49-1-604 - Mixed model methodologies.
49-1-604. Mixed model methodologies.
As used in this part, mixed model methodologies that provide for best linear unbiased prediction, or similar language setting forth the methodology used for evaluating measured progress of students, teachers, schools or school districts, has the meaning and shall be interpreted as set forth in the following references:
(1) A Unified Approach to Mixed Linear Models, McLean, Sanders, and Stroup; The American Statistician, February 1991; Vol. 45, No. 1;
(2) Extension of the Gauss-Markov Theorem to Include the Estimation of Random Effects, Harville; The Annals of Statistics, 1976; Vol. 4, No. 2, 384-395;
(3) Analysis of Variance in the Mixed Model: Higher Level, Nonhomogeneous, and Random Regressions, Henderson; Biometrics, September 1982; No. 38, 623-640;
(4) Maximum Likelihood Approaches to Variance Component Estimation and to Related Problems, Harville; Journal of the American Statistical Association, July 1977; Vol. 72, No. 358;
(5) Approximations for Standard Errors of Estimators of Fixed and Random Effects in Mixed Linear Models, Kackar and Harville; Journal of the American Statistical Association, December 1984; Vol. 79, No. 388; and
(6) The Analysis of Unbalanced Linear Models with Variance Components, Engel; Statistica Neerlandica, 1990; Vol. 44, No. 4.
[Acts 1992, ch. 535, § 4.]