Despite its convenience, the process of self-report in personality testing can be impacted by a variety of cognitive and perceptual biases. One bias that violates local independence, a core criterion of modern test theory, is the order effect. In this bias, characteristics of an item response are impacted not only by the content of the current item but also the accumulated exposure to previous, similar-content items. This bias is manifested as increasingly stable item responses for items that appear later in a test. Previous investigations of this effect have been rooted in classical test theory (CTT) and have consistently found that item reliabilities, or corrected item-total score correlations, increase with the item's serial position in the test. The purpose of the current study was to more rigorously examine order effects via item response theory (IRT). To this end, the FACETS modeling approach (Linacre, 1989) was combined with the Generalized Partial Credit model (GPCM; Muraki, 1992) to produce a new model, the Generalized Partial Credit FACETS model (GPCFM). Serial position of an item serves as a facet that contributes to the item response, not only via its impact on an item's location on the latent trait continuum, but also its discrimination. Thus, the GPCFM differs from previous generalizations of the FACETS model (Wang&Liu, 2007) in that the item discrimination parameter is modified to include a serial position effect. This parameter is important because it reflects the extent to which the purported underlying trait is represented in an item score. Two sets of analyses were conducted. First, a simulation study demonstrated effective parameter recovery, though measurements of error were impacted by sample size for all parameters, test length for trait level estimates, and the size of the order effect for trait level estimates, and an interaction between sample size and test length for item discrimination. Secondly, with respect to real self-report personality data, the GPCFM demonstrated good fit as well as superior fit relative to competing, nested models while also identifying order effects in some traits, particularly Neuroticism, Openness, and Agreeableness.

Thesis (M. Ed.)--University of Hong Kong, 1982. / Includes bibliographical references (leaf 106-111).

Thesis (Ph. D.)--University of California, San Diego, 2000. / Vita. Includes bibliographical references (leaves 154-163).

Thesis (M.Ed.)--University of Hong Kong, 1982. / Includes bibliographical references (leaves 106-111). Also available in print.

This study compared three methods of detecting test item bias, the chi-square approach, the transformed item difficulties approach, and the Linn-Harnish three-parameter item response approach which is the only Item Response Theory (IRT) method that can be utilized with minority samples relatively small in size. The items on two tests which measured writing and reading skills were examined for evidence of sex and ethnic bias. Eight sets of samples, four from each test, were randomly selected from the population (N=7287) of sixth, seventh, and eighth grade students enrolled in a large, urban school district in the southwestern United States. Each set of samples, male/female, White/Hispanic, White/Black, and White/White, contained 800 examinees in the majority group and 200 in the minority group. In an attempt to control differences in ability that may have existed between the various population groups, examinees with scores greater or less than two standard deviations from their group's mean were eliminated. Ethnic samples contained equal numbers of each sex. The White/White sets of samples were utilized to provide baseline bias estimates because the tests could not logically be biased against these groups. Bias indices were then calculated for each set of samples with each of the three methods. Findings of this study indicate that the percent agreement between the Linn-Harnish IRT method and the chisquare and transformed difficulties methods is similar to that found in previous studies comparing the latter approaches with other IRT methods requiring large minority samples. Therefore, it appears that the Linn-Harnish IRT approach can be used in lieu of other more restrictive IRT methods. Ethnic bias appears to exist in the two tests as measured by the large mean bias indices for the White/Hispanic and White/Black samples. Little sex bias was found as evidenced by the low mean bias indices of the male/ female samples and the fact that the male/female mean bias indices were lower than those of the White/White in 33% of the samples.

The purpose of this study was to examine the effects of behavioral checklist critical item content on subsequent global, Likert-type ratings. It was hypothesized that assessment center participants rated with positive critical items would receive higher scores on subsequent global ratings than would participants rated with negative critical items. Additionally, it was hypothesized that volunteers would receive better ratings than nonvolunteers. Finally, it was hypothesized that behavioral ratings would show less susceptibility to halo effect than global ratings.

This research is an investigation into the Apil B as a culture fair assessment tool for the purpose of recruitment and selection. The Employment Equity Act stipulates that "psychological testing and other similar assessments are prohibited unless the test or assessment being used :-a) has been shown to be valid and reliable b) can be fairly applied to all employees and c) is not biased against any employee" (Employment Equity Act, 1998). The primary objective of this research is to evaluate the Apil B with regard to validity. reliability, cultural fairness, and bias via a consideration of the attitudes, opinions, and feelings of psychometric test administrators in the Kwa-Zulu Natal region. The secondary objective is to ascertain whether the Employment Equity legislation has influenced the attitudes of test administrators toward psychometric testing, and the implications for psychometric testing in South Africa following the Act. The sample in this research consists of 20 qualified test administrators of the Apil B. The administrators are affiliated with the following companies : Beacon, Durban Electricity, Profiled Appointments, Mondi, McCann and Associates, Saunders and Associates, Tetrapak, lthaJa, and Mangosuthu Technikon. Qualitative and quantitative methods are used. A self-administered questionnaire is used to investigate the attitudes, opinions, and feelings of the respondents toward the Apil B as a culture fair assessment, specifically in relation to validity, reliability, cultural fairness, and bias. Although most of the questions are qualitative, quantitative questions are also included. Therefore, this research requires numerical data as well . The quantitative questions include yes and no responses, as well as rating scales. The quantitative data supplements the qualitative data and therefore facilitates a more concrete data base. The results indicate that all respondents feel that the Apil B is valid, reliable, culture fair and fairly applied . Respondents feel more positively toward the new updated psychometric tests. They expressed greater confidence in terms of knowing which tests are inappropriate and which tests are relevant to specific jobs. It is also evident that the Employment Equity Act has strongly influenced the attitudes of respondents, in a very positive light, toward the use of psychometric testing. The respondents do however mention some concerns with regard to language based tests, and the ethical use of tests. According to the research conducted, the Apil B is regarded as a valid, reliable, and culture fair assessment tool in the opinion of the respondents utilised in this study. It is therefore recommended that the Apil B is used in organisations as a test that does not discriminate against any culture or subculture. / Thesis (M.A.)-University of Natal, Durban, 2000.

