Extended caution indices (ECI) specify the degree of confidence that can be placed in an individual's test score by analyzing patterns of item response. Among the most promising of such indices are the standardized ECIs. Contrary to the literature, several instances were found, in a previous study, of nonnormal distributions of ECIs with samples…

The stability of annual normal curve equivalent (NCE) on the Iowa Test of Basic Skills over a period of six years was established in a previous study of status (position within a reference group). This study investigated the relationship between status stability and student achievement level. Third grade students in a suburban school district who…

This study was designed to determine the feasibility of using the Teacher Performance Assessment Instruments (TPAI) to evaluate student teacher performance in northern Nevada by a more accurate and mutually acceptable method. Developed and validated by the University of Georgia, passing the TPAI is required for teacher certification in Georgia.…

The Student Evaluation of Teaching instrument (SET) was administered in grades 4 through 12, to examine its construct validity and usefulness for grades 4 to 6. The original SET was developed for grades 7 to 12. This sample included 419 students in grades 4 to 6, and 612 students in grades 7 to 12. Teachers were graduates of Western Kentucky…

Discusses the flaws in online catalogs that lead to ineffective library use, including confusing subject access, outmoded classification systems, lack of authority control, split files and closed catalogs, lack of both menu and command modes, and lack of circulation information. Recommendations are made for improving library catalogs. (EM)

Reports on the development of a 19-item, Likert-type instrument that measures the attitudes of preservice and inservice teachers toward energy conservation. Focuses on the nine-step process used in developing the instrument. (JN)

Discusses process of continuous assessment that involves many tutors grading papers at Great Britain's Open University and the problem of grade unreliability within such a system. A longitudinal experiment involving grading of papers in one course by all tutors and comparing their grades to determine grading reliability is described. (MBR)

Five common play criteria were applied by subjects to a videotape of nursery school children's behavior, rated separately for occurrence of play. The idea that play is best predicted by a combination of criteria was supported by the finding that, when more criteria occurred simultaneously, the more certainly a judgment of play was implied.…

The author argues that scoring, reporting, and deriving final grades can be considerably assisted by using a computer. He also contends that the savings in time and the computer database formed will allow instructors to determine test quality and reflect on the quality of instruction. (BW)

This book offers a resource guide for qualitative researchers in education, discussing data collection techniques, data analysis, reporting, and the issues of validity, reliability, and ethics. Part 1 reviews the nature and design of qualitative research; it discusses various types of qualitative research (including case studies), and how to…

Reliability and construct validity of over 1,000 mothers' responses to the Disruptive Behavior Rating Scale-Parent Version were studied. Results indicated excellent internal consistency and test-retest reliability on three subscales (Oppositional, Distractible, and Impulsive-Hyperactive) but only marginal reliability on the Antisocial Conduct…

Examines a new approach to university economic impact research that views institutional expenditures as a means to increase the state's skill base, and finds that while the approach yields favorable results for higher education it fails to consider fully the effects of migration. Advises researchers to avoid this approach and use the traditional,…

Responds to Crowley's and Wilson's articles on the reliability of the van Hiele Geometry Test. Discusses the issue of the difficulty of trying to maximize content validity while minimizing measurement error. (YP)

This comparison of regression discrepancy model results for learning disability identification used published IQ-achievement and reliability coefficients and generic, mean values. Use of mean IQ-achievement values resulted in higher discrepancy rates for a clinical sample. Use of both mean IQ-achievement and reliability values resulted in higher…

This study first found that an adaptation of the Pictorial Scale of Perceived Competence and Social Acceptance for Young Children exhibited construct validity and test-retest reliability in assessing preschool children's sensitivity to age-related differences in peers' developmental abilities. Second, children with disabilities were viewed as less…