In this technical report, we describe the development and piloting of a series of mathematics progress monitoring measures intended for use with students in grade 1. These measures, available as part of easyCBM [TM], an online progress monitoring assessment system, were developed in 2008 and administered to approximately 2800 students from schools…

A study was conducted to extend the sequential probability ratio testing (SPRT) procedure with the polytomous model under some practical constraints in computerized classification testing (CCT), such as methods to control item exposure rate, and to study the effects of other variables, including item information algorithms, test difficulties, item…

Computerized adaptive tests (CATs) are individualized tests that, from a measurement point of view, are optimal for each individual, possibly under some practical conditions. In the present study, it is shown that maximum information item selection in CATs using an item bank that is calibrated with the one- or the two-parameter logistic model…

The convention for selecting starting points (that is, initial items) on a computerized adaptive test (CAT) is to choose as starting points items of medium difficulty for all examinees. Selecting a starting point based on prior information about an individual's ability was first suggested many years ago, but has been believed unimportant provided…

Item response theory (IRT) is based on the assumption that a direct relationship exists between an examinee's total performance on a set of items and the difficulty of each item on the test. The Rasch model represents this relationship mathematically on an equal interval scale. This paper argues that IRT, under the required conditions, provides…

When examinees are allowed to review responses on an adaptive test, can they "cheat" the adaptive algorithm in order to take an easier test and improve their performance? Theoretically, deliberately answering items incorrectly will lower the examinee ability estimate and easy test items will be administered. If review is then allowed,…

A network-flow model is formulated for constructing parallel tests based on classical test theory while using test reliability as the criterion. Practitioners can specify a test-difficulty distribution for values of item difficulties as well as test-composition requirements. An empirical study illustrates the reliability of generated tests. (SLD)

Under certain conditions it may be possible to determine the difficulty of previously untested test items. Although no recipe can be provided, reflections on this topic are presented, drawing on concepts of item banking. A functional constructive method is suggested as having the most potential. (SLD)

In an effort to improve the ways in which words are learned, the Johnson O'Connor Research Foundation (JOCRF) is attempting to determine the difficulty level of all non-technical words in the English language. This item banking project entails: (1) identifying words that should be calibrated; (2) writing a test item for each word; (3) testing the…

A 1992 study by R. Gershon found discrepancies when comparing the theoretical Rasch item characteristic curve with the average empirical curve for 1,304 vocabulary items administered to 7,711 students. When person-item mismatches were deleted (for any person-item interaction where the ability of the person was much higher or much lower than the…

One of the item types in the Listening Comprehension section of the Test of English as a Foreign Language (TOEFL) test is the dialogue. Because the dialogue item pool needs to have an appropriate balance of items at a range of difficulty levels, test developers have examined items at various difficulty levels in an attempt to identify their…

Items with the highest discrimination parameter values in a logistic item response theory (IRT) model do not necessarily give maximum information. This paper shows which discrimination parameter values (as a function of the guessing parameter and the distance between person ability and item difficulty) give maximum information for the…

Effects of the following variables on performance of computerized adaptive testing (CAT) procedures for the partial credit model (PCM) were studied: (1) stopping rule for terminating CAT; (2) item pool size; and (3) distribution of item difficulties. Implications of findings for CAT systems based on the PCM are discussed. (SLD)

When an examination consists in whole or part of constructed response test items, it is common practice to allow the examinee to choose a subset of the constructed response questions from a larger pool. It is sometimes argued that, if choice were not allowed, the limitations on domain coverage forced by the small number of items might unfairly…

Two frequently cited guidelines for writing multiple choice test item stems are: (1) the stem can be written in either a question or statement-to-be-completed format; and (2) only positively worded stems should be used. These guidelines were evaluated in a survey of the test item banks of 13 nationally administered examinations in the physician…