Value-Added Models (VAMs) are both common and controversial in education policy and accountability research. While the sensitivity of VAMs to model specification and covariate selection is well documented, the extent to which test scoring methods (e.g., mean scores vs. IRT-based scores) may affect VA estimates is less studied. We examine the…

This report summarizes some of the evidence supporting interpretations of scores from the enhanced ACT, focusing on reliability, concurrent validity, predictive validity, and score comparability. The authors argue that the evidence presented in this report supports the interpretation of scores from the enhanced ACT as measures of high school…

Performance evaluation may change employee effort and decisions in unintended ways, for example, in multitask jobs where the evaluation measure captures only a subset of (differentially weights) the job tasks. We show evidence of this multitask distortion in schools, with teachers allocating effort across students (tasks). Teachers are evaluated…

Path models with observed composites based on multiple items (e.g., mean or sum score of the items) are commonly used to test interaction effects. Under this practice, researchers generally assume that the observed composites are measured without errors. In this study, we reviewed and evaluated two alternative methods within the structural…

ACT reporting categories and ACT Readiness Ranges are new features added to the ACT score reports starting in fall 2016. For each reporting category, the number correct score, the maximum points possible, the percent correct, and the ACT Readiness Range, along with an indicator of whether the reporting category score falls within the Readiness…

As a property of test scores, reliability/dependability constitutes an important psychometric consideration, and it underpins the validity of measurement results. A review of interpreter certification performance tests (ICPTs) reveals that (a) although reliability/dependability checking has been recognized as an important concern, its theoretical…

Errors don't exist in our data, but they serve a vital function. Reality is complicated, but our models need to be simple in order to be manageable. We assume that attributes are invariant over some conditions of observation, and once we do that we need some way of accounting for the variability in observed scores over these conditions of…

The aim of this study is to determine the test-retest reproducibility of the seven-item Short-Form Berg Balance Scale (SFBBS) and the five-item Short-Form Postural Assessment Scale for Stroke Patients (SFPASS) in individuals with chronic stroke. Fifty-two chronic stroke patients from two rehabilitation departments were included in the study. Both…

We discuss generalizability (G) theory and the fair and valid assessment of linguistic minorities, especially emergent bilinguals. G theory allows examination of the relationship between score variation and language variation (e.g., variation of proficiency across languages, language modes, and social contexts). Studies examining score variation…

This discussion paper argues that both the use of Cronbach's alpha as a reliability estimate and as a measure of internal consistency suffer from major problems. First, alpha always has a value, which cannot be equal to the test score's reliability given the inter-item covariance matrix and the usual assumptions about measurement error. Second, in…

The 12th annual William H. Angoff Memorial Lecture was presented by Dr. Michael T. Kane, ETS's (Educational Testing Service) Samuel J. Messick Chair in Test Validity and the former Director of Research at the National Conference of Bar Examiners. Dr. Kane argues that it is important for policymakers to recognize the impact of errors of measurement…

State testing programs today are more extensive than ever, and their results are required to serve more purposes and high-stakes decisions than one might have imagined. Assessment results are used to hold schools, districts, and states accountable for student performance and to help guide a multitude of important decisions. This report describes…

IQ test scores should be corrected for high stakes decisions that employ these assessments, including capital offense cases. If scores are not corrected, then diagnostic standards must change with each generation. Arguments against corrections, based on standards of practice, information present and absent in test manuals, and related issues,…

The work reported in this paper reflects a collaborative effort of many individuals representing multiple organizations. It began during a session at the October 2008 meeting of TILSA when a representative of a member state asked the group if any of their programs had experienced unexpected fluctuations in the annual state assessment scores, and…

Test score validity takes center stage in the debate over the use of high school exit exams. Scant literature addresses the amount of conditional standard error of measurement (CSEM) present in individual student results on high school exit exams. The purpose of this study is to fill a void in the literature and add a national review of the CSEM,…