South African organisations face the challenge of creating organisations that will engage employees in ways that allow for the optimisation of their strengths. This can be achieved by following a strength-based approach (SBA). An SBA aims to achieve optimisation of human functioning, where talents and strengths are the focus and weaknesses are understood and managed. Although previous research suggests that an SBA has positive influences on individual and organisational outcomes, no instrument exists within the South African context that measures employees’ perception of the extent to which they believe their organisation makes use of their strengths and talents. Recently, a new scale was developed to address this need. However, no studies have been done to see how well the items of this scale function. The objectives of this research were to 1) conceptualise an SBA according to literature, 2) determine whether the items in the SBA scale are unidimensional, 3) to determine the internal validity and reliability of the new SBA scale, and 4) make recommendations for future research. A cross-sectional quantitative research design was used whereby online and hardcopy versions of the questionnaire were distributed to participants. A sample of 699 participants was collected from the Gauteng and North-West provinces. Rasch analysis was used to determine the reliability and validity of the items. Acceptable item reliability was found. Both the item and person separation indices were acceptable. Mean infit and outfit indices for both person and item were acceptable. The seven-point frequency-based Likert scale worked satisfactorily overall, although categories 0, 1, and 2 were under-utilised. Finally, the infit and outfit statistics for all eight items functioned satisfactorily, except for one item. Recommendations are made for practice as well as for future research. / Thesis (MA (Industrial Psychology))--North-West University, Potchefstroom Campus, 2012

Diversos estudos têm demonstrado associação entre o temperamento e os níveis de ansiedade pré-operatória em crianças. Entretanto, a natureza desta associação não é consenso. Como a maioria desses estudos utilizou o questionário EASI (emocionalidade, atividade, sociabilidade e impulsividade) como instrumento para mensurar o temperamento, o presente estudo examina em profundidade as propriedades psicométricas do mesmo utilizando a análise de Rasch, para determinar se o instrumento mede de forma adequada o temperamento de crianças. Além disso, outras análises são aplicadas para explorar novas estruturas fatoriais do instrumento original. A habilidade da nova versão do EASI para discriminar diferentes níveis de ansiedade no período préoperatório imediato em crianças também foi investigada. Trata-se de um estudo transversal, com amostra de conveniência, onde foram incluídas crianças entre dois e seis anos agendadas para procedimentos cirúrgicos eletivos ambulatoriais. Os cuidadores das crianças completaram um formulário sócio-demográfico, o questionário EASI e uma Escala Análogo-Visual (EAV) de 100mm. Também, a Escala de Ansiedade Pré-Operatória de Yale (mYPAS) foi aplicada imediatamente antes da administração da medicação préanestésica, pelo anestesista pediátrico. Uma vez que o escore médio de compreensão das questões, medido pela EAV, foi superior a 80mm, considerou-se que o nível de compreensão da versão traduzida do EASI foi adequado. O questionário foi respondido por 110 cuidadores. Os resultados da análise de Rasch evidenciaram que as quatro subescalas da estrutura original apresentam desempenho inadequado (especialmente baixo índice de classificação correta dos aspectos do temperamento). A análise de componentes principais gerou uma solução com dois fatores. O Fator 1 é composto de itens de atividade e impulsividade, e o Fator 2 é predominantemente composto por itens das sub-escalas de sociabilidade e atividade. As sub-escalas originais do EASI não se correlacionaram com os escores da mYPAS, à exceção da sociabilidade (r=-0,449; P <0,001). O Fator 1 apresentou correlação positiva (r=0,239; P =0,0034) com a mYPAS, enquanto o Fator 2 apresentou correlação negativa (r=-0,404, P <0,0001). A análise de Rasch indicou que o instrumento original teve poder de discriminação insuficiente. Dentre os dois fatores propostos, os doze itens que compõem o Fator 1 apresentam perfil estatístico adequado, com alto poder de discriminação e relevância clínica. O Fator 2 não tem poder de discriminação adequado no seu formato atual, e ainda necessita de ajustes. / Some studies have shown correlation between temperament and preoperative anxiety levels in children. However, the nature of the association is not consensual. As most of these studies have used the EASI (emotionality, activity, sociability and impulsivity) questionnaire as an instrument to assess temperament, the present study examines the psychometric properties of this questionnaire in depth using the Rasch analysis to determine whether it is suitable for measuring children’s temperament. In addition, further analyses are carried out to explore potentially new factorial structures of the original instrument. The ability to discriminate between different levels of a child’s anxiety in the immediate preoperative period was also examined. It is a cross-sectional study, with a convenience sample of children aged between 2 and 6 years, to be submitted to outpatient elective surgeries. Children’s caregivers completed a socio-demographic form, the EASI questionnaire and a 0-100mm Analogue-Visual Scale (VAS). In addition, the Yale Preoperative Anxiety Scale (mYPAS) was performed immediately prior to the administration of pre-anesthetic medication by the pediatric anesthetist. As the mean comprehension level of the EASI-questions on VAS score was above 80mm, the translation of EASI was considered adequate. The questionnaire was answered by 110 caregivers. Results from Rasch analysis showed that the four subscales have inadequate perfomance (especially low person separation indices). Principal component analysis yielded a two-factor solution. Factor 1 is composed of activity and impulsivity items, and factor 2 is predominantly composed of items from the sociability and activity subscales. The original EASI subscales did not correlate with m-YPAS scores, with exception of sociability (r=-0.449; P <0.001). Factor 1 had a positive correlation (r=0.239; P=0.0034) with m-YPAS, while factor 2 showed a negative correlation (r=-0.404, P <0.0001). Rasch analysis indicated that the original EASI instrument has insufficient separation power. The twelve items that compose Factor 1 show adequate fit statistics, high separation power, and clinical meaning. Factor 2 is not sufficiently powerful in its current state, and still requires refinements.

Objetivo: Este estudo avalia a estrutura do IDATE (Inventário de Ansiedade Traço-Estado) utilizando a análise psicométrica de Rasch, propondo-se ao refinamento e redução do instrumento. Métodos: Consiste em estudo transversal em amostra de 900 pacientes em perioperatório de cirurgias eletivas, com idade variando de 18 a 60 anos e ASA I-III. Informações demográficas foram coletadas empregando-se um questionário estruturado. O instrumento de mensuração IDATE foi aplicado na tarde que antecedia a cirurgia, antes da sedação pré-operatória. Resultados: As escalas de estado e traço de ansiedade foram analisadas separadamente pelo modelo de Rasch. A análise demonstrou invariância no nível de ansiedade traço e estado no formato original de ambas as escalas, resultando em instabilidade no desempenho dos itens. Subsequentemente, a escala refinada foi retestada em duas amostras randomizadas de 300 sujeitos cada, sendo confirmados os resultados. O desempenho foi adequado, independentemente do gênero. Na análise, alguns itens da escala estado foram eliminados (itens 3,4,9,10,12,15 e 20) devido à inadequação ao modelo estatístico. Os itens restantes mostraram unidimensionalidade, independência local e adequado índice de consistência interna. Na escala original de traço, identificaram-se muitas falhas. Primeiro, a escala de resposta tipo Likert 4 pontos provou ser inadequada, e desordens threshold foram encontradas nos 20-itens. A escala original de traço mostrou insuficiente interação e muitos itens individuais mal ajustados. Seguindo a análise e retestando a segunda amostra randomizada, alguns itens foram excluídos (itens 4, 6, 11, 14, 15 e 19). A versão refinada obteve independência local, unidimensionalidade e adequados modelos estatísticos para a fase de sedação pré-operatória. Discussão: Os resultados indicam que a aplicação do modelo de Rasch conduz ao refinamento das escalas clássicas de IDATE estado e traço, sugerindo também que as versões sucintas apresentam desempenho psicométrico mais adequado e estão livres de desordens no limiar de resposta (threshold) e DIF (funcionamento diferencial de itens). / Objetive: This study evaluates the STAI structure using a Rasch psychometric approach and it proposes a refined and shorter STAI version. Methods: A cross-sectional study was performed with 900 inpatients scheduled for elective surgery. Age varied from 18 to 60 years (ASA physical status I-III). Demographic information was collected using a structured questionnaire. The measuring instrument (the State-Trait Anxiety Inventory) was applied to all patients in the afternoon before the surgery, before of the patients Results: Rasch analysis of the state and trait anxiety scales was