Two approaches of statistical testing are critically reviewed. A new approach, which is a hybrid of the two, is proposed. The new approach requires the researcher to think about the two types of potential inferential errors and an explicit alternative hypothesis of interest. (VM)

This paper reports on the application of Yukl's Management Behavior Survey in 47 elementary schools. Three problems with the instrument are discussed: (1) lack of response; (2) interrater disagreement; and (3) ceiling effects. The dimensionality of the measure is evaluated through factor analysis. (Author/LMO)

Reports the efforts of the California Assessment Program (CAP) to interject critical thinking items into its new statewide social studies tests. Provides definitions of the critical thinking skills tested by the 8th grade social studies CAP test, and lists 100 critical thinking skills vocabulary words. (JDH)

Coordination between research and practice in performance appraisal is essential. Specifically, researchers should focus on utilization methods while practitioners should focus on appraisal content. Cognitive process research may increase raters' ability to judge accurately but fails to address raters' willingness to provide accurate ratings. (BL)

The validity of the Pearson-Johnson taxonomy is supported by the finding that the most correct answers was obtained on the textually explicit scale while the least correct answers occurred on the scriptally implicit scale. (FL)

This study finds that the chief existing teacher evaluation methods (measuring teacher characteristics, student achievement test scores, and teacher performance rating scales) are subjective, open to bias, closed to public scrutiny, and based on irrelevancies. The study advocates evaluation procedures that are performance-based and empirically…

Confirms that in cloze testing, it is unnecessary to use standard size spaces and reveals a high correlation between synonymic scoring and verbatim scoring. Indicates also that a specific probability concepts test is comprehensible and readable by the great majority of students for whom it was devised. (FL)

As a preliminary step in customizing the Stanford Achievement Test, a study was made to compare the test items at each level with the curriculum objectives of a Florida school district, classifying them as either matching, above or below grade level, or unmatching. Results are shown in table form. (TE)

This article lists psychometric problems encountered when measuring change in achievement by improved test scores as is done in performance contracting. Examined are: regression toward the mean, reliability of gain scores, the nature of grade norms, and test content as representation of educational objectives. (SB)

There are many reasons to align tests with curricular standards, but this alignment is not sufficient to protect against score inflation. This report explains the relationship between alignment and score inflation by clarifying what is meant by inappropriate test preparation. It provides a concrete, hypothetical example that illustrates a process…

Successful reform of science education requires careful orchestration of a number of factors including technological developments, organizational issues, teacher preparation and enhancement, and advances in the scientific disciplines themselves. Exemplary innovations in exam practices that assess scientific understanding in new and more…

In June 1995, the Scottish Council for Research in Education began a 5-month study of the Standard Entrance Examination (SET) to the police in Scotland. The first phase was an analysis of existing recruitment and selection statistics from the eight Scottish police forces. Phase Two was a study of two police forces using a case study methodology.…

A study was conducted to determine if providing additional time to complete classroom tests increases the test scores of college students with learning disabilities. Whether the accommodation allows the extra time necessary for students with learning disabilities to process the information, leveling the playing field, or whether it creates a…

The focus of this study was on the validity and feasibility of test accommodation strategies on a small-scale level. Both limited English proficiency (LEP) students and non-LEP students were tested under accommodated and nonaccommodated conditions and their performance was compared. The study was conducted in two public school districts and at…

This study tested whether mental rotation performance of 186 high school students (80 males and 106 females) in grades 9 through 12 in art and nonart classes on Vandenbergs Mental Rotations test (S. Vandenberg and Kuse, 1978) was affected by gender, visual-spatial activities, strategies used while performing the test, and the ease of test taking.…