Presents a formal method for making important one-shot decisions based on an elimination process which uses sequentially acquired information and is based on a conditional belief structure. The routine updating of the structure as information accumulates is based on Dempster's rule of combination. (15 references) (EAM)

An extension of the discussion of loglinear models presents post hoc procedures for statistically evaluating treatment effects, contrasts, and confidence intervals, illustrating methods for main effect and interaction contrasts and paying special attention to odds ratios and their interval estimates. Procedures for treating variables as…

The problems posed for latent variable analysis by populations with different sets of parameter values are discussed. An overview of methodology to address heterogeneity is presented, including a review of structural modeling by the multiple indicator-multiple causes model and the use of the LISCOMP computer program. (SLD)

A new method is presented for determining the minimum number of observations per subject needed to achieve a specific generalizability coefficient. The method, consisting of a branch-and-bound algorithm, enables an investigator to specify an acceptable threshold for generalizability coefficients. (SLD)

Two methods of estimating parameters in the Rasch model are compared. The equivalence of likelihood estimations from the model of G. J. Mellenbergh and P. Vijn (1981) and from usual unconditional maximum likelihood (UML) estimation is demonstrated. Mellenbergh and Vijn's model is a convenient method of calculating UML estimates. (SLD)

Describes habituation and addiction, both psychological and physiological, using simple equations of mathematical model of ideodynamics, optimized to smoking data from Multiple Risk Factor Intervention Trial (MRFIT) program. With only four constant parameters, it was possible to calculate accurate time trends for recidivism to smoking among…

A general approach to least-squares unidimensional scaling is presented. Ordering information contained in the parameters is used to transform the standard squared error loss function into a discrete rather than continuous form. Monte Carlo tests with 38,094 ratings of 261 senators, and 1,258 representatives demonstrate the procedure's…

Relations between latent trait and latent class models for item response data are examined. Conditions are given for two-latent class and two-parameter normal ogive models to agree, and relations between their item parameters are presented. Results are illustrated with data from the Law School Admission Test. (SLD)

Total, direct, and indirect effects in linear structural equation models are examined. Formulas currently given for direct and total effects are reported, and causation is considered. It is concluded that in many instances the effects do not support the interpretations given in the literature. (SLD)

It is contended that this book's conceptually rigorous and complete treatment of meta-analysis is written with the consumer in mind. Some of the models presented are not appropriate for studies that meta-analyses commonly consider. Sampling issues are not covered adequately, and much of the supporting theory is omitted. (SLD)

Conditions necessary and sufficient for the determination of LISREL model solutions are identified. The reasons for indeterminacy of LISREL solutions are discussed, and an index of determinacy is presented and related to the covariance matrix of latent variables. (SLD)

Describes the recall precision trade-off in information retrieval as the deceleration in the proportion of relevant documents retrieved over time. Several mathematical functions sharing this property are examined and it is concluded that the equation that best models recall as a function of time is the logarithm of a quadratic equation. (21…

Problems of estimation and interpretation are discussed that arise when a statistical package, which does not incorporate a dedicated weighted least-squares (WLS) routine, performs WLS regression by misapplication of a case-weighting strategy. A strategy is offered for adjusting WLS regression estimates after a case weighting strategy has been…

Two methods of handling unequal variances in the two-way fixed effects analysis of variance (ANOVA) model are described. One is based on an improved Wilcox (1988) method for the one-way model, and the other is an extension of G. S. James' (1951) second order method. (TJH)

This paper shows that redundancy maximization with J. K. Johansson's extension can be accomplished via a simple iterative algorithm based on H. Wold's Partial Least Squares. The model and the iterative algorithm for the least squares approach to redundancy maximization are presented. (TJH)