Using estimates of item ease and item discrimination, procedures are provided for computing estimates of the reliability and percentage of failing scores for tests assembled from these items. Two assumptions are made: that the average item coefficient will be approximately equal to the average of the estimated coefficients and that the score…

Items from a widely used standardized reading achievement test were rated by trained judges according to the degree to which they required an understanding of hierarchical classification. Two subtests were constructed from subsets of items that were identified by their extreme ratings. Subtest A was judged to require classification operations from…

The Classroom Behavior Rating Scale (CBRS) was developed to provide school psychologists and educators with a measure of learning-related classroom behaviors of elementary school students. The initial item pool consisted of 100 behavioral statements reflecting learning-related behaviors such as attention and persistence. These behaviors were…

The development of the communicative competence test for college business majors described in this paper consisted of three phases. The first phase identified as desirable the following communication behaviors: planning, logic, perception, communication without fear or anxiety, listening, oral communication, conference leadership, and nonverbal…

The literature suggests a "concrete" to "abstract" hierarchy in concept acquisition. However, typical tests do not reflect this hierarchy. Studies have been reported in which an "abstract" option for an item was included in with a "concrete" option and with distractors. This procedure assumes that "abstract" and "concrete" behavior is on a common…

In order to determine if multiple choice spelling tests are as effective as handwritten tests, 265 college freshmen were tested on the same spelling words by traditional oral dictation and then by five-option multiple choice questions. Results were compared to examine the efficacy of multiple choice testing. Of a possible 20 points, the mean score…

Because criterion referenced measurement is a relatively new approach to testing, systematic guidelines for constructing criterion referenced tests (CRTs) have not yet been developed. A number of questions and issues face the test developer who wishes to provide a test that will be helpful in a particular decision-making situation. There are three…

Performance on high and low level objective test items were compared for two groups of business writing students. Students in individualized classes with one teacher for 50 students did as well on high level tasks as did students in traditional classes with one teacher for 25 students. (Author/GDC)

High-level behavioral objectives did not produce greater learning than low-level, contrary to previous findings using study questions interspersed through written prose. Overt use of objectives at both levels produced greater learning, supporting the idea that procedures requiring semantic encoding are instructionally superior to those requiring…

Presents research data to examine two questions: What do teacher-constructed social studies questions measure? How does the skill of critical thinking relate to a teacher-constructed test? (Author)

This study reports the progress in developing and validating an instrument that can be used to measure critical aspects of a parent's attitudes and behavior that influence child development. This instrument should aid in reaching the goal of treating participants in parent education programs as individuals. (Editor/KC)

The teacher who holds enlarged perspectives is regarded more highly than the teacher whose views tend to narrow perspectives. Since children will be influenced by their teacher's attitudes, an examination of the variables related to enlarged perspectives in teachers is pertinent to a broad education. This research looks at those variables among…

Compares the mathematical concepts and language of the 1937, 1948, and 1962 editions of the Essex Local Education Authority mathematics tests with the 1977 Testing of Attainment in Mathematics Survey (TAMS). Sample test items are given along with two tables analyzing the concepts tested and the language used. For journal availability, see SO 506…

Assuming vectors of item responses depend on ability through a fully specified item response model, this paper presents maximum likelihood equations for estimating the population parameters without estimating an ability parameter for each subject. Asymptotic standard errors, tests of fit, computing approximations, and details of four special cases…

New theory and practice in testing is replacing the standard test by the test item bank and classical test theory by item response theory. Eight papers and a commentary are presented in this special issue concerning test item banking. (SLD)