William A. Mehrens

Curriculum‐based measurement: Conceptual and psychometric considerations

This review of published literature and research critically examines the conceptual and psychometric problems associated with curriculum-based measurement (CBM) as they relate to eligibility decision making and programming for special education. It is concluded that although CBM can provide a useful supplement in assessing and remediating academic difficulties, it suffers from many of the criticisms leveled at traditional assessments as well as some unique limitations of its own. It is concluded that CBM, to be of most value, needs to be part of a larger systematic psychoeducational assessment program rather than a replacement for it.

Using multiple choice test items to improve classroom testing of professional accounting students

Abstract The paper develops support for multiple choice testing as an alternative to the traditional “problem” or essay test used for evaluation of academic achievement in accounting courses. Basic measurement concepts are discussed and defined. The relationships between test reliability and validity are explained. The reader is encouraged to develop objective test items for use in the classroom.

Predicting Performance in a Computer Programming Course

COMPUTER programming is one of the fastest growing of the new technical professions. Large numbers of programmers are needed immediately, and it is predicted (Seiler, 1965) that by 1970 there will be a half million new jobs in electronic data processing. Computers are more and more being forced to perform well below their potential for lack of programmers. Automatic data processing has had a significant impact on private business schools; most of the larger ones now have computertraining programs for operators and technicians (Fulton, in press). Too often, however, specialized schools accept students with scant aptitude for computer training; and unqualified graduates ultimately heighten rather than alleviate the critical shortage of personnel in this area. Thus the need is apparent for tests capable of identifying potentially adequate computer programmers. Katz (1962) administered several tests from the Army Classification Battery to 190 trainees in the Army’s Automatic Data Processing Programming course. He used these test results in an attempt to reduce the wasted training time and costs associated with the prevailing high attrition rate. Test scores were correlated with final grade in the course. A combination of verbal and arithmetic reasoning tests from the army battery yielded a correlation with

A Critical Analysis of the Psychometric Properties of the K-ABC:

The K-ABC is reviewed with respect to (a) the manuals, (b) the test construction and norming procedures, (c) reliability evidence, (d) validity evidence, (e) administration and scoring procedures, (f) types of scores and interpretative guidelines, and (g) the bias issue. The psychometric properties of the test are in line with what we would hope for in a newly constructed instrument. Particularly praiseworthy features are the theoretical bases for the instrument, the completeness of the manuals, the treatment and discussion of the difference scores and profile score comparisons, and the number of validity studies that were conducted. Some of the limitations are the lack of long-term stability, reliability, and validity studies, some lack of clarity about how to use the sociocultural norms, and a lack of a definition of bias.

Psychometric Limitations of Guilford's Structure-of-Intellect Model for Identification and Programming of the Gifted

The purpose of this presentation is (a) to analyze critically reseach and promotional literature on Guilfords Structure-of-Intellect (SOI) model as it relates to the cognitive functioning of gifted students, and (b) to evaluate the implications of this analysis for the identification and programming of gifted students. This evaluation is based on a review of promotional literature and available research, and is written from a measurement basis. It is concluded that the promotional literature promises more than it has delivered and that major psychometric limitations seriously limit the SOI models usefulness.

Defending a State Graduation Test: "GI Forum v. Texas Education Agency." Measurement Perspectives from an External Evaluator.

In this article, I present the perspectives of an independent measurement expert regarding the Texas state graduation test, the Texas Assessment of Academic Skills (TAAS). I conclude, among other things, that (a) the TAAS was constructed according to acceptable professional standards; (b) it tests curricular material that Texas Board of Education views important for graduates to have mastered; (c) students have had ample opportunity to learn the material tested on the TAAS; (d) providing instruction on the domain the test samples seems commendable; (e) the TAAS requirement should increase efforts to educate all students in Texas; (f) the TAAS is sufficiently reliable and valid; (g) conjunctive decision making is an appropriate decision making model; and (h) the standard setting was done in a manner that met professional standards.

Social issues in teacher testing

As of 1984 there were about 30 national reports on educational reform, and the states had appointed nearly 300 task forces to study ways to improve education (Cross, 1984). One of the ways chosen was to focus on the quality of teachers (Carnegie Task Force, 1986; Holmes Group, 1986). To improve the quality of teachers, testing programs were implemented for entry into and exit out of teacher education programs, eventually leading to receipt of a license to teach. Current plans are under way to implement a test as requirement for national certification (Shulman, 1987). The vast, fast effort to implement teacher testing programs has not been without controversy, centering around psychometric issues (Mehrens, 1987a), legal issues (Rebell, 1986, 1990), and social issues surrounding the testing of preservice and practicing teachers (hereafter referred to as teacher testing). This article will focus on the social issues. Is the pervasive practice of teacher testing likely to be a positive or negative force in improving education in the United States? As a prelude to the major focus of the article, this section will cover briefly the extensiveness of and support for teacher testing and the reasons for both.

Stability of educational goal orientations held by teachers

Abstract This study assesses the temporal stability of teacher goal orientation and identifies factors associated with change. Seventy-four elementary teachers responded to questionnaires and interviews over a 4-year period. Teacher goal orientation was found to be a relatively stable construct, as few teachers made major and enduring shifts in goal orientation. Subjects identified a variety of factors as contributing to change in goal orientation, with student behavior emerging as a primary impetus for change. Contrary to expectation, shifts in goal orientation were most often exhibited by the more highly experienced teachers in the study. Implications are discussed.

FACTORS INFLUENCING EDUCATORS' DECISIONS ON COLLEGE ATTENDANCE: A STUDY OF A THEORETICAL MODEL'

The processes involved in individual decision making have received considerable attention in the last fifteen years and several mathematical models have been developed to account for, and predict, decisions reached. The more widely accepted decision theory models suggest that people do not make decisions on the basis of the objective probabilities and monetary values of the possible outcomes. Rather, decisions are made on the basis of the subjective probabilities and utility values of the possible outcomes. A subjective probability differs from an objective probability in that the former is really a degree of belief that an individual has that an outcome will occur while the latter is determined through the observation of repetitions of an event. Utility can be thought of as synonomous with preferability or desirability. The utility value of an outcome is the degree to which that outcome is desirable.