Psychology in South Africa is facing a dire need for valid and reliable mental health instruments for all its citizenry. There presently exists a reliance on instruments of foreign origin. Very often such instruments are used without their psychometric properties having been tested in the local setting. The present study employed a multi-stage process for translating the General Health Questionnaire (GHQ) into Zulu. A simplified English version of the GHQ and the translated Zulu version were subsequently administered to a sample of two hundred and fifty seven (257) bilingual high school students. The data from this sample was used to assess the equivalency between the Zulu version and the English version. At the scale level, both versions of the GHQ showed adequate internal consistency and reliability. Item analysis revealed certain differences between the two versions. Possible explanations regarding semantic differences are discussed. Substantial overlap between the factor solutions of the two versions was found. These factor solutions were found to correspond well with those recorded in the literature. The present sample scored much higher on the GHQ than foreign samples do. Suggestions for raising the cutting scores for South African samples are made. On the whole, the Zulu version displayed evidence of reasonable equivalence to the English version. A comprehensive research programme for the GHQ in South Africa is presented. / Thesis (M.A.)-University of Natal, Pietermaritzburg, 1996.

There is a dire need for psychological tests to be standardized for the African child and for appropriate norms to be developed. This study is part of two larger studies that began in 2008, namely the National Early Learning Standards Age Validation Study (NELDS) (Kvalsvig, Govender & Taylor; 2008) (Principal Investigator – Dr. S. Kauchali), and the INSTAPA Work Package 6 Study (Kvalsvig, Govender & Taylor; 2008) (Principle Investigator – Dr. J.D. Kvalsvig). This study considers aspects of the Kilifi Development Inventory (KDI) as a measurement tool used to provide a structured and standardized procedure for monitoring the cognitive development of children from the age of 6 to 35 months. It is situated in the positivist paradigm and has adopted the classical test theory in attempting to establish the reliability of the Kilifi Developmental Inventory. Stratified purposive sampling techniques were used to select the South African sample, which comprised 120 participants categorized by gender (males and females), age (6 – 18 months and 18 – 35 months) and area (urban and peri urban/rural). The internal reliability of the KDI was investigated. Results from the data analysis (using SPSS 18.0) suggest that the measurement tool as a whole had inadequate internal reliability and therefore, further investigations and modifications have to be undertaken in certain subtests of the measurement tool before the assessment tool can be standardized for the South African population. / Thesis (M.Soc.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.

Increasing focus is being placed on fair assessment practices in South Africa. Most psychological tests used in this country were designed and developed in other countries. Research has shown that when using these tests on a different target population, it is important to determine the suitability of the test for that population. This involves investigating reliability, validity and possible sources of bias. It is equally important to have local normative data with which scores can be compared and interpreted. Research in this area has been limited. The purpose of this study was to undertake such research by evaluating a widely used test, the Raven's Coloured Progressive Matrices (RCPM), and its existing norms, for Zulu children in South Africa. The study also aimed to provide raw data, which could be used to assist with the compilation of local norms. The test was administered to a convenience sample of 522 primary school children between the ages of 5 and 12 years old. The sample was divided into two groups to represent two different geographical locations: rural and urban. The sample consisted of 284 rural children and 237 urban children. There were 263 males and 259 females in the sample. Findings indicated good test reliability for this sample. Test validity however could not be confirmed as possible sources of bias at the content and item levels were shown. Age had a significant effect on performance, but level of education appeared to be the strongest predictor of performance on the RCPM for this sample. Significant gender differences were found. There was a tendency for males to outperform females on the test. Location had a significant effect on scores, where urban children performed better than rural children. Comparison of Zulu sample scores with those produced by the British standardisation sample, revealed a large discrepancy in performance between the two different cultural groups. The Zulu sample mean scores were significantly lower than the British mean scores. These findings highlight the need for locally developed normative data. The validity of the test for the sample requires further investigation. Results indicate that test adaptation may be required for Zulu children. / Thesis (M.A.)-University of KwaZulu-Natal, Pietermaritzburg, 2005.