The effects on Rasch measurement of both response underfit (noise) and overfit (mutedness or superuniformity) are described and illustrated. Misfit is identified by mean-square fit statistics. Person separation and reliability are shown to be deceptive indicators of measurement effectiveness when some items exhibit marked overfit. Theoretical…

The work of R. MacCallum et al. (1992) was extended by examining chance modifications through a Monte Carlo simulation. The stability of post hoc model modifications was examined under varying sample size, model complexity, and severity of misspecification using 2- and 4-factor oblique confirmatory factor analysis (CFA) models with four and eight…

Statistical significance is often misinterpreted to mean replicability or generalizability of results, although a statistically significant difference does not equal a reliable difference. Sample splitting procedures may be a more accurate way of estimating research result generalizability. This type of cross-validation involves randomly dividing…

A study investigated the impact of "thought over time" as a variable that may influence recall of discrepant information encountered in the media. Earlier studies indicated that human subjects are capable of improving their recall of "consistent" sets of news information over time without the benefit of additional exposure to…

The analysis of covariance as a procedure for statistical correction of the effects for an extraneous variable, called a "covariate," is presented. An heuristic data set is used to make the discussion of the calculation of ANCOVA partitions easier to follow. A discussion of homogeneity of regression as an essential condition to be met…

The relevance of ethnographic research to evaluation of educational programs is discussed. The author focuses on evaluation questions that deal with finding out what is happening in a program and if what is happening is desirable. Ethnography suits the purposes of evaluators who answer such questions because they want to know how something works…

A sample of prominent psychotherapists were asked to fill out the A-B therapist "type" scale and comment on their possible differential effectiveness in treating schizoid/schizophrenic versus neurotic patients. The data suggest that B therapists desire and seek more complex and exciting sensory-cognitive inputs during therapy hours than A…

Multinomial-response models are available that correspond implicitly to tests in which a total score is computed as the sum of polytomous item scores. For these models, joint and conditional estimation may be considered in much the same way as for the Rasch model for right-scored tests. As in the Rasch model, joint estimation is only attractive if…

This paper discusses the applicability and adaptability for evaluating reference interviews in library science of the Arizona Clinical Interview Rating Scale (ACIRS), a process-oriented scale originally developed in 1976 at the Arizona College of Medicine to evaluate the interview performance of medical students. The analysis surveys the content…

While attributions have traditionally been viewed as social cognition variables, some researchers have proposed that individual differences exist in attributional patterns and that these patterns are differentially related to and predictive of various behaviors. A study was conducted to examine the question of attribution style as a trait-like…

This report provides preliminary technical information on a method for analyzing the factors involved in the closing of elementary schools. Included is a presentation of data and a brief discussion bearing on descriptive statistics, reliability, and validity. An intercorrelation matrix is also examined. The method employs 9 factors that have a…

In today's complex computer-based teaching (CBT)/computer-assisted instruction market, flashy presentations frequently prove the most important purchasing element, while instructional design and content are secondary to form. Courseware purchasers must base decisions upon either a vendor's presentation or some published evaluator rating.…

This paper describes how a naturalistic inquirer can demonstrate rigor, or trustworthiness, and how an auditor might assess it. Looking at both inquiry products and processes, Part 1 of this paper describes how an audit may assess whether an inquiry's findings and conclusions are grounded in the raw data, and whether appropriate conventions were…

This paper discusses four criteria for selecting covariates when utilizing the analysis of covariance in quasiexperimental research: completeness, causation, independence, and fallibility. Completeness is the extent to which the known or suspected sources of bias are accounted for by covariates used in an experiment. Causation is the requirement…

Noting that public opinion polls are here to stay, this report focuses on the difficulties of good poll reporting in newspapers that use (1) syndicated polls; (2) private polls from political candidates, or (3) their own polls, and argues that newspapers must learn how to report them well. The first section of the report traces the history of…