performed separately. This analysis demonstrated that the original format of state and trait scales fails to show invariance across the trait-state anxiety level, which results in the unstable performance of items. The refined scale was retested in two subsequent random samples of 300 subjects each, and the results were confirmed. The performance was adequate regardless of gender. In the analysis, some items of the state scale (items 3,4,9,10,12,15 and 20) were deleted due to poor fit statistics. The remaining 13 items showed unidimensionality, local independence, and adequate index of internal consistency. Also, the original trait scale displayed several weaknesses. First, the 4 point Likert response scale proved to be inadequate, and threshold disorders were found in all twenty items. Also, the original trait scale showed insufficient item-trait interaction and several individual item misfits. Following the rescoring process, and retesting in a second random sample, items were excluded (namely items 4, 6, 11, 14, 15 and 19). The refined version showed local independence, unidimensionality and adequate fit statistics.receive preoperative sedatives. Discussion: The results indicate that the application of the Rasch Model led to the refinement of the classic state and trait STAI scales. In addition, suggest that these shorter versions present more adequate psychometric performance and are free of threshold disorders and differential item functioning problems.

Diversos estudos têm demonstrado associação entre o temperamento e os níveis de ansiedade pré-operatória em crianças. Entretanto, a natureza desta associação não é consenso. Como a maioria desses estudos utilizou o questionário EASI (emocionalidade, atividade, sociabilidade e impulsividade) como instrumento para mensurar o temperamento, o presente estudo examina em profundidade as propriedades psicométricas do mesmo utilizando a análise de Rasch, para determinar se o instrumento mede de forma adequada o temperamento de crianças. Além disso, outras análises são aplicadas para explorar novas estruturas fatoriais do instrumento original. A habilidade da nova versão do EASI para discriminar diferentes níveis de ansiedade no período préoperatório imediato em crianças também foi investigada. Trata-se de um estudo transversal, com amostra de conveniência, onde foram incluídas crianças entre dois e seis anos agendadas para procedimentos cirúrgicos eletivos ambulatoriais. Os cuidadores das crianças completaram um formulário sócio-demográfico, o questionário EASI e uma Escala Análogo-Visual (EAV) de 100mm. Também, a Escala de Ansiedade Pré-Operatória de Yale (mYPAS) foi aplicada imediatamente antes da administração da medicação préanestésica, pelo anestesista pediátrico. Uma vez que o escore médio de compreensão das questões, medido pela EAV, foi superior a 80mm, considerou-se que o nível de compreensão da versão traduzida do EASI foi adequado. O questionário foi respondido por 110 cuidadores. Os resultados da análise de Rasch evidenciaram que as quatro subescalas da estrutura original apresentam desempenho inadequado (especialmente baixo índice de classificação correta dos aspectos do temperamento). A análise de componentes principais gerou uma solução com dois fatores. O Fator 1 é composto de itens de atividade e impulsividade, e o Fator 2 é predominantemente composto por itens das sub-escalas de sociabilidade e atividade. As sub-escalas originais do EASI não se correlacionaram com os escores da mYPAS, à exceção da sociabilidade (r=-0,449; P <0,001). O Fator 1 apresentou correlação positiva (r=0,239; P =0,0034) com a mYPAS, enquanto o Fator 2 apresentou correlação negativa (r=-0,404, P <0,0001). A análise de Rasch indicou que o instrumento original teve poder de discriminação insuficiente. Dentre os dois fatores propostos, os doze itens que compõem o Fator 1 apresentam perfil estatístico adequado, com alto poder de discriminação e relevância clínica. O Fator 2 não tem poder de discriminação adequado no seu formato atual, e ainda necessita de ajustes. / Some studies have shown correlation between temperament and preoperative anxiety levels in children. However, the nature of the association is not consensual. As most of these studies have used the EASI (emotionality, activity, sociability and impulsivity) questionnaire as an instrument to assess temperament, the present study examines the psychometric properties of this questionnaire in depth using the Rasch analysis to determine whether it is suitable for measuring children’s temperament. In addition, further analyses are carried out to explore potentially new factorial structures of the original instrument. The ability to discriminate between different levels of a child’s anxiety in the immediate