A common method of checking person-fit in Bayesian item response theory (IRT) is the posterior-predictive (PP) method. In recent years, more powerful approaches have been proposed that are based on resampling methods using the popular L*[subscript z] statistic. There has also been proposed a new Bayesian model checking method based on pivotal…

In equating, smoothing techniques are frequently used to diminish sampling error. There are typically two types of smoothing: presmoothing and postsmoothing. For polynomial log-linear presmoothing, an optimum smoothing degree can be determined statistically based on the Akaike information criterion or Chi-square difference criterion. For…

Research on equating with small samples has shown that methods with stronger assumptions and fewer statistical estimates can lead to decreased error in the estimated equating function. This article introduces a new approach to linear observed-score equating, one which provides flexible control over how form difficulty is assumed versus estimated…

This study investigates a sequence of item response theory (IRT) true score equatings based on various scale transformation approaches and evaluates equating accuracy and consistency over time. The results show that the biases and sample variances for the IRT true score equating (both direct and indirect) are quite small (except for the mean/sigma…

Score equating based on small samples of examinees is often inaccurate for the examinee populations. We conducted a series of resampling studies to investigate the accuracy of five methods of equating in a common-item design. The methods were chained equipercentile equating of smoothed distributions, chained linear equating, chained mean equating,…

Studied three procedures for estimating the standard errors of school passing rates using a generalizability theory model and considered the effects of student sample size. Results show that procedures differ in terms of assumptions about the populations from which students were sampled, and student sample size was found to have a large effect on…

Information loss occurs when continuous data are grouped in discrete intervals. After calculating the squared correlation coefficients between continuous data and corresponding grouped data for four population distributions, the effects of population distribution, number of intervals, and interval width on information loss and recovery were…

Some procedures for estimating internal consistency reliability may be superior mathematically to the more commonly used methods such as Coefficient Alpha. One problem is computational difficulty; the other is the possibility of overestimation due to capitalization on chance. (Author/CTM)

While major test publishers randomly select school districts for their national norming studies, a survey of "accepting" and "declining" districts supports the hypothesis that self-selection bias results in overrepresentation of districts which already use a specific publisher's tests or instructional materials. (Author/BW)

Multiple matrix sampling is applied to a variation of Angoff's standard setting method. Thirty-six experts (internists) and 190 items were divided into five groups, and borderline examinee performance was estimated. There was some variability in the cutting scores produced by the individual groups, but various components were well estimated. (SLD)

When the precise basis of selection effect on correlation and regression equations is unknown but can be modeled by selection on a variable that is highly but not perfectly related to observed scores, the selection effects can lead to the commonly observed "overprediction" results in studies of predictive bias. (Author/PN)

The suitability of the beta binomial test model for criterion referenced testing was investigated, first by considering whether underlying assumptions are realistic, and second, by examining the robustness of the model. Results suggest that the model may have practical value. (Author/RD)

For the purpose of obtaining data to use in test development, multiple matrix sampling plans were compared to examinee sampling plans. Data were simulated for examinees, sampled from a population with a normal distribution of ability, responding to items selected from an item universe. (Author/LMO)

A Monte Carlo study, which simulated 10,000 examinees' responses to four tests, investigated the effect of item stratification on parameter estimation in multiple matrix sampling of achievement data. Practical multiple matrix sampling is based on item stratification by item discrimination and a sampling plan with moderate number of subtests. (SLD)

Linear conversions were developed relating scores on recent forms of the Graduate Record Examinations. Conversions based on specially selected subpopulations were compared with total-group conversions and evaluated. Conclusions indicated that the data clearly support the assumption of population independence for homogenoeous tests, but not quite…