Includes bibliographical references. / International portfolio diversification is an area of popular academic interest. Most of the research is based on linkages between developed markets or between developed markets and emerging markets. Emerging markets have only been a feature of research for periods of crisis and contagion such as the emerging market crisis of 1997 and 1998. The common argument used to explain the cause of the stock market linkages is based on economic slowdowns (or deep-recession), large balance of payment problems or other macro-economic problems in the emerging markets. Contagion or herding behaviour is an alternative explanation to the cause of the co-movement of emerging markets. Contagion or Herding behaviour is based on the assumption that it is too costly to acquire information. Investors thus remain uninformed in the countries in which they invest. Investors try to infer future movements in one market based on how the rest of the market is reacting. The uniformed investors follow the supposedly informed investor. These information frictions can make investors follow the market, rather than take the time and expense to make their own assessments about the market fundamentals. This research aims to identify countries that are regarded as similar by international investors. Countries that are regarded as similar can be grouped together to form separate clusters. This research attempts to ascertain as whether there has been any change of perceptions since the emerging market crisis of 1997 and 1998, hence any change in the composition of above-mentioned clusters.

Interviewers are a principal source of error in quantitative surveys. While surveys are often self-administered (e.g. in online surveys), it is often required to administer these face-to- face. This is the case, for example, in census surveys in low-income areas where there is little internet penetration, like that of the quality-of-life surveys presently being conducted in multiple countries through a residential door-to-door approach (Carr et al., 2018). In such situations, the social interaction between the interviewer conducting the survey and the interviewee is likely to introduce bias into the survey data collected. Interviewer effects (IE) can influence both item non-response and answer quality, i.e., participants not providing the true answer (Harling, et al., 2019). In an attempt to gain more representative data, this study conducted an exploratory analysis on the possible antecedents and consequences of interviewer effects using the Living Wage survey presently being conducted in South Africa, as the study context. To this end, I examine the systematic biasing effects associated with deploying the same group interviewers (n = 10), of the same ethnicity, age, and of equal gender distribution across five sampling areas in Cape Town in a quasi-experimental design (n = 282). This study highlighted that each interviewer is associated with a unique set of systematic bias that varies dependent on the survey item type. Sensitive items requiring respondents to disclose personal information were the most prone to bias, followed by interviewer-referencing and attitudinal items sequentially. Furthermore, this study found that gender differences in the interview had a marginal influence on the attitudes respondents are willing to share. I hope to contribute to an understanding and critical consideration of the antecedents and consequences of deploying human interviewers for collecting quantitative surveys, especially in a context where ethnic, gender and political differences are loaded in social interactions and are likely to contribute to respondents obscuring their responses.

Includes abstract. / Includes bibliographical references (p. 248-287). / Winning the commitment and support of employees for organisational transformation is a major objective of the leaders of organisational change. However, the determinants and outcomes of employee commitment to organisational change (Herscovitch & Meyer, 2002) are still not yet fully understood, especially with regard to a mandatory information system change in a non-Western environment.

