The goals of revision for the Wechsler Intelligence Scale for Children-Third Edition included enhancement of the factor structure, improvement of subtests, and revision of norms. The researchers reported that the very few items that were found to be biased were replaced. The WISC-III performance of a clinical sample of African American, Caucasian, and Hispanic children was analyzed to determine if the test bias was eliminated as claimed in the goals of the revision.

Two procedures for detection of differential item functioning (DIF) for polytomous items were generalized to detection of differential testlet functioning (DTLF). The methods compared were the logistic discriminant function analysis procedure for uniform and non-uniform DTLF (LDFA-U and LDFA-N), and the Mantel score test procedure. Further analysis included comparison of results of DTLF analysis using the Mantel procedure with DIF analysis of individual testlet items using the Mantel-Haenszel (MH) procedure. Over 600 chi-squares were analyzed and compared for rejection of null hypotheses. Samples of 500, 1,000, and 2,000 were drawn by gender subgroups from the NELS:88 data set, which contains demographic and test data from over 25,000 eighth graders. Three types of testlets (totalling 29) from the NELS:88 test were analyzed for DTLF. The first type, the common passage testlet, followed the conventional testlet definition: items grouped together by a common reading passage, figure, or graph. The other two types were based upon common content and common process. as outlined in the NELS test specification.

This study investigated bias at the item level in six subtests of the California Achievement Tests (CAT). Variability of performance across all individual items of the CAT for fourth graders from three ethnic groups was examined. A two-factor (item scores and ethnicity) ANOVA procedure was used to examine the interaction between Anglo and Hispanic children and between Anglo and black subjects on individual test items of the subtests. Significant F-ratios for the Items x Groups interaction were further examined by using Bonerroni's post-hoc test for the purpose of identifying specific items reflecting cultural bias. A total of twenty-one items was identified as culturally biased. Of these items, sixteen were biased against Hispanics, three items were found to contain bias against blacks, and two items reflected bias against both Hispanic and black children. Of these twenty-one items identified as biased, eighteen belonged to four verbal subtests and three items are part of the two mathematics subtests. In addition to these items identified as being statistically biased, this study suggests that ethnocultural differences exist on overall performance levels between groups. For example, on the verbal subtests, there was a total of only three items on which Hispanic children scored higher than Anglo subjects, and only one item which reflected a better performance by black children than Anglo students. Higher performance levels by Anglo subjects were also noted on mathematic subtests, wherein Hispanic children scored higher on six items than their Anglo counterparts, and black subjects outperformed Anglo children on only one item. These data reflected a tendency of higher performance by Anglo students across all subtests when by an examination of the number of items passed or failed by members of each ethnic group was made. The examination of the verbal subtests additionally showed that Anglos passed sixty-five items, Hispanic children passed twenty-four items and thirty-two items were passed by black subjects. This trend continued on the mathematics subtests, where thirty-one items were passed by Anglo students and seventeen and fifteen items were passed by Hispanic and black children respectively. The findings of this study led to the conclusion that the majority of items on the CAT does not reflect evidence of cultural bias. There were, however, a limited number of items on which either Hispanic or black children out-performed their Anglo counterparts. Implications of these findings were discussed and recommendations were made for future studies to examine bias at the item level.

The purpose of this study was to investigate bias in the areas of examiner disability, and examiner and examinee gender on test scores of selected subtests of the WAIS-R. The subjects of this study were 101 nondisabled University of Arizona Introductory Psychology students who were randomly assigned to treatment and control groups. The treatment group was administered the Arithmetic and Picture Completion subtests of the WAIS-R by a wheelchair bound examiner (a nondisabled examiner posing as a disabled examiner). The control group was administered the same subtests by a nonwheelchair bound examiner. The data were analyzed by two three-way ANOVAs. The three-way analysis of variance tested for significant relationships between main effects of the three independent variables--examiner disability, examiner gender and examinee gender, and the scores of the arthmetic and Picture Completion subtests, and also for two-way and three-way interactive effects of the three variables and the subtest scores. Results indicated that in terms of bias operating to influence the individually administered subtest scores, only examinee gender was found to have a significant effect, with subtest scores for male examinees significantly higher than those for female examinees, on both Arithmetic and Picture Completion subtests. Nonsignificant relationships were found between the other two independent variables of examiner disability and examiner gender, and subtest scores. The presence of a visible, physical disability on the part of the examiner, and the gender of the examiner were not seen to be operating as bias, and were not seen to be impacting on subtest scores. Due to limitations of the study that included instrumentation, interaction of testing and treatment, interaction of selection and treatment, reactive arrangements and history, and due to lack of supportive research in the area of examinee gender as bias influencing test scores, the findings of this study need to be interpreted with caution, and should be considered speculative at best.