preoperative period was also examined. It is a cross-sectional study, with a convenience sample of children aged between 2 and 6 years, to be submitted to outpatient elective surgeries. Children’s caregivers completed a socio-demographic form, the EASI questionnaire and a 0-100mm Analogue-Visual Scale (VAS). In addition, the Yale Preoperative Anxiety Scale (mYPAS) was performed immediately prior to the administration of pre-anesthetic medication by the pediatric anesthetist. As the mean comprehension level of the EASI-questions on VAS score was above 80mm, the translation of EASI was considered adequate. The questionnaire was answered by 110 caregivers. Results from Rasch analysis showed that the four subscales have inadequate perfomance (especially low person separation indices). Principal component analysis yielded a two-factor solution. Factor 1 is composed of activity and impulsivity items, and factor 2 is predominantly composed of items from the sociability and activity subscales. The original EASI subscales did not correlate with m-YPAS scores, with exception of sociability (r=-0.449; P <0.001). Factor 1 had a positive correlation (r=0.239; P=0.0034) with m-YPAS, while factor 2 showed a negative correlation (r=-0.404, P <0.0001). Rasch analysis indicated that the original EASI instrument has insufficient separation power. The twelve items that compose Factor 1 show adequate fit statistics, high separation power, and clinical meaning. Factor 2 is not sufficiently powerful in its current state, and still requires refinements.

Objetivo: Este estudo avalia a estrutura do IDATE (Inventário de Ansiedade Traço-Estado) utilizando a análise psicométrica de Rasch, propondo-se ao refinamento e redução do instrumento. Métodos: Consiste em estudo transversal em amostra de 900 pacientes em perioperatório de cirurgias eletivas, com idade variando de 18 a 60 anos e ASA I-III. Informações demográficas foram coletadas empregando-se um questionário estruturado. O instrumento de mensuração IDATE foi aplicado na tarde que antecedia a cirurgia, antes da sedação pré-operatória. Resultados: As escalas de estado e traço de ansiedade foram analisadas separadamente pelo modelo de Rasch. A análise demonstrou invariância no nível de ansiedade traço e estado no formato original de ambas as escalas, resultando em instabilidade no desempenho dos itens. Subsequentemente, a escala refinada foi retestada em duas amostras randomizadas de 300 sujeitos cada, sendo confirmados os resultados. O desempenho foi adequado, independentemente do gênero. Na análise, alguns itens da escala estado foram eliminados (itens 3,4,9,10,12,15 e 20) devido à inadequação ao modelo estatístico. Os itens restantes mostraram unidimensionalidade, independência local e adequado índice de consistência interna. Na escala original de traço, identificaram-se muitas falhas. Primeiro, a escala de resposta tipo Likert 4 pontos provou ser inadequada, e desordens threshold foram encontradas nos 20-itens. A escala original de traço mostrou insuficiente interação e muitos itens individuais mal ajustados. Seguindo a análise e retestando a segunda amostra randomizada, alguns itens foram excluídos (itens 4, 6, 11, 14, 15 e 19). A versão refinada obteve independência local, unidimensionalidade e adequados modelos estatísticos para a fase de sedação pré-operatória. Discussão: Os resultados indicam que a aplicação do modelo de Rasch conduz ao refinamento das escalas clássicas de IDATE estado e traço, sugerindo também que as versões sucintas apresentam desempenho psicométrico mais adequado e estão livres de desordens no limiar de resposta (threshold) e DIF (funcionamento diferencial de itens). / Objetive: This study evaluates the STAI structure using a Rasch psychometric approach and it proposes a refined and shorter STAI version. Methods: A cross-sectional study was performed with 900 inpatients scheduled for elective surgery. Age varied from 18 to 60 years (ASA physical status I-III). Demographic information was collected using a structured questionnaire. The measuring instrument (the State-Trait Anxiety Inventory) was applied to all patients in the afternoon before the surgery, before of the patients Results: Rasch analysis of the state and trait anxiety scales was performed separately. This analysis demonstrated that the original format of state and trait scales fails to show invariance across the trait-state anxiety level, which results in the unstable performance of items. The refined scale was retested in two subsequent random samples of 300 subjects each, and the results were confirmed. The performance was adequate regardless of gender. In the analysis, some items of the state scale (items 3,4,9,10,12,15 and 20) were deleted due to poor fit statistics. The