Latent Variable Fit to Interlaboratory Studies

  • Published : 2000.12.01


The use of an unweighted mean and of separate tests is part of the current practice for analyzing interlaboratory studies, and we hope to improve on this method. We fit, using maximum likelihood(ML), a rather intricate, multi-parameter measurement model with the material's true value as a latent variable in a situation where quite serviceable regression and ANOVA calculations have already been developed. The model fit leads to both a weighted estimate of he overall mean, and to tests for equality of means, slopes and variances. Maximum likelihood tests for difference among variances poses a challenge in that the likelihood can easily becoem unbounded. Thus the major objective become to provide a useful test of variance equality.


  1. Proceedings of the Third Berkeley Symposium v.V Statistical inference in factor analysis Anderson, T.W.
  2. Standard practice for conducting an interlaboratory study to determine the precision of a test method, designation : E-691-92 ASTM Standards on Precision and Bias for Various Applications(4th ed.) ASTM
  3. Measurement Error Models Fuller, W.A.
  4. J. Assoc. Off. Anal. Chem. v.63 no.6 Quality Control : Quality Assurance in the Analysis of foods Trace Constituents Horwitz, W.;Kemp, L.R.;Boyer, K.W.
  5. Factor Analysis as a Statistical Method Lawley, D.N.;Maxwell, A.E.
  6. Paper presented in Atlanta at Symposium Analyzing interlaboratory data according to ASTM Standard E-691 Mandel, J.
  7. Communication in Statistics, Simulation v.15 no.1 Computational algorithm for the factor model Pantula, S.G.;Fuller, W.A.
  8. Paper given in Atlanta v.11 Quick fitting covariance models to interlaboratory trial data Proctor, C.H.