This dissertation was designed to foster a deeper understanding of population invariance in the context of composite-score equating and provide practitioners with guidelines for addressing score equity concerns at the composite score level. The purpose of this dissertation was threefold. The first was to compare different composite equating…

Test equating requires collecting data to link the scores from different forms of a test. Problems arise when equating samples are not equivalent and the test forms to be linked share no common items by which to measure or adjust for the group nonequivalence. Using data from five operational test forms, we created five pairs of research forms for…

The aim of the study was to examine the common items in the mixed format (e.g., multiple-choices and essay items) contain parameter drifts in the test equating processes performed with the common item nonequivalent groups design. In this study, which was carried out using Monte Carlo simulation with a fully crossed design, the factors of test…

We investigate the extent to which language versions (English, French and Arabic) of the same science test are comparable in terms of item difficulty and demands. We argue that language is an inextricable part of the scientific literacy construct, be it intended or not by the examiner. This argument has considerable implications on methodologies…

Sampling errors limit the accuracy with which forms can be linked. Limitations on accuracy are especially important in testing programs in which a very large number of forms are employed. Standard inequalities in mathematical statistics may be used to establish lower bounds on the achievable inking accuracy. To illustrate results, a variety of…

The test equating accuracy of content-representative anchors (subsets of items in common) versus nonrepresentative, but substantially longer, anchors was compared for a professional certification examination. Through a chain of equatings, it was found that content representation in anchors was critical. (Author/GDC)

Five indices are discussed that should logically discriminate between situations in which: (1) the linear equating method (LEM) adequately adjusts for difference between score distributions of two approximately parallel test forms; or (2) a method more complex than the linear equating method is needed. (RL)

A common problem arises in scale transformation when independent estimates of item parameters from two separate data sets must be expressed in the same metric. These item parameter estimates will be different because the metric or scale defined by each independent calibration of the items is different. The problem is frequently confronted in…

Five statistical indices are developed and described which may be used for determining (1) when linear equating of two approximately parallel tests is adequate, and (2) whan a more complex method such as equipercentile equating must be used. The indices were based on: (1) similarity of cumulative score distributions; (2) shape of the raw-score to…

Although there are many advantages to using computer-based tests (CBTs) linked to computer-based instruction (CBI), there are also several difficulties. In certain instructional settings, it is difficult to conduct psychometric analyses of test results. Several measurement issues surface when CBT programs are linked to CBI. Testing guidelines…

In a study to determine whether a shift from Formula scoring to Rights scoring can be made without causing a discontinuity in the test scale, the analysis of special administrations of the Scholastic Aptitude Test and Chemistry Achievement Test and the variable section of an operational form of the Graduate Management Admission Test (GMAT) is…

Applications of item response theory to practical testing problems including equating, differential item functioning, and computerized adaptive testing, require that item parameter estimates be placed onto a common metric. In this study, three methods for developing a common metric under item response theory are compared: (1) linking separate…

The proponents of test disclosure argue that disclosure is a matter of fairness; the opponents argue that fairness is enhanced by score equating which is dependent on test security. This research simulated disclosure on a professional licensing examination by placing response keys to selected items in some examinees' records, and comparing their…

This study investigates the amount of uncertainty added to National Assessment of Educational Progress (NAEP) estimates by equating error under both ideal and less than ideal circumstances. Data from past administrations are used to guide simulations of various equating designs and error due to equating is estimated empirically. The design…

A method is proposed to equate different sets of items administered to different groups of individuals using the Rasch model. A Rasch equating model was formulated to describe one common Rasch scale in different groups with different but overlapping sets of items. The item parameters can then be estimated simultaneously, avoiding different…