This research project examines a theoretical gap (impact of culture on entrepreneurial orientation EO) to try and mitigate a practical problem (unemployment) among graduates in three East African countries: Kenya, Tanzania and Uganda. Unemployment would be mitigated if only graduates embraced entrepreneurship by starting their own business projects, and many governments have encouraged their youth/graduates to do this. Unfortunately, graduate student interest in entrepreneurial activities in many countries in Africa is very low, and the three study countries are no exception to this trend. While many explanations have been advanced for this phenomenon, a key theoretical gap left unattended in entrepreneurship research is the way in which culture impacts upon EO and how this affects entrepreneurial intention (EI), yet EO is often considered the most important variable in the formative stage of a given project. A close look at how culture influences EO is important because EO is unexplainable without considering the socio-cultural framework in which it is embedded, since it can be supportive or a hindrance to entrepreneurship in different contexts. Thus the key question which this study tries to address is: „Does culture constrain the development of a strong EO, eventually leading to low start-up in this region?‟ Specifically, can the low graduate start-up in these countries be explained by the impact of five cultural orientation dimensions (ambiguity intolerance, power distance, masculinity, independence and interdependence) and ability perception variables (achievement motivation and learning goal orientation LGO) on two variables of EO namely risk taking and proactiveness? How does gender affect these relationships? Besides gender, this study also sought to know the level of prevalence in the study population of three other important culturally influenced variables in entrepreneurship literature namely experience, fear of failure and modernity in order to throw more light on the study problem In particular, given that individuals with a modern outlook are somehow liberated from firm cultural norms, this study sought to establish whether students with a more modern outlook differ from those with a less modern one in terms of the study variables. Further, does optimism another important culturally inclined characteristic of entrepreneurs moderate the relationship between EO and entrepreneurial intention? Some authors argue, however, that culture does not matter; rather, what matters are the entrepreneurial competencies of an entrepreneur. Hence another major question addressed in the current study is to what extent do entrepreneurial competencies (such as knowledge/networks) moderate the relationship between EO and entrepreneurial intention? Using the theory of planned behavior (TPB), upper echelons theory and image theory, this study seeks to address these questions based on a pragmatic paradigm and thus a mixed methods approach in two phases. Phase one of the study was qualitative consisting of non- structured interviews and conversations with various stake holders and is the basis upon which the study instrument was refined. Phase two was quantitative, utilizing a cross-sectional survey research design based on a non random sampling to gather data from finalists in business faculties in three public and two private universities in the study countries (N=1086) during their classes. Data analysis consisted of three phases, comprising ten steps. Phase one was more of a preliminary analysis and consisted of five steps: Generation of descriptive statistics such as frequencies, percentages, and normality tests in step 1, T-tests to gauge the prevalence of experience, fear of failure, and modernity as well as a MANOVA to gauge the prevalence of the cultural dimensions in each study country in step 2, Exploratory factor analysis (EFA) by Varimax/promax rotation to examine the factor structures of the study dimensions in step 3, followed by examination of validity (construct validity, discriminant validity) and reliability for all study instruments (alpha and composite reliability, CR) in Step 4, while step 5 confirmed the factor structure of the measures using confirmatory factor analysis CFA (Lisrel 8.8). The second phase utilized structural equation modeling (SEM) based on latent variables (using AMOS 23) to first estimate a CFA model, followed by a structural baseline model for all data combined (omnibus model) in step 6. This was followed by fitting the baseline model into each country data set in step 7. In step 8, data was divided by gender into male and female samples and by modernity into low and high modernity groups and the baseline model was fitted into each of these four data sets. This was followed by invariance tests between the gender sets and modernity sets as a basis for their meaningful comparison (step 9). The third phase utilized the process macro in SPSS (step 10) to conduct the moderation analysis. Study findings indicate that in all three countries, only 50% of the respondents had some sort of start-up experience. A third (31%) of the students in the three countries indicated that fear of failure would prevent them from starting a business, while the rest indicated that it would not. Further, students who do not regard fear of failure as a barrier to entrepreneurial activities scored significantly higher on proactiveness, knowledge, achievement motivation and modernity in all the three countries, while in at least two of these countries, these people scored significantly higher on risk taking, networking and learning goal orientation. This finding confirms that fear of failure is an important barrier to graduate entrepreneurship in this region. Turning to the structural models in SEM, findings indicate that the low start-up rate in these countries can be attributed to the negative impact of ambiguity intolerance (the most problematic variable), power distance, and lack of an optimistic bias as well as possible negative attitude towards those with an independent cultural orientation. However, Independence and Interdependence support EO, in agreement with researchers who assert that both cultural variables are good for entrepreneurship. Theoretically, the study makes an extension of the TPB since achievement motivation predicts intention in all study samples (apart from Kenya and Tanzania). In terms of gender, there are no significant differences on the reported levels of risk taking; however females score significantly higher on proactiveness. Further structural models indicate that males are more achievement oriented than females, while the low modernity group seems to be more entrepreneurial than the high modernity group. Lastly, networks and knowledge moderate the relationship between risk taking/proactiveness and intention, while optimism does not. The study calls for a revision of the curriculum to include tolerance for ambiguity, proactiveness and autonomy courses in entrepreneurship education as well as a change in the mode of delivery of this subject. A transformation in the education systems of the three countries is needed to produce critical thinkers and to introduce entrepreneurship early in the education system to make everyone appreciate entrepreneurship, thus nurturing an entrepreneurial culture.

Increased sales of counterfeit luxury branded handbags are a significant transnational concern. Research effort concerning understanding this purchase behaviour has largely been contextualised outside of Africa, with varying results relating to the impact of attitude on the purchase decision-making process having emerged. With South Africa serving as one of the most profitable counterfeit luxury handbag markets on the continent, this study sought to fundamentally determine whether consumers' attitude effected their buying rationale. The study was conducted in Durban, home to a large female middle-class population, deemed to satisfy many of the characteristics attributed to the typical counterfeit handbag consumer. The study adopted a post-positivist research paradigm and employed a causal research design. In terms of research strategy, the study made use of a quantitative approach. The target population regarding the investigation pertained to Durban's emergent Black middle-class females. In order to segment this population, nonprobability sampling was utilised, with convenience and snowball sampling having been selected. The total sample size amounted to 350 individuals, who provided data by means of a self-administered questionnaire. Once collected, this data was interpreted through factor and regression analyses. The findings of the study confirmed that individual, product and service factors impacted upon the attitude to purchase of counterfeit luxury handbags among Durban's emergent Black Middle-class females. Furthermore, it was discovered that the attitudinal antecedent's of knowledge, perceived risk, ethical obligation, product price and service quality significantly influenced attitude formation, and in turn, attitude to purchase. As such, the information collated allows for those combating counterfeit sales activity to focus their efforts and resources in a more predetermined manner. It is therefore primarily recommended that anti-counterfeiting agents place greater emphasis upon defending intellectual property rights through demand, or ‘consumer focused' initiatives which target specific purchase-prompting variables. This investigation also provides interesting opportunities for future research, including determining whether attitude continues to play an important role across different South African provinces, as well as within international settings.

