Many of the measurement models currently used in testing require that the items that make up the test span a unidimensional space. The assumption of unidimensionality is difficult to satisfy in practice since item pools are arguably multidimensional. Among the causes of test multidimensionality are the presence of minor dimensions (such as test motivation, speed of performance and reading ability) beyond the dominant ability the test is supposed to measure. The consequences of violating the assumption of unidimensionality may be serious. Different item selection procedures when used for constructing tests will have unknown and differential effects on the reliability and validity of tests. The purposes of this research were (1) to review research on test dimensionality, (2) to investigate the impact of test dimensionality on the ability estimation and the decision accuracy of criterion-referenced tests, and (3) to examine the effects of interaction of item selection methods with test dimensionality and content categories on ability estimation and decision accuracy of criterion-referenced tests. The empirical research consisted of two parts: in Part A, three item pools with different dimensionality structures were generated for two different tests. Four item selection methods were used to construct tests from each item pool, and the ability estimates and the decision accuracies of the 12 tests were compared in each test. In Part B, real data were used as an item bank, and four item selection methods were used to construct short tests from the item bank. The measurement precision and the decision accuracies of the resulted tests were compared. It was found that the strength of minor dimensions affect the precision of the ability estimation and decision accuracy of mastery tests, and that optimal item selection methods perform better than other item selection methods, especially when test data are not unidimensional. The differences in measurement precision and decision accuracy among data with different degrees of multidimensionality and among the different item selection methods were statistically and practically significant. An important implication of the study results for the practitioners are that the presence of minor dimensions in a test may lead to the misclassification of examinees, and hence limit the usefulness of the test.

The purpose of the study was to collect predictive validity data on the cognitive subtests and composite of Early Screening Profiles, a screening instrument that will be published in 1990. Data collection involved 135 children, ages 3-6 through 6-11. The scores on Early Screening Profiles were compared to scores on the Achievement Scale of the Kaufman Assessment Battery for Children (K-ABC), the Peabody Picture Vocabulary Test-Revised (PPVT-R), and, for the 85 children in kindergarten or grade one at the time of follow-up testing, a teacher rating scale, Teacher Rating of Academic Performance (TRAP). Time between testing ranged from 5 $1\over2$ to 8 months. For the population studied, statistically significant, strong correlations of.75,.73, and.70 were found between the composite of Early Screening Profiles and K-ABC Achievement, PPVT-R, and TRAP (p $<$.01). Strong or moderate correlations, all significant at the.01 level, resulted when Early Screening Profiles cognitive subtests were compared to criterion subtests. High agreement rates were found for standard scores of one standard deviation above the mean (82%) and one standard deviation below the mean (84%). Comparison of the Early Screening Profiles cognitive composite score with the total scores of all three criterion measures yielded average specificity and sensitivity rates of.80 and.74, respectively, for scores of 115 or higher. For scores of 85 or lower, the average specificity was high (.97) and the average sensitivity rate was modest (.32). No significant differences emerged based on sex. The older group of children scored higher than the younger on the K-ABC Achievement Scale. Research results indicate that the cognitive subtests and composite of Early Screening Profiles show promise of becoming useful and valid additions to the field of early childhood screening.

The purpose of this study was to describe and document those criteria utilized by TEAM members to determine which students receive special education service within the Salem, Massachusetts Public Schools. The central question inherent in this study was "How is special needs status assigned, through the TEAM evaluation process, to children within the Salem Public Schools?" The focus of this study was to describe how members of a special education evaluation TEAM, in one suburban community, went about determining which students were deemed eligible to receive special education services. The process for determining who is eligible for special education services is a source of concern and discussion at all levels of public education. The incidence of special needs placements increases throughout the Commonwealth of Massachusetts and the Salem Public Schools is no exception to this trend. The commonly held admission criteria to special needs programs include lack of student progress combined with a handicapping condition. Both criteria, however, are not defined and are subjective in nature. The design of the study was descriptive research in which the investigator was a participant observer in TEAM meetings scheduled during October and November of 1989. The focus was to record TEAM dynamics and student assessment data to identify and define criteria which impact on the decision making processes at seventeen TEAM meetings. Taped transcriptions were analyzed individually in keeping with content analysis as a prescribed research methodology using simple frequency counts and cross-tabulations. The results of this investigation showed, clearly, that students receive special education service more as a direct result of TEAM dialogue than any concrete entrance criteria or diagnostic findings. This dialogue confirmed five wide-spread assumptions. Pre-determination, Ititerant Services, Parents as Advocates, Somebody to Care, Special Education: A Remedial Program were found to be the predominant attitudes of TEAM participants in deciding who assigns special needs status.