remaining 13 items showed unidimensionality, local independence, and adequate index of internal consistency. Also, the original trait scale displayed several weaknesses. First, the 4 point Likert response scale proved to be inadequate, and threshold disorders were found in all twenty items. Also, the original trait scale showed insufficient item-trait interaction and several individual item misfits. Following the rescoring process, and retesting in a second random sample, items were excluded (namely items 4, 6, 11, 14, 15 and 19). The refined version showed local independence, unidimensionality and adequate fit statistics.receive preoperative sedatives. Discussion: The results indicate that the application of the Rasch Model led to the refinement of the classic state and trait STAI scales. In addition, suggest that these shorter versions present more adequate psychometric performance and are free of threshold disorders and differential item functioning problems.

Objetivo: Este estudo avalia a estrutura do IDATE (Inventário de Ansiedade Traço-Estado) utilizando a análise psicométrica de Rasch, propondo-se ao refinamento e redução do instrumento. Métodos: Consiste em estudo transversal em amostra de 900 pacientes em perioperatório de cirurgias eletivas, com idade variando de 18 a 60 anos e ASA I-III. Informações demográficas foram coletadas empregando-se um questionário estruturado. O instrumento de mensuração IDATE foi aplicado na tarde que antecedia a cirurgia, antes da sedação pré-operatória. Resultados: As escalas de estado e traço de ansiedade foram analisadas separadamente pelo modelo de Rasch. A análise demonstrou invariância no nível de ansiedade traço e estado no formato original de ambas as escalas, resultando em instabilidade no desempenho dos itens. Subsequentemente, a escala refinada foi retestada em duas amostras randomizadas de 300 sujeitos cada, sendo confirmados os resultados. O desempenho foi adequado, independentemente do gênero. Na análise, alguns itens da escala estado foram eliminados (itens 3,4,9,10,12,15 e 20) devido à inadequação ao modelo estatístico. Os itens restantes mostraram unidimensionalidade, independência local e adequado índice de consistência interna. Na escala original de traço, identificaram-se muitas falhas. Primeiro, a escala de resposta tipo Likert 4 pontos provou ser inadequada, e desordens threshold foram encontradas nos 20-itens. A escala original de traço mostrou insuficiente interação e muitos itens individuais mal ajustados. Seguindo a análise e retestando a segunda amostra randomizada, alguns itens foram excluídos (itens 4, 6, 11, 14, 15 e 19). A versão refinada obteve independência local, unidimensionalidade e adequados modelos estatísticos para a fase de sedação pré-operatória. Discussão: Os resultados indicam que a aplicação do modelo de Rasch conduz ao refinamento das escalas clássicas de IDATE estado e traço, sugerindo também que as versões sucintas apresentam desempenho psicométrico mais adequado e estão livres de desordens no limiar de resposta (threshold) e DIF (funcionamento diferencial de itens). / Objetive: This study evaluates the STAI structure using a Rasch psychometric approach and it proposes a refined and shorter STAI version. Methods: A cross-sectional study was performed with 900 inpatients scheduled for elective surgery. Age varied from 18 to 60 years (ASA physical status I-III). Demographic information was collected using a structured questionnaire. The measuring instrument (the State-Trait Anxiety Inventory) was applied to all patients in the afternoon before the surgery, before of the patients Results: Rasch analysis of the state and trait anxiety scales was performed separately. This analysis demonstrated that the original format of state and trait scales fails to show invariance across the trait-state anxiety level, which results in the unstable performance of items. The refined scale was retested in two subsequent random samples of 300 subjects each, and the results were confirmed. The performance was adequate regardless of gender. In the analysis, some items of the state scale (items 3,4,9,10,12,15 and 20) were deleted due to poor fit statistics. The remaining 13 items showed unidimensionality, local independence, and adequate index of internal consistency. Also, the original trait scale displayed several weaknesses. First, the 4 point Likert response scale proved to be inadequate, and threshold disorders were found in all twenty items. Also, the original trait scale showed insufficient item-trait interaction and several individual item misfits. Following the rescoring process, and retesting in a second random sample, items were excluded (namely items 4, 6, 11, 14, 15 and 19). The refined version showed local independence, unidimensionality and adequate fit statistics.receive preoperative sedatives. Discussion: The results indicate that the application of the Rasch Model led to the refinement of the classic state and trait STAI scales. In addition, suggest that these shorter versions present more adequate psychometric performance and are free of threshold disorders and differential item functioning problems.

