An information-correction method for testlet-based tests is introduced. This method takes advantage of both generalizability theory (GT) and item response theory (IRT). The measurement error for the examinee proficiency parameter is often underestimated when a unidimensional conditional-independence IRT model is specified for a testlet dataset. By…

For decades, standardized reading comprehension tests have consisted of a series of passages and associated multiple-choice questions. Although widely used in and out of the classroom, there continues to be considerable disagreement regarding how or whether such tests have net value in the service of advancing educational progress in reading. This…

In this study, we describe the development of rapid reading measures, sentences presented to students in a nearly subliminal manner, with a literal comprehension question asked following their removal. After administering alternate forms of these measures to students, we present the results from three statistical analyses to ascertain their…

Discusses, for the benefit of research workers and other test users, the ongoing controversy concerning the relative merits of conventional test theory and Rasch scaling in the construction of reading tests. Concludes that a great deal of further research is required to see whether these approaches are educationally valid. (JD)

The purpose of this study was to document the instrument development of maze measures for grades 3-8. Each maze passage contained twelve omitted words that students filled in by choosing the best-fit word from among the provided options. In this technical report, we describe the process of creating, reviewing, and pilot testing the maze measures.…

Reading comprehension involves a number of distinctly different intellectual skills that can be assessed if the proper techniques are employed. As part of a reading assessment system, two measures of literal comprehension were developed: the Literal Comprehension Details Test (LCDT) and the Paraphrase Reading Test (PRT). Both the LCDT and the PRT…

Discusses issues in reading assessment, including the concern that tests can not measure the reading process. Concludes that tests will never be more than useful indicators; calls for more research and examination of existing and new types of reading comprehension tests. (MM)

A behavioral domain is well-defined when it is clear to both test developers and test users which categories of performance should or should not be considered for potential test items. Only those tests that are keyed to well-defined domains meet the definition of criterion-referenced tests. The greatest proliferation of criterion-referenced tests…

Describes a study to pilot-test a new reading assessment instrument designed to function in a computerized adaptive testing (CAT) environment. Indicates that the measure showed fair internal consistency and correlated well with other tests. Discusses advantages and disadvantages of CAT systems and describes the HyperCAT testing program. (23…

Questions about the appropriateness of the Nelson-Denny Reading Test have led to consideration of the use of a criterion referenced test for assessing students' reading abilities before and after each semester of developmental reading instruction. In constructing such a test, a number of steps should be followed. The first step is to clarify the…

Although informal reading inventories are widely used, they are not without their shortcomings, regardless of whether they have been published commercially or have been constructed by the teacher. There are at least two significant weaknesses in inventories developed by the teacher: (1) passages selected randomly from the graded basal readers that…

Analyzes four informal reading inventories to determine the text type of each passage, whether narrative passages are well formed, and whether expository passages are well organized. Finds almost half the narratives poorly formed. Concludes that the lack of continuity in text type and organization could result in children's comprehension scores…

Criterion-referenced testing was used to complement norm-referenced procedures in a revision of a university's English-as-a-Second-Language placement test for reading. Test validation results indicated that the revised test better matched the university's program and included more items related to the content and skills that students were…

Three different item-bank maintenance procedures were compared for the purpose of investigating their cumulative effects on a series of longitudinal equatings of tests measuring the same characteristic. Item data used resulted from Rasch one-parameter latent trait test calibrations. Comparison of the results in the various final banks were made.…

The autonomous model of literacy--including its two features autonomy of text and autonomy of skill, which have been particularly influential in reading assessment--does not account for important aspects of reading. The pragmatic model of literacy accounts more adequately for how readers respond to text. An analysis of the 1976 and 1987 versions…