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.

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

This study was designed to examine the variability among college of education faculty members' grading behavior. A survey consisting of forty items related to the different aspects of the grading behavior was employed. A total of 95 college of education instructors at the Florida State University participated in the study. The main purpose of the study was to investigate the factorial structure of the grading behavior. It was hypothesized that grading behavior is comprised of three components and each component is multi-dimensional. These components and their factorial structure are: (1) Philosophy Component (Traditionalist, Eclectivist, Relativist), (2) Purposes of Grading Component (Institutional-oriented, Instructional-oriented, Student-oriented), (3) Systems of Grading Component (Pass-fail, Competition, Motivation). / A secondary purpose of the study was to examine the interrelationships among factors underlying the grading behavior components. The following relationships were hypothesized to exist: (1) A positive relationship between the traditionalist philosophy and the institutional and the instructional purposes. (2) A positive relationship between traditionalist philosophy and motivational and competition methods. (3) A positive relationship between relativist philosophy and instructional purposes. (4) A positive relationship between relativist philosophy and motivation methods. (5) A positive relationship between the eclectivist philosophy and student-oriented purposes. (6) A positive relationship between eclectivist philosophy and motivational and pass-fail methods. (7) A positive relationship between student-oriented purposes and motivational and pass-fail methods. (8) A positive relationship between institutional-oriented purposes and competition methods. (9) A positive relationship between instructional-oriented purposes and motivational methods. / Conventional methods of factor analysis and path analyses were employed in carrying out the study. Results of the factorial modeling method were compared with those of the factor and path analysis. The general findings of the factor and path analyses tended to confirm, with one exception, the hypothesized factorial structure of grading behavior and several of the expected interrelationships. Results of the factorial modeling approach were not always consistent with those of the factor and path analysis. The findings, based primarily on the factor and path analyses, are summarized as follows: (1) Only two philosophies were identified. The first was the relativist factor as hypothesized. The second was a combination of the traditionalist and the eclectivist. This factor was named social science traditionalist. All of the factors hypothesized for purposes and methods emerged in the analyses. (2) Significant positive correlations were found between scores on the social science traditionalist factor and all grading purpose factors. Significant negative correlations were found between scores on the relativist factor and student and institutional purposes factors. (3) Scores on the relativist factor had a significant positive correlation with scores on the pass-fail methods factor. Relativist scores correlated positively with scores on the motivation factor but the correlation was not statistically significant. No significant correlations were found between the methods scores and the social science traditionalist scores. (4) Scores on the institutional purposes factor showed a significant negative relationship with pass-fail factor scores and a significant positive correlation with competition scores. / Source: Dissertation Abstracts International, Volume: 41-09, Section: A, page: 4003. / Thesis (Ph.D.)--The Florida State University, 1980.

Source: Dissertation Abstracts International, Volume: 40-09, Section: A, page: 5016. / Thesis (Ph.D.)--The Florida State University, 1979.

This study empirically investigated Rasch model equating for three separate equating situations. In all three equating situations a test was equated to itself through an anchor test. Two internal anchor tests, with varying degrees of fit to the model, were designed to investigate anchor test length and degree of misfit influences upon equating. Random samples were employed for each of three equating situations. In one of the equating situations persons with valid scores, nonzero and not perfect, were included in equating. In the other two situations, persons were deleted on the basis of the Rasch person fit t statistic and on the basis of a heuristic technique. An index of discrepancy between each of the sets of equated scores and a criterion set was calculated. Although further investigation is required, these results should be beneficial in providing a framework for person deletion decisions for equating situations. / Source: Dissertation Abstracts International, Volume: 42-10, Section: A, page: 4416. / Thesis (Ph.D.)--The Florida State University, 1981.