Objective. To evaluate the validity and reliability evidence of the preceptor assessment of student tool (PAST) which was designed to assess doctor of pharmacy (PharmD) student rotation performance. Methods. Evaluation data were loaded into WINSTEPS software to conduct a Rasch rating scale analysis. Validity evidence was examined from construct and content validity perspectives, and reliability was assessed via student and item separation index and reliability coefficient. Data from 435 observations were included in the analysis. Results. All 19 items measured the same construct of interest and the five-point rating scale functioned appropriately and differentiated students' ability. However, the item/person map indicated an absence of items at the end of the measurement continuum. Conclusion. Although adding items at the end of the measurement continuum may be beneficial, PAST showed good validity and reliability evidence when used to evaluate PharmD student rotations and is suitable to assess mastery learning.

Adolescent students’ perceptions of parental influence in relation to mathematics education is an under-researched area, since most studies in this area focus on parental involvement in primary mathematics. This research study aims to fill this gap in the literature by exploring adolescent students’ perceptions of parental influence on their dispositions towards studying mathematically-demanding courses in Higher Education (HE). This study employs mixed research methods to investigate students’ perceptions of parental influence with a survey (N=563) and semi-structured interviews with six students and their parents. Additional interviews were conducted with three immigrant families in Cyprus. The study builds on Bourdieu’s theory of reproduction of social inequalities and extends some concepts of Bourdieu’s theoretical framework to discuss parental influence. The main findings of this research study are reported in three journal papers whilst the papers are under review for publication. The first paper of the thesis reports the analysis of the quantitative data of this study by combining Rasch analysis with statistical modelling. The statistical analysis showed that parental influence is not statistically significant for predicting students’ mathematical dispositions in some models, when other background variables are included in the models. However, further statistical analysis showed that the effect of parental influence is mediated through students’ choice of mathematics course and their mathematical inclination. The non-statistical significance of parental influence in some models was interpreted as a ‘misrecognition’ after Bourdieu (1980). The same phenomenon was noted in the second paper of the thesis, where students and their parents ‘denied’ parental influence during their interviews. This was again interpreted as ‘misrecognition’ and parental influence is conceptualised as a form of ‘symbolic violence’ that parents exercise on their children. Arguably, parents possess more capital in the family field and their influence on their children’s educational choices might be unconscious, thus students’ misrecognise their parents’ influence but they draw significantly on their family’s capital to make their choices for future studies in HE.Lastly, the third paper of the thesis explored cultural differences in students’ perceptions of parental influence in England and Cyprus; stronger perceptions of parental influence were identified in immigrant students’ interviews compared to indigenous students in both countries. Bourdieu’s (1977) concept of ‘hysteresis’ was adapted to theorise this phenomenon. Arguably, while immigrant students’ habitus adjust to the new field, students become more reflective on their parents’ influence because of the reflexivity the hysteresis effect entails. In all three papers Bourdieu’s theoretical framework was used to operationalise students’ mathematical dispositions and to interpret the findings of this study. The main contributions to knowledge of this study is the operationalisation of students’ mathematical habitus; the new theoretical conceptualisation of parental influence as a form of symbolic violence in the family field and the extension of the hysteresis effect to interpret immigrant students’ stronger perceptions of their parents’ influence.

