Порівняння методів

	Multitrait-Multimethod Matrix	Common Method Bias Remedies
Галузь	Організаційна поведінка	Організаційна поведінка
Родина	Process / pipeline	Process / pipeline
Рік появи	1959	2003
Автор методу	Donald T. Campbell & Donald W. Fiske; Keith F. Widaman	Philip Podsakoff, Scott MacKenzie, Jeong-Yeon Lee & Nathan Podsakoff; Michael Lindell & David Whitney
Тип	Construct-validation matrix and covariance-structure modeling pipeline	Procedural and statistical remedies for method-induced bias
Основоположне джерело	Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56(2), 81-105.	Podsakoff, P. M., MacKenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879-903.
Інші назви	MTMM, Multitrait-Multimethod Analysis, Campbell-Fiske Matrix, CFA-MTMM	Common Method Variance Remedies, CMV Controls, Harman's Single-Factor Test, Marker Variable Technique
Пов'язані	3	3
Підсумок	The multitrait-multimethod matrix is the classic framework for establishing construct validity by measuring several traits with several methods and examining the resulting pattern of correlations. Donald Campbell and Donald Fiske introduced it in 1959, arguing that validating a construct requires showing both convergent validity — different methods of measuring the same trait agree — and discriminant validity — measures of different traits diverge even when they share a method. The matrix lays out every correlation among trait-method combinations so that these patterns can be read off systematically, while also exposing method variance, the tendency of measures sharing a method to correlate for the wrong reasons. Campbell and Fiske's original criteria were inspectional rules of thumb; Keith Widaman's 1985 work recast the matrix as a family of nested confirmatory factor models, providing formal significance tests for convergent validity, discriminant validity, and method variance. The MTMM matrix remains a foundational tool for asking whether a measure captures the construct it claims to.	Common method bias remedies are the procedural and statistical tools researchers use to detect and reduce the spurious covariance that arises when constructs are measured with the same method — typically a single self-report survey. Podsakoff, MacKenzie, Lee, and Podsakoff's 2003 review crystallized the problem, cataloguing the many sources of method bias and the design and analysis safeguards available, and it became the field's reference point. Because the same respondent, rating scale, and occasion can inflate correlations among unrelated constructs, method variance can manufacture or distort relationships that researchers then mistake for substance. The remedies fall into two families: procedural design choices that prevent method variance from entering the data, and statistical techniques that diagnose or partial it out afterward. Lindell and Whitney's marker-variable approach and Williams, Hartman, and Cavazotte's confirmatory-factor-analysis marker technique are the leading statistical correctives. Used together, these remedies make method bias a problem to be designed against and tested for rather than assumed away.