In this age of financial restraint and increasing demands on behalf of consumers, quality is becoming an essential component of any product or service. While this study focuses on aspects of quality within the private rehabilitation segment of the health care industry, there are lessons that may be applied to the field of education. As expectations change and teachers become increasingly accountable for their students' education, issues of quality and cost effectiveness will come to light. Ideally, quality assurance represents a belief that a product or service can exist that satisfies the needs and expectations of both the consumer and producer or the service provider. Traditionally, quality assurance programs in the health care industry focused either on cost containment or risk management. The practitioners' perspective regarding quality is often overlooked. This study identifies benchmarks of quality based on the health care practitioner's perspective. Through the administration of a survey, the researcher identified four categories of benchmarks that were common to physical therapists, occupational therapists, and speech pathologists. These professional groups identified benchmarks that focused on education, communication, attitude, and the success of the therapeutic intervention. This study has potential for the field of education. Like the health care industry, the field of education is also evolving. As people become increasingly concerned about the quality of their children's education and how best their educational dollars are spent, the demands for quality assurance within education will increase.

Due to the impact of Public Law 99-457, many school psychologists will become increasingly involved in the psychological assessment of preschoolers. Since the law does not specify what assessment practices should be followed, current practices vary according to the setting, employer and practitioner. This study addresses the issue of an information need by establishing a database on current assessment practices. Information provided by school psychologists currently involved in the assessment of preschool children in public school settings would help define techniques specifically designed for this age group. To determine current preschool assessment practices, a survey was conducted of 100 school psychologists who work with preschool children enrolled in public school programs for early intervention in the New England region. Site visits to 12 preschool programs in Massachusetts and Connecticut were conducted to interview school psychologists regarding their preschool assessment practices. The survey endeavored to ascertain what assessment techniques are employed, which instruments are perceived to be most effective, and what the factors are that influence the selection of preschool assessment instruments. The role of the school psychologist as a member of an early intervention team was also examined. Results indicated a great variety of tools/techniques in use for assessment. The most frequently used tests/techniques were medical and developmental history, file review, interviews, classroom observation and norm-referenced tests; in sum, preferred measures sample a broader range of behaviors and are more useful in designing effective educational interventions. School psychologists, however, still need more knowledge of curriculum-based measures and dynamic assessment procedures. Results also indicated that tests are selected on the basis of their effectiveness. The most common rationale for selecting a particular tool cited was the match between referral question and intervention. The role of the school psychologist consisted primarily of assessment for classification/placement decisions. School psychologists and teachers were found to have the highest levels of involvement in assessment, followed by parental participation. Implications of the survey are discussed in terms of improving preschool assessment practices, as well as with respect to training of school psychologists.

This descriptive study examined the perceptions of teachers toward grouping and evaluation of pupil learning progress in reading. Instructional grouping decisions by teachers were viewed as part of the evaluation function of teaching. Teacher practices in grouping and evaluation were assumed to represent significant classroom level indicators of equity and excellence in the education provided for children. A major priority for the inquiry was to discover whether the evaluation information collected about students lead teachers to change the practice of ability grouping. Three major research questions guided the study: (1) What criteria do teachers report they use to group pupils for instruction in reading? (2) What are teachers' perceptions of the similarities and differences in ways that they evaluate pupil learning progress across instructional groups for reading? (3) What are teachers' perceptions of ways they use data from their evaluation of pupil learning progress in reading? The study adapts assumptions from interpretive and critical theoretical perspectives complemented by a qualitative research design to describe the perceptions of teachers for grouping and evaluating students in reading. Data are drawn from interviews with 23 experienced teachers representing grades 5 and 6 in ten schools in western Massachusetts. Findings indicate that teachers tend to rely on previous teacher recommendations and the results of their own informal assessments as criteria for grouping students by ability within the classroom. Teachers assign students to within-class ability groups during the first few weeks of school. Students tend to remain in the reading groups to which they were initially assigned. Teachers' evaluation methods and performance criteria vary depending on whether they are in a basal or non-basal reading curriculum. Teachers reported that they expect students placed in higher ability groups to proceed at a faster pace through the reading curriculum and to produce written work that was more detailed and of higher quality than students placed in lower ability groups. Teachers report numerous uses of data from pupil evaluation. The primary uses are for communicating with parents and students and for improving instruction. The study concludes that despite the diverse ways that teachers evaluate student learning progress in Reading, little student mobility occurs across the instructional groups, nor does the evaluation data collected by teachers suggest to them a need to change the practice of ability grouping.

The purpose of this dissertation was to use the technique of meta-analysis to combine and analyze the results from studies examining the peer status of mainstreamed elementary school students with learning disabilities. Twenty-one data sets from fourteen studies were analyzed to determine whether or not learning disabled children in mainstreamed settings occupy a lower status than classroom peers as measured by a sociometric instrument. The variables of gender of both rater and child rated, type of sociometric instrument used, date of publication and degree of integration were also examined. Results showed that learning disabled elementary school children in mainstream settings occupied a significantly lower status than their peers when measured by either peer rating or peer nomination instruments. Children were rated lower in status when a peer rating instrument was used versus when a peer nomination instrument was used. There was a trend toward higher status in students with learning disabilities in studies that were published in the 1980s versus those published in the 1970s. There was also a trend toward higher status in students with learning disabilities who were integrated more than 50% of the time versus those integrated less than 50% of the time.

