Hostname: page-component-745bb68f8f-grxwn Total loading time: 0 Render date: 2025-01-08T04:26:56.121Z Has data issue: false hasContentIssue false
  • English
  • Français

A Widely Applicable Extension of the Random Effects Two-Way Layout: Its Definition and Statistical Analysis Based on Group Invariance

Published online by Cambridge University Press:  01 January 2025

Heng Li
Heng Li*
Affiliation:
University of Rochester
*
Requests for reprints should be sent to Heng Li, FDA/CDRH (HFZ-550), 1350 Piccard Dr., Rockville MD 20850.
Get access Rights & Permissions [Opens in a new window]

Abstract

A type of data layout that may be considered as an extension of the two-way random effects analysis of variance is characterized and modeled based on group invariance. The data layout seems to be suitable for several scenarios in psychometrics, including the one in which multiple measurements are taken on each of a set of variables, and the measurements can be divided into exchangeable subsets. The algebraic structure of the model is studied, which leads to results that are applicable to such problems as estimating correlation matrix corrected for attenuation and testing symmetry hypotheses.

Keywords

disattenuationgroup invariancemeasurement errorreliability
Type
Theory And Methods
Information
Psychometrika , Volume 69 , Issue 1 , March 2004 , pp. 55 - 64
Copyright
Copyright © 2004 The Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

The author would like to thank NIH for support through grants M01 RR00044 and P01 DE13539, and NSF and ASA Survey Section for a travel grant which allowed him to present this paper as an invited young researcher at the Conference in Celebration of Wayne A. Fuller's 70th birthday.

References

Andersson, S. A. (1975). Invariant normal models. Annals of Statistics, 3, 132–154CrossRefGoogle Scholar
Andersson, S. A., Madsen, J. (1998). Symmetry and lattice conditional independence in a multivariate normal distribution. Annals of Statistics, 26, 525–572CrossRefGoogle Scholar
Box, G. E. P., Tiao, G. C. (1973). Bayesian Inference in Statistical Analysis. Reading, MA: Addison-WesleyGoogle Scholar
Cronbach, L.J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297–334CrossRefGoogle Scholar
Cronbach, L.J., Gleser, G. C., Nanda, H., Rajaratnam, N. (1972). The Dependability of Behavioral Measurements. New York: John Wiley and SonsGoogle Scholar
Cronbach, L.J., Schonemann, P., McKie, D. (1965). Alpha coefficients for segmented parallel tests. Educational and Psychological Measurement, 25, 291–312CrossRefGoogle Scholar
Dawid, A. P. (1986). Symmetry analysis of the mixed model. Research Report No. 53, Department of Statistical Science, University College London.Google Scholar
Dawid, A. P. (1988). Symmetry models and hypotheses for structured data layouts (with discussion). Journal of Royal Statistical Society B, 50, 1–34CrossRefGoogle Scholar
Feldt, L. S., Brennan, R. L. (1989). Reliability. In Lynn, R. L. (Eds.), Educational Measurement (pp. 105–146). New York: MacmillanGoogle Scholar
Hakstian, A.R., Schroeder, M.L., Rogers, W.T. (1989). Inferential procedures for correlation-coefficients corrected for attenuation. Psychometrika, 53, 27–43CrossRefGoogle Scholar
Hakstian, A.R., Schroeder, M.L., Rogers, W.T. (1989). Inferential theory for partially disattenuated correlation-coefficients. Psychometrika, 54, 397–407CrossRefGoogle Scholar
Li, H. (1997). A unifying expression for the maximal reliability of a linear composite. Psychometrika, 62, 245–249CrossRefGoogle Scholar
Li, H. (2001). Nested designs with multivariate measurement: An illustration of the structural approach to random effects multivariate analysis of variance. Communications in Statistics: Theory and Methods, 30, 887–895CrossRefGoogle Scholar
Li, H. (2002). Likelihood based quadratic analyses for diallel cross and other experiments involving data collected on ordered pairs of sampling units. Statistics and Probability Letters, 58, 61–69CrossRefGoogle Scholar
Li, H. (2002). Modeling through group invariance: An interesting example with potential applications. Annals of Statistics, 30, 1069–1080CrossRefGoogle Scholar
Li, H., Loken, E. (2002). A unified theory of statistical analysis and inference for variance component models for dyadic data. Statistica Sinica, 12, 519–535Google Scholar
Li, H., Wainer, H. (1997). Toward a coherent view of reliability in test theory. Journal of Educational and Behavioral Statistics, 22, 478–484Google Scholar
Li, H., Rosenthal, R., Rubin, D. B. (1996). Reliability of measurement in psychology: From Spearman-Brown to maximal reliability. Psychological Methods, 1, 98–107CrossRefGoogle Scholar
Lord, F.M., Novick, M.R. (1968). Statistical Theories of Mental Test Scores. Reading, MA: Addison-WesleyGoogle Scholar
McDonald, R. P. (1999). Test Theory: A Unified Treatment. Mahwah, NJ: Lawrence Erlbaum AssociatesGoogle Scholar
Perlman, M. D. (1987). Comment: Group symmetry covariance models. Statistical Science, 2, 421–425CrossRefGoogle Scholar
Rajaratnam, N., Cronbach, L.J., Gleser, G.C. (1965). Generalizability of statified parallel test. Psychometrika, 30, 39–56CrossRefGoogle Scholar
Rosner, B., Willett, W.C. (1988). Interval estimates for correlation coefficients corrected for within-person variation: Implications for study design and hypothesis testing. American Journal of Epidemiology, 127, 377–386CrossRefGoogle ScholarPubMed
Scheffé, H. (1959). Analysis of Variance. New York: John Wiley and SonsGoogle Scholar
Searle, S.R., Casella, G., McCulloch, C.E. (1992). Variance Components. New York: John Wiley and SonsCrossRefGoogle Scholar
Tjur, T. (1984). Analysis of variance models in orthogonal designs (with discussion). International Statistical Review, 52, 33–81CrossRefGoogle Scholar