| Muller, KE Relationship between redundancy analysis, canonical correlation, and multivariate regressionPsychometrika198146139142 Google Scholar | Crossref | |
| Newton, RG, Spurrell, DJ A development of multiple regression for the analysis of routine dataApplied Statistics1967a165164 Google Scholar | Crossref | |
| Newton, RG, Spurrell, DJ Examples of the use of elements for clarifying regression analysisApplied Statistics1967b16165172 Google Scholar | Crossref | |
| Ogasawara, H Standard errors of matrix correlationsMultivariate Behavioral Research199834103122 Google Scholar | Crossref | |
| Pedhazur, EJ Kerlinger, FN Analytic methods in studies of educational effectsReview of research in education 31975Itasca, ILPeacock Google Scholar | |
| Pedhazur, EJ Multiple regression in behavioral research19973rd edForth Worth, TXHarcourt Google Scholar | |
| Pratt, JW Pukkila, T, Puntanen, S Dividing the indivisible: Using simple symmetry to partition variance explainedProceedings of the Second International Tampere Conference in Statistics1987Tampere, FinlandUniversity of Tampere245260 Google Scholar | |
| Scott, A, Wild, C Transformations and R2The American Statistician199145127129 Google Scholar | |
| Soofi, ES, Retzer, JJ Information indices: Unification and applicationsJournal of Econometrics20021071740 Google Scholar | Crossref | |
| Soofi, ES, Retzer, JJ, Yasai-Ardekani, M A framework for measuring the importance of variables with applications to management research and decision modelsDecision Sciences200031595625 Google Scholar | Crossref | |
| Stewart, D, Love, W A general canonical correlation indexPsychological Bulletin196870160163 Google Scholar | Crossref | Medline | |
| Tatsuoka, MM Multivariate analysis: Techniques for educational and psychological research19882nd edNew YorkMacmillan Google Scholar | |
| Theil, H How many bits of information does an independent variable yield in a multiple regression?Statistics & Probability Letters19876107108 Google Scholar | Crossref | |
| Theil, H, Chung, C-F Information-theoretic measures of fit for univariate and multivariate linear regressionsThe American Statistician198842249252 Google Scholar | |
| Thomas, DR, Hughes, E, Zumbo, BD On variable importance in linear regressionSocial Indicators Research: An International and Interdisciplinary Journal for Quality-of-life Measurement199845253275 Google Scholar | |
| Thomas, DR, Zumbo, BD Using a measure of variable importance to investigate the standardization of discriminant coefficientsJournal of Educational and Behavioral Statistics199621110130 Google Scholar | SAGE Journals | |
| Timm, NH Multivariate analysis with applications in education and psychology1975Monterey, CABrooks/Cole Google Scholar | |
| Timm, NH Multivariate linear regression models: Model selection, fit, association, and prediction2002Unpublished manuscript Google Scholar | |
| Van den Burg, W, Lewis, C Some properties of two measures of multivariate associationPsychometrika198853109122 Google Scholar | Crossref |