Thesis (M.Tech.: Human Resources Management)-Dept. of Human Resources Management, Durban University of Technology, 2008. ix, 132 leaves / The research outlines the statement of the problem, motivation for the study and the related legislated requirements for all research in the field of psychological assessment. It will also take a closer look at related issues supporting the background of psychometric testing and present day themes and trends that are becoming prevalent in contemporary practice. There is an in-depth discussion on Personality which is one of the main foundation components of the Discus Behaviour Profiling System which is personality. Points on, ethics, translation, validity, reliability, test construction and cross cultural matters are also discussed. / M

This research focused on the cultural bias in the items the California Achievement Tests (CAT). Performance variability was examined across all individual items of the CAT for the third graders from four ethnic groups. A sample of 1600 third grade children was randomly selected from population of children attending various elementary schools in the state of Arizona. Four hundred subjects within each ethnic group were matched for sex, ethnicity, and grade level. A two-factor (items scores and ethnicity) ANOVA procedure was used to examine the interaction between the item performances and ethnicity for groups of Anglo and Black, Hispanic, and American Indian on all individual test items of the eight subtest of the CAT. An examination of obtained findings revealed that a total of 31 items were found to be as culturally biased against Hispanic, Blacks, and Native-American children. Of these items, thirty were biased toward American Indians, six items were biased toward Hispanics, and four items were biased toward Blacks. Some items were biased toward more than one ethnic group. Twenty-eight items identified as biased belonged to five of the six language subtests and three items are part of one of the two mathematics subtests. It should be noted that even though most of the items (98%) did not reveal any statistical evidence of bias, there were only four items (1.9%) on which minority group children performed higher than did the Anglo children. The overall direction of the findings would seem to suggest that most of the content of the CAT is free from cultural bias.

Differential item functioning (DIF) can occur across age, gender, ethnic, and/or linguistic groups of examinee populations. Therefore, whenever there is more than one group of examinees involved in a test, a possibility of DIF exists. It is important to detect items with DIF with accurate and powerful statistical methods. While finding a proper DIP method is essential, until now most of the available methods have been dominated by applications to large scale testing contexts. Since the early 1990s, Ramsay has developed a nonparametric item response methodology and computer software, TestGraf (Ramsay, 2000). The nonparametric item response theory (IRT) method requires fewer examinees and items than other item response theory methods and was also designed to detect DIF. However, nonparametric IRT's Type I error rate for DIF detection had not been investigated. The present study investigated the Type I error rate of the nonparametric IRT DIF detection method, when applied to moderate-to-small-scale testing context wherein there were 500 or fewer examinees in a group. In addition, the Mantel-Haenszel (MH) DIF detection method was included. A three-parameter logistic item response model was used to generate data for the two population groups. Each population corresponded to a test of 40 items. Item statistics for the first 34 non-DIF items were randomly chosen from the mathematics test of the 1999 TEVISS (Third International Mathematics and Science Study) for grade eight, whereas item statistics for the last six studied items were adopted from the DIF items used in the study of Muniz, Hambleton, and Xing (2001). These six items were the focus of this study.

Thesis (EdD)--Montana State University--Bozeman, 2008. / Typescript. Chairperson, Graduate Committee: Jayne Downey. Includes bibliographical references (leaves 129-146).

Thesis (Ph. D.)--Georgia State University, 2007. / Title from file title page. Carolyn F. Furlow , committee chair; Phillip Gagne, T. Chris Oshima, Valerie A. Miller, committee members. Electronic text (113 p. : ill.) : digital, PDF file. Description based on contents viewed June 24, 2008. Includes bibliographical references (p. 93-100).

Thesis (Ph. D.)--University of Texas at Austin, 2003. / Vita. Includes bibliographical references (leaves 157-175). Also available in an electronic version.