The current investigation is post-hoc in nature and is nested in a larger research project, which aimed to explore and compare the personality characteristics, coping mechanisms and psychological well-being of South African and Swedish police trainees. The purpose of this particular study was to explore the psychometric properties of the Cloninger’s Temperament and Character Inventory (TCI) in a sample of South African Police trainees. A literature review highlighted that the TCI has the potential to be applied across various cultures without the risk of any ethnic or gender bias; this characteristic is attributed to the theoretical model underlying the TCI, which assumes that personality consists of seven universal factors, which manifest in an invariant manner across all humans. Despite this, the majority of international research focussing on the TCI version nine has been undertaken with primarily European populations and Eastern populations. Research exploring the construct equivalence, factor structure and the level of instrument bias of the TCI in any African is virtually non-existent. The current study endeavoured to address the aforementioned knowledge gap by exploring the psychometric properties of the TCI in a multi-cultural South African sample. The primary goal of the research endeavour was to explore whether the TCI can in the future be established as a valid and reliable personality assessment measure in a multi-cultural context like South Africa. Literature indicates that in the current South African psychometric context personality measures should adhere to the stipulations of the employment equity act (EEA), which especially in its amended form requires fair and just measurement. Studies such as this one can be used to adjudicate whether the TCI has the potential to be used as a fair and reliable measure, which does not violate the stipulation of the employment equity act. In this way the measure may contribute to provide evidence which can be used to make fair, just and reliable decisions not only in the South African Police Service, but also within the general public. A quantitative investigation was conducted using analysis base on Item Response Theory, specifically the Rasch model, which is considered more accurate than Classical Test Theory in assessing the psychometric functioning of dichotomous personality assessment measures. The analyses rendered information with which the researcher was able to evaluate the validity, reliability, levels of gender and cultural bias, as well as the factors rendered by the TCI. The research sample was a convenient one, comprising 1144 police trainees whom completed a test-battery of four tests, which included the TCI. The results derived from this investigation show that the primary TCI scales each measured a single factor, the presence of these factors among the current sample provide some support for the universality of the TCI; however most of these scales showed a high level of bias when measuring their respective constructs across ethnic and gender groups. The results also pointed out that numerous items and sub-scales possess a considerable level of ethnic and gender bias. There was also no attempt made to investigate the reasons underlying bias, bias may yield important information about cross-cultural differences and can also be seen as a phenomenon that requires explanation (Poortinga and Van der Flier, 1989), which means that the study created a launch pad for future investigations to explore the sources of bias. These findings have stern implications for the larger research project, as it might decrease the validity of findings derived from comparing scores across groups within the current sample, and to a lesser degree if the performance of the current sample is compared to that of the Swedish sample. It can also be argued that another implication of the study’s findings is that the information derived from the TCI cannot legally be used to make clinical or selection decisions based partially on the personality profile of individuals; however the convenient nature of the sample limits the generalizability of the investigation’s findings. This means that additional research is first required before the legitimacy of the use of the TCI in a South African context can be evaluated. / Dissertation (MA)--University of Pretoria, 2014. / tm2015 / Psychology / MA / Unrestricted

No description available.