In this article, we examine eighth graders' incorrect responses to a specific task on a national standardized assessment of mathematics. The task asked students to write the formula for the perimeter of a given figure as a function of a variable. We focus on incorrect responses to better understand students' difficulties with the algebraic…

We present a logistic function of a monotonic polynomial with a lower asymptote, allowing additional flexibility beyond the three-parameter logistic model. We develop a maximum marginal likelihood based approach to estimate the item parameters. The new item response model is demonstrated on math assessment data from a state, and a computationally…

Schools are scrambling to prepare students for the writing assessments aligned to the Common Core State Standards. In some states, writing has not been assessed for over a decade. Yet, with the use of computerized grading of the student's writing, many teachers are wondering how to best prepare students for the writing assessments that will…

We use a natural experiment to show that the presence of an external examiner has both a direct and an indirect negative effect on the performance of monitored classes in standardized educational tests. The direct effect is the difference in the test performance between classes of the same school with and without external examiners. The indirect…

Recent studies conclude that teachers are important for student learning but it remains uncertain what actually determines effective teaching. This study directly peers into the black box of educational production by investigating the relationship between lecture style teaching and student achievement. Based on matched student-teacher data for the…

Possible distortions in standardized test scores due to guessing, particularly in inner-city situations, are examined. A procedure is presented for estimating true knowledge levels from test scores. Results show scores are overstatements of subject mastery which means many programs are being planned and evaluated using inaccurate data. (Author/DWH)

In this paper, a proposal is made by which a content validity coefficient can be calculated. An example of the use of the coefficient is given, demonstrating that different question combinations in a CSE biology examination in which a choice of questions was given gave different levels of content validity. (Author)

The norm-referenced score scale used by the National Foundation for Educational Research (NFER) is described. The usefulness of standardized scores is explained by a simple numerical example, and the formulas and computations are shown for calculating a mean, a standard deviation, and a deviation or z score. The need for a representative sample is…

A mathematical model for predicting the size of the dental school applicant pool using the number of Dental Admission Test participants is presented. (MSE)

The Naval Academy has instituted a system of averaging the Scholastic Aptitude Test scores of each applicant rather than using the most recent score alone as an admission criterion, having found that average scores are better predictors of academic success at the institution. (MSE)

Some little-understood facts about standardized test scores and how they are reported and interpreted are explained. In particular, the practice of score equating ensures that different test forms have comparable scoring. This and other practices are designed to enhance the equity of testing. (MSE)

A statistical model is presented for processing quantitative admission data that weights equally two or more quantitative elements to ensure that the N most consistently high scorers qualify for the N available positions. The quantitative elements considered are overall grade point average (GPA), GPA in selected courses, and standardized test…

This guide explains the concept of out-of-level testing and suggests a formula for estimating the occurrence of floor effects and ceiling effects, within the context of models for evaluating Elementary Secondary Education Act (ESEA) Title I programs. An analogy explains floor and ceiling effects as if test items are stored in different levels in a…

Efforts to improve the reliability or prediction in dental school admissions by combining several preprofessional academic averages and selected nongraded personal attributes into a single admission index score are outlined. Use of the index is also proposed for use in monitoring less-qualified students' progress. (Author/MSE)

The validity of prediction equations based on American College Testing (ACT) Program test scores and high school grades is investigated. Presented are separate results for prediction equations based on ACT scores only, on high school grades only, and on both kinds of prediction jointly. Also included is a discussion of the relationship between…