A Monte Carlo simulation study was conducted to investigate the accuracy of several item parameter standard error (SE) estimation methods in item response theory (IRT) when the marginal maximum likelihood (MML) estimation method was used and the distribution of the underlying latent trait was nonnormal in the two-parameter logistic (2PL) model. The manipulated between-subject factors were sample size (N), test length (TL), and the shape of the latent trait distribution (Shape). The within-subject factor was the SE estimation method, which includes the expected Fisher information method (FIS), the empirical cross-product method (XPD), the supplemented-EM method (SEM), the forward difference method (FDM), the Richardson extrapolation method (REM), and the sandwich-type covariance method (SW). The commercial IRT software flexMIRT was used for item parameter estimation and SE estimation. Results showed that when other factors were hold equal, all of the SE methods studied were apt to produce less accurate SE estimates when the distribution of the underlying trait was positively skewed or positively skewed-bimodal, as compared to what they would produce when the distribution was normal. The degree of inaccuracy of each method for an individual item parameter depended on the magnitude of the relevant a and b parameter, and were affected more by the magnitude of the b parameter. On the test level, the overall average performance of the SE methods interact with N, TL, and Shape. The FIS was not viable when TL=40 and was only run when TL=15. For such a short test, it remained to be the “gold standard” as it estimated the SEs most accurately among all the methods, although it requires relatively longer time to run. The XPD method was the least time-consuming option and it generally performed very well when Shape is normal. However, it tended to produce positively biased results when a short test was paired with a small sample. The SW did not outperform other SE methods when Shape is nonnormal as the theory suggests. The FDM had somewhat larger variations when TL=1500 and TL=3000. The SEM and REM were most accurate among the SE methods in this study and appeared to be a good choice both for normal or non-normal cases. For each simulated condition, the average shape of the raw-score distribution was presented to help practitioners better infer the shape of the underlying distribution of latent trait when the truth about the latent trait distribution shape is unknown, thereby leading to more informed decisions of SE methods using the results of this study. Implications, limitations and future directions were discussed. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester 2018. / April 9, 2018. / Includes bibliographical references. / Insu Paek, Professor Directing Dissertation; Fred Huffer, University Representative; Betsy Jane Becker, Committee Member; Yanyun Yang, Committee Member.

In research synthesis, researchers may aim at summarizing peoples' attitudes and perceptions of phenomena that have been assessed using different measures. Self-report rating scales are among the most commonly used measurement tools to quantify such latent constructs in education and psychology. However, self-report rating-scale questions measuring the same construct may differ from each other in many ways. Scale format, number of response options, wording of questions, and labeling of response option categories may vary across questions. Consequently, variations across the measures of the same construct bring about the issue of comparability of the results across the studies in meta-analytic investigations. In this study, I examine the complexities of summarizing the results of different survey questions about the same construct in the meta-analytic fashion. More specifically, this study focuses on the practical problems that arise when combining survey items that differ from one another in the wording of question stems, numbers of response option categories, scale direction (i.e., unipolar and bipolar scales), response scale labeling (i.e., fully-labeled scales and endpoints-labeled scales), and response-option labeling (e.g., "extremely happy" - "completely happy" - "most happy", "pretty happy", "quite happy"- "moderately happy", and "not at all happy" - "least happy" - "most unhappy"). In addition, I propose practical solutions to handle the issues that arise due to such variations when conducting a meta-analysis. I discuss the implications of the proposed solutions from the perspective of meta-analysis. Examples are obtained from the collection of studies in the World Happiness Database (Veenhoven, 2006), which includes various single-item happiness measures. / A Dissertation submitted to the Department of Educational Psychology and Learning Systems in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Fall Semester 2018. / November 9, 2018. / meta-analysis, scale transformations, survey / Includes bibliographical references. / Betsy J. Becker, Professor Directing Dissertation; Fred W. Huffer, University Representative; Yanyun Yang, Committee Member; Insu Paek, Committee Member.

Thesis (Ed.D.)--Boston University.

This study investigated the effects of test speededness and context of items on Rasch model parameters. Two questions were formulated for this investigation: "To what extent are the Rasch model parameters affected by partially speeded test administration?" and "To what extent are the Rasch model item difficulty parameters affected by changes in item position and experimental calibration situation?" / The investigations were conducted using data from a mandatory statewide basic skills examination of eighth graders. The effects of test speededness were examined by comparing the Rasch model parameters of samples which were specifically constructed to exhibit various levels of test speededness. Context effects were investigated by longitudinal comparisons of item calibration values obtained in different item positions and experimental versus regular situations. / This research has revealed that the item parameters of a partially speeded test were significantly changed at different levels of speededness. No demonstration of the effects on ability estimates was found. The analyses of context effects showed that item difficulty estimates were affected by changes in item position and calibration situation. / Source: Dissertation Abstracts International, Volume: 43-12, Section: A, page: 3885. / Thesis (Ph.D.)--The Florida State University, 1982.

Source: Dissertation Abstracts International, Volume: 40-10, Section: A, page: 5410. / Thesis (Ph.D.)--The Florida State University, 1979.