This study measured the computer competency of New Hampshire high school seniors, using the National Assessment of Educational Progress (NAEP) 1986 assessment. Six different test booklets, cumulatively containing 124 cognitive items and 75 demographic items were used. An additional 35 demographic questions beyond the NAEP items were collected from a survey designed for this dissertation. One hundred and sixty-eight students (95 males, 68 females, 5 unknown) from eight public and two private high schools across the state were sampled based upon an enrollment size distribution. The total enrollment of the sample schools represented 15% of the total state high school enrollment of 52,400 students. 8.4% of the seniors at participating schools were assessed. Essentially all students have completed a one semester computer competency course, as required by state regulations. The data was analyzed utilizing non-parametric statistics for demographics and Z-Tests for comparisons to the NAEP national sample. The microcomputer statistical packages of MINITAB and EXECUSTAT were used. Conclusions. First, the computer competency levels of New Hampshire are significantly greater than both the national average and the higher New England average as measured during the NAEP national survey in 1986 at the 95% confidence level. Secondly, it made no difference whether the computer competency course was taken in (1) In high school or junior high school. (2) A Public or private school. (3) A small, medium or large school. (4) Different types of communities (i.e. city, rural etc.). Third, the cognitive outcome was significantly correlated at the 95% confidence level with: (1) Sex of the student. (2) Number of years of computer usage. (3) The curriculum content of the first course. (4) Semesters of computer courses studied. (5) Time of the first computer course. (6) Word processing usage. Finally, the cognitive outcome was not significantly affected by the following characteristics: (1) Attitude. (2) Self-assessment. (3) Home computer. (4) Programming courses. (5) Timing of last computer class. Students perceived that computers were not integrated into the curriculum as many in the state expected. Classroom computer usage was substantially limited to computer classes. Students overwhelmingly desired more computer usage in classes.

Elementary school teachers serve a key role in the use and interpretation of standardized achievement test batteries (SATBs). If this role is to be properly exercised, teachers must be willing and able to use score information from such tests. Evidence suggests that teachers may not be skilled in making the interpretations that psychometricians intend from SATB score reports. To learn what purposes teachers believe SATB score reports appropriately serve, and what contents and formats teachers find useful in these reports, a survey research study was conducted. A questionnaire was administered to a sample of 671 elementary school teachers from Texas, Illinois, and Massachusetts, focusing on the purposes of SATB score reports in general, the purposes of specific formats and types of score reports presented in the questionnaire, and interpretations of the sample score reports. Background information about the respondents was gathered. Of particular interest was the amount of testing-related training they had experienced. A relationship between amount of testing-related training and teachers' opinions and interpretations of test score reports was hypothesized. Respondents generally approved of the use of SATB score reports for certain purposes, especially those that had no permanent consequences for students and no external evaluative implications for the classroom. Teachers disapproved of the use of SATB score reports for student grading, promotion/retention decisions, and teacher evaluation. Other proposed uses were moderately approved. Respondents were equally supportive of numerical and graphical formats for a class-level report, but strongly preferred a numerical/pictorial version of an individual student score report (i.e., displaying subscores as sets of confidence bands) to a narrative version. Teachers' interpretations of score reports generally did not agree with strict psychometric interpretations. This lack of a psychometric perspective was found regardless of the amount of training in testing issues the respondents had experienced. The author concludes with recommendations for action regarding the design of score reports to meet teachers' needs and psychometricians' intentions more effectively, the training of teachers to match their professional demands, and further exploration of this topic through qualitative research methods.

Special education placement determinations, due to State and Federal legislation, are largely the responsibility of an evaluation team convened to develop an Individualized Education Program (IEP) designed to ensure that each student receives a free and appropriate public education. Decisions about individual students require the sharing and synthesis of student-specific information by members of the Team. In Massachusetts, the lack of specific entry and exit criteria within the definition of a "child in need of special education" does not restrain the interpretation or weight Teams attribute to the student-specific information they use to make placement determinations. Consequently, this variability is frequently cited as a primary contributor to the state's high special education enrollment figures. The purpose of this study was to compare information from the records of special needs students with the reports of evaluation Team members on the importance of the information in determining the amount of special education services a student receives. The study collected student-specific information from the IEP and special education records of 150 special needs students in grades 3, 6, and 9. Questionnaires were mailed to 250 special education directors, parents, special education teachers, school principals, and school psychologists. The questionnaire asked the participants to indicate their feelings about the importance and use of the same student-specific information that was collected from the student records. Research questions that guided this study were designed to compare the predictions about the amount of special education services a student receives based on IEP and student record information with the importance and use of the same information as reported by members of evaluation Teams. Specific findings to suggest that: (1) Within the set of student-specific information analyzed, there is no subset of data can be used to reliably predict the amount of special education a student receives. (2) Between group differences among students suggest that the differences are artifacts of legal procedural requirements rather than specific student characteristics such as handicapping conditions or academic deficits. (3) Team members agreed that reading and math achievement and intelligence are the most important factors in making placement decisions. Mainstreaming opportunities and student behavior were also perceived to be of primary importance. (4) Team members believe that placement decisions are influenced by the Team's composition and are not based primarily on specific student information. The implications and limitations of the study are discussed with recommendations for further research and changes in the Team evaluation process.