This book presents an approach to developing performance assessments. It begins with four steps for "Info In" and moves to four "Info Out" modes through which students can make their content understanding explicit for evaluation purposes. The first chapter is an overview of performance assessment in the classroom. Chapter 2 discusses the "Info In"…

The Reading Framework for the 1992 National Assessment of Educational Progress (NAEP) contains the rationale for the aspects of reading assessed in 1992 and criteria for development of the assessment. Developed through a national consensus process as a part of an effort to move assessment forward, the framework presented in the booklet is more…

Item bias has been a major concern for test developers during recent years. The Mantel-Haenszel statistic has been among the preferred methods for identifying biased items. The statistic's performance in identifying uniform bias in simulated data modeled by producing various levels of difference in the (item difficulty) b-parameter for reference…

This study attempts to develop a satisfactory child version of the linguistic task of judging grammaticality. With a nondifferentially reinforced forced-choice procedure, it was found that responses of 48 children (aged 5 and 7) varied as a function of the grammatical complexity of stimulus sentences. (Author/GO)

Describes the construction and use of an experimental instrument for monitoring science teaching opinions, initially conceived for use by pre-service teachers but considered valuable with in-service teachers participating in workshops, seminars and institutes. (EB)

Presents a scale of thirteen positive and seven negative statements designed to measure attitudes of teachers toward behavior modification. Results of the item analysis indicate that the individual statements discriminate very effectively between high and low-scoring groups. (Author/PC)

Three dimensions of emotionality are distinguished: responsiveness, expressiveness, and orientation. Three measures of these dimensions were analyzed in a multitrait-multimethod matrix. There was substantial support for convergent and discriminant validity, and the measures were unconfounded with adjustment or social desirability. Females scored…

A pilot project was undertaken to more clearly determine teacher attitudes toward Black English. A Teacher Attitude Scale (TAS) was constructed, based upon the literature and interviews with teachers and administrators, and administered to 15 teachers. Items significant at the .05 level were retained, forming the final edition of the TAS.…

Critically analyzes selected intelligence test items from commonly used standardized tests. (DW)

Discusses diversity which has arisen in testing of productive skills at GCE O level. Criteria to apply in assessment of foreign language acquisition, and writing and speaking tests in particular, are discussed, as well as the weighting of writing and speaking at O level. (RM)

Reports the procedure followed by members of the Testing and Evaluation Section, Schools Council Modern Languages Project, York, in producing test batteries to accompany the Nuffield/Schools Council language courses, the French course "En Avant" in particular. (RM)

Describes a math review unit in a mastery-learning format designed to ensure that every student entering a physics course can actually use the mathematics essential for the course. (GS)

Criterion-referenced tests (CRTs) are constructed to permit the interpretation of examinee tests performance in relation to a set of well-defined competencies. CRTs are currently used extensively in schools, industry, and the armed services because they provide valuable and different information from norm-referenced tests. Test publishers, school…

Alcohol expectancies have been found to predict later onset of drinking among adolescents. This study examined whether the relationship between level of alcohol use and expectancies is paralleled with cigarette smoking, and attempted to identify the content of smoking expectancies. An instrument to measure the subjective expected utility of…