The tests we use to evaluate student achievement may well be sound measures of what students know, but they are faulty indicators at best of how well they have been taught. A remedy to this this situation of judging teachers by the performance of their students on high-stakes tests may be in hand already. We should look to the methods successfully…

Most implementations of the Mantel-Haenszel differential item functioning procedure delete records with missing responses or replace missing responses with scores of 0. These treatments of missing data make strong assumptions about the causes of the missing data. Such assumptions may be particularly problematic when groups differ in their patterns…

The use of standardized tests for outcome assessment has grown dramatically in recent years. Two driving factors have been the No Child Left Behind legislation, and the increase in outcome assessment measures by accrediting agencies such as AACSB, the international accrediting body for business schools. Despite the growth in usage, little effort…

The aim of the present research was to compare the use of multiple-choice questions (MCQs) as an examination method, to the examination based on constructed-response questions (CRQs). Despite that MCQs have an advantage concerning objectivity in the grading process and speed in production of results, they also introduce an error in the final…

Paul Newton (2010), with his characteristic concern about theory, has set out two different ways of thinking about the basis upon which equivalences of one sort or another are established between test score scales. His reason for doing this is a desire to establish "the defensibility of linkages lower on the continuum than concordance."…

Recently, educational researchers and practitioners have turned to value-added models to evaluate teacher performance. Although value-added estimates depend on the assessment used to measure student achievement, the importance of outcome selection has received scant attention in the literature. Using data from a large, urban school district, I…

This article presents the author's rejoinder to thinking about linking from issue 8(1). Particularly within the more embracing linking frameworks, e.g., Holland & Dorans (2006) and Holland (2007), there appears to be a major disjunction between (1) classification discourse: the supposed basis for classification, that is, the underlying theory…

"Linking" is a term given to a general class of procedures by which one represents scores X on one test or measure in terms of scores Y on another test or measure. A recent taxonomy by Holland and Dorans (2006; Holland, 2007) organizes the various types of links into three broad categories: prediction, scale aligning, and equating. In…

Newton's article (2010) makes three main contributions to the literature. First, it is transatlantic, bringing together literatures that have been dealing with similar problems, using sometimes different methods and certainly with distinctive educational, cultural perspectives. He points out that neither of these literatures has all of the…

The article "Thinking About Linking" by Newton (2010) presents a novel philosophical perspective on the way that educational assessments should be linked. Newton starts by describing the linking framework as it was characterized in various publications and identifies a cross-cultural dimension in the definitions and uses of test…

In this study, we describe a framework for monitoring rater performance over time. We present several statistical indices to identify raters whose standards drift and explain how to use those indices operationally. To illustrate the use of the framework, we analyzed rating data from the 2002 Advanced Placement English Literature and Composition…

Although the Angoff procedure is among the most widely used standard setting procedures for tests comprising multiple-choice items, research has shown that subject matter experts have considerable difficulty accurately making the required judgments in the absence of examinee performance data. Some authors have viewed the need to provide…

In this article, we introduce a person-fit statistic called the hierarchy consistency index (HCI) to help detect misfitting item response vectors for tests developed and analyzed based on a cognitive model. The HCI ranges from -1.0 to 1.0, with values close to -1.0 indicating that students respond unexpectedly or differently from the responses…

This study demonstrates that multiple matrix sampling procedures can be used to collect assessment data efficiently, unabstrusively, and reliably. (MB)

Describes the Dutch rating system for test quality, which evaluates a test for seven criteria, and analyses the results of test ratings from the past 18 years. Results show a steady increase in test quality in the Netherlands that can be attributed to use of better tests and declining use of tests of less quality after evaluation. (SLD)