The purpose of this study was to examine the effect of the Developmental and Deviance Models on services to mentally retarded individuals by addressing the following questions: Do client characteristics such as level of intelligence or skill levels affect the placement of mentally retarded individuals in group homes in the community? How completely are their identified service needs met in community settings? What types of variables affect the extent to which services, documented as needed, are provided? What types of variables affect the extent to which provided services are congruent with the goals in the individual habilitation plans? How often, once placed do group home residents move to residential settings of less or greater restrictiveness? / A total of 477 group home residents were randomly selected from group homes stratified by size (the number of residents: 4-6, 7-10, 11-20, and more than 20 residents). For each client, an instrument was completed which dealt with client characteristics, the adequacy of the community service delivery system and the rate and type of movement experienced by these clients once placed in a group home. / The results indicate that 41% of the sample are severely and profoundly retarded and that these individuals possess minimal, if any, ambulation or communication abilities or self-care skills (bathing, toileting, self-feeding, drinking and dressing). / More services were provided to these individuals than were documented to meet treatment goals of the ihp. Eighty-seven percent of the 1,512 services documented as needed to meet treatment goals in the ihp were provided. A total of 1,945 services were actually received by group home residents, many not documented to meet ihp treatment goals. The service district and the group home size in which the individual lived were the only variables to significantly explain the variance in the amount of ihp-related services which were provided. The service district, level of intelligence and age were significant in explaining the percentage of received services which were congruent with ihp treatment goals. / Client movement was minimal, moves from group homes to institutions decreased as the admission's law made institutional avoidance possible. Moves to settings less restrictive than a group home increased as funding for community support services increased. Too few moves were made to settings more restrictive than group homes to make further analysis possible. / Although this study was only a preliminary examination of the effect of the developmental and deviance models on services affecting the mentally retarded, the developmental model is enjoying at least moderate influence. Services, based on identified need, are provided to an almost complete extent; only administrative differences in service districts and home sizes affect the provision of services. Many unskilled individuals, formerly confined to institutions, are living in the community. In all, this study reflects to a surprising degree, the influence of a progressive perspective on the services of mentally retarded residents of group homes. / Further study is needed to examine the habilitation planning process to determine whether the treatment goals are adequately developed for mentally retarded individuals. Additional investigations should also be undertaken to determine what the service district variable is measuring; it is likely that this variable could be measuring resources, resource utilization, and the differing management structures in the field. / Source: Dissertation Abstracts International, Volume: 41-10, Section: A, page: 4372. / Thesis (Ph.D.)--The Florida State University, 1980.

In this study the technique of using the diagnostic study as a tool for improving supervisory effectiveness was investigated. A specially designed diagnostic study was administered to a group of industrial supervisors to determine what, if any, effects the knowledge of perceived supervisory behavior by the subordinates had in changing the behavior of the supervisor as measured by a second administration of the same diagnostic study. / A group of 672 industrial supervisors with different levels of responsibilities constituted the population used in the study. This group was first asked to rate the behavior of their supervisors who then were given the results of the subordinate's ratings. With the exception of an interpretive session explaining the meanings of the ratings, no formal treatment was made to bring about any change in the behavior of the supervisors. Three months later the diagnostic study was repeated to determine if any change in behavior, as perceived by the subordinates, had taken place on the part of the supervisors. / Using a test-retest design, change in behavior of the supervisors as perceived by subordinates was measured. Changes in behavior are attributed to the knowledge gained by the supervisors as a result of the initial diagnostic study. Within the limitations of the study the following conclusions seem justified. (1) The diagnostic study provided useful criteria as base measures for evaluating supervisory behavior. (2) The diagnostic study provided a measurement of behavior in relation to these criteria. (3) The record of variance of actual performance from standards identified areas of supervisory behavior needing change. (4) Once areas of supervisory behavior that varied from standards as established by superintendents were identified by the supervisor, behavior patterns changed. / Source: Dissertation Abstracts International, Volume: 41-10, Section: A, page: 4372. / Thesis (Educat.D.)--The Florida State University, 1980.

The primary purpose of this study was to apply the Rasch measurement model to investigate sex-item bias in Chemistry, Biology and English tests from the Joint Higher Education Entrance Examination (JHEEE) in Thailand. This research proposed to examine three questions. (1)Does each test measure the same things for both males and females? (2)Are test items within each test potentially biased against either males or females? (3)If potential sex-biased items exist, what are the content characteristics of the items, identified by the Rasch model as being potentially biased? / To answer these questions, male and female students were randomly selected from the population of examinees who took the tests in science area in the 1979 JHEEE. The test data obtained from the selected samples were analyzed by the BICAL program. / The definition of item bias in this study was based on item difficulty and item fit statistics derived from calibrations done by the Rasch model. The criteria for identifying discrepant items were two standard deviations of the differences of item difficulties and of fit statistics which were computed from the random halves of male and female samples. Each criterion was used to establish a band on the 45(DEGREES) line of the scattergrams for difficulty values and fit mean squares of the total male and female samples. Items which fell outside the band were declared as potentially biased items and their content characteristics were examined. / The major findings from this study indicated that: (1)Each test measured approximately the same things for both males and females. The empirical evidence was supported by the patterns of the scattergrams and high correlation coefficients of the item difficulties. (2)There were some items within each test identified as potentially biased against males or females. The English test produced the largest number of potentially biased items, followed by the Biology and Chemistry tests. (3)The content analyses of potentially biased items were not absolutely conclusive. However, the results indicated close agreement between the content characteristics of those items identified as potentially biased by the item difficulty criterion and the judges' ratings relative to sex bias. / The reseacher suggested that the potentially biased items which were identified by the item difficulty criterion and also confirmed by the judges' evaluation should be removed or rewritten so as to eliminate the bias. Also, the researcher recommended the use of the item difficulty criterion to explore the potentially biased items in other entrance tests using other samples having different characteristics. / Source: Dissertation Abstracts International, Volume: 42-03, Section: A, page: 1112. / Thesis (Ph.D.)--The Florida State University, 1981.

Source: Dissertation Abstracts International, Volume: 40-10, Section: A, page: 5412. / Thesis (Ph.D.)--The Florida State University, 1979.

Source: Dissertation Abstracts International, Volume: 40-09, Section: A, page: 5017. / Thesis (Ph.D.)--The Florida State University, 1979.