The purpose of this study was to evaluate two linear equating methods where the abilities of nonrandom equating groups vary. The two equating methods investigated in the study were the Tucker and Levine methods for equally-reliable tests and internal anchor forms. The study was particularly relevant for minimum competency and basic skills testing programs since conditions were simulated where successive equating groups increased in mean ability while decreasing in variance. / Six equating groups of approximately 300 examinees each were selected according to scores on the communication section of the 1978 Florida Fifth-Grade State Student Assessment Test. The examinees' scores were selected using a stratified random sampling plan intended to yield six ability groups for scores on the mathematics section of the test. The mathematics score distributions of the groups varied from flatter-than-normal to a skew of -2.420. The average mathematics p-values ranged from .64 to .92. / The mathematics scores of the groups were equated to the three lowest-ability groups using each of the two methods in order to simulate conditions where successive equating groups increase in ability from a group whose scores are used to form the criterion scale. Since a test was equated to itself, the differences between the equated and observed scores of examinees was considered to be error produced by the equating methods. / Six anchor tests were selected to investigate the accuracy of equating with the two methods where the anchor form is not content--representative of the operational form. One anchor form was constructed to be parallel in content with the operational form. The five nonrepresentative forms varied in difficulty from easy to hard. / The Levine method generally produced smaller average, absolute errors of conversions of examinees scores than the Tucker method except where the abilities of the equating groups were similar. The Levine method also generally produced less error for the conversion of group means. However, the Levine method conversion lines were characterized by high, positive intercept values which generally increased as the abilities of the equating groups diverged. Therefore, low scores were equated with considerable error. / Except where the differences between the ability groups were greatest, both equating methods generally equated the scores in the upper 80% of the groups with error of less than one raw score point. Also, the distances between the converted means and the actual raw score means of the equated groups generally did not vary by more than 5%. / In light of these data, it was concluded that the use of either equating method would probably not result in serious error for comparisons of equating group means where the abilities of equating groups vary substantially. Furthermore, either method could be used if critical decision points occur high on the score scale. However, the Levine method would be recommended in each of these cases since it consistently produced less average error in this study. Caution would be advised for the use of either method where critical decision points occur low on the scale. / In spite of the fact that three of the nonrepresentative anchor forms correlated better than .95 with the operational form, each nonrepresentative anchor form produced a substantially larger average error than the representative anchor form for each method. These results demonstrated the necessity of constructing anchor forms which are content-representative of the operational forms. / Source: Dissertation Abstracts International, Volume: 41-11, Section: A, page: 4686. / Thesis (Ph.D.)--The Florida State University, 1980.

Source: Dissertation Abstracts International, Volume: 40-09, Section: A, page: 5016. / Thesis (Ph.D.)--The Florida State University, 1979.

The purpose of this study was to determine whether scores from a third year, General Secondary school exam could be predicted from scores from an earlier exam for students in The Ismailia Educational District in Egypt. Students take the first exam at the end of grade 9 and the second exam at the end of grade 12. Presently, the grade 12 exam is used to determine whether students have "passed" and for admission to higher education. One reason for the investigation was to determine whether the ninth grade scores might have utility in guiding students in the selection of major areas, which occurs at the end of grade 10. / This study differed from others in that the Egyptian exams are free response or essay type rather than multiple choice. In addition, the exams yield several scores, providing both multiple predictor and multiple outcome variables. Multivariate Multiple Regression (MMR) techniques were used to analyze the data. / Subjects for the study were randomly drawn from students in The Ismailia Education District for whom both sets of scores existed. Males (340) and females (303) were distributed over the three major areas--Literary (LM, 234), Mathematics (MM, 158) and Science (SM, 251). / Standard MMR procedures were used. The total sample was split in two to provide for cross validation. Tests for parallelism and coincidence were conducted to determine whether separate models were needed for males and females. Three sets of predictors were used: (1) original variables, (2) the sum of scores from (1), and (3) principal component transformations of (1) and (2). For (1), the logical order of entry differed for LM, MM and SM and was derived on the basis of differences in the three curricula. / Criteria were established for determining the best prediction equation for the several outcomes. For LM, and logical order of variables proved to be best; for MM and SM, the total score proved to be the best predictor. / The R('2) values ranged from low .20's to a high of .52. A sample of 95 percent conficence intervals were constructed for single observations. The intervals were wide, indicating less accuracy than might be desired. / Accuracy of decisions on major groups was estimated by comparing the predicted pass/fail with an actual pass/fail, using a median cut-off score as criteria. Accuracy ratios for males and females in the three groups ranged from .65 to .83. / The basic question dealt with the predictability of scores and was answered with a qualified "yes." This is consistent with prior research. Accuracy of decisions, predicted "pass" to actual "pass", indicated that the scores may play an important role in advising as well as in selection/admission. / The exams are part of the Egyptian Educational System. The scores have not been used for helping students make decisions about major areas of study. To do so implies the existence of trained counselors, without whom the use of scores could take on more importance than they should. Hence, it is obvious that much more work must be done before the scores can assume a new use. / Source: Dissertation Abstracts International, Volume: 41-07, Section: A, page: 3064. / Thesis (Ph.D.)--The Florida State University, 1980.