According to Dimson (1998), modem financial theory is founded on the assumption that markets are highly efficient. The presence of anomalous stock market behaviour has therefore attracted a great amount of research internationally. This thesis investigates the presence and exploitability of style anomalies on the London Stock Exchange (LSE) and is divided into three main branches of research.

Word processed copy. / Includes bibliographical references (p. 168-180). / The difference in career success among male and female academics is welldocumented and a number of qualitative studies have offered explanations about the challenges faced by women academics. This study provided an empirical investigation into the relationship between selected career psychology variables and the career success of women academics in South Africa. This research employed organisational theory to explain career success. The impact of work centrality, motivation, career anchors and self-efficacy on career success was examined. Care-giving responsibility was included as a moderating variable on work centrality. The examination of career theory and the testing of these particular career variables in relation to the career success of academic women have not been conducted before in South Africa. Other studies in the field have typically been qualitative in nature or have focused on explanations why women are often unsuccessful in academia. Those few studies that address success amongst academic women have been conducted outside of South Africa and thus offer findings from a different context. This study builds on this previous body of knowledge by examining the constructs empirically.

Includes bibliographical references. / Most of today's knowledge of Business-to-Business marketing is grounded on studies conducted in so-called western countries. Recently some researchers have started to question the validity of concepts, ideas and measures conceived in developed markets to explain business marketing phenomena in non-western contexts. Moreover research has attributed this gap in the literature on the relative absence of Business-to-Business marketing research from emerging, non-western countries (see, for example, Biggemann and Fam, 2011.) In emerging markets firms also operate in large networks that contains multiple, complex, direct and indirect business relationships among buyers and sellers. The thesis draws on four different empirical studies to integrate key sets of knowledge spawned from the realms of the Industrial Marketing and Purchasing group (IMP) to observe Business-to-Business relationships in a South African context.

Includes abstract. / Includes bibliographical references (leaves 186-231). / Infection with the human immunodeficiency virus (HIV) has reached pandemic proportions. In 2007, UNAIDS (2007) reported that 33.2 million people were living with HIV, 2.5 million were newly infected, and 2.1 million people died due to HIV and its associated disease, Acquired Immunodeficiency Syndrome (AIDS). Approximately two-thirds of these people reside in sub-Saharan Africa, a region where economic, infrastructural, and human resources are severely constrained. Infection rates in the range of 10%-35% make substantial demands on the health infrastructure of most sub-Saharan African countries and limit the amount of care provided to people infected and affected by HIV/AIDS.

Includes bibliographical references. / Little is known about the antecedents of work-family conflict (WFC) among Hindu working women in South Africa, a minority subgroup shaped by a unique set of historical, political, and cultural dynamics. Responding to repeated calls in the literature for the examination of work-family issues in unique cultural contexts, this study began with 20 in-depth interviews to elicit the subjective meaning that Hindu women in South Africa give to their work-family experiences. The qualitative data were analysed adhering to the principles of thematic analysis. These findings, together with a review of extant literature, were used to develop a new and culturally nuanced explanatory model of the antecedents of WFC for this specific context. The antecedents of WFC in the explanatory model include role stressors, sources of social support, and specific individual-level cultural variables. The study’s propositions were tested with survey data from 317 respondents. Psychometric analyses confirmed the portability of the measures and the bi-directional nature of WFC; work-to-family conflict and family- to-work conflict. Multiple regression analyses showed that a significant amount of variability in work-to-family conflict and family-to work conflict were explained by within- domain and cross-domain role stressors; with work overload having the strongest predictive effect on both directions of WFC. The results further highlight the salience of family in Hindu culture, noting that family involvement functioned as an important resource in reducing both directions of WFC and that food-work overload had a distinctive effect on WFC as a significant within-domain and cross-domain stressor. Results of further moderated multiple regression analyses confirmed co-worker support as an important resource for alleviating work-to-family conflict and for buffering the negative effects that work stressors can have on work-to-family conflict. Likewise, the results confirmed spousal support as an important resource for reducing family- to-work conflict; however, paid domestic support increased family- to-work conflict directly and when interacting with food-work overload. Moderated multiple regression analyses additionally showed that work involvement interacted significantly with gender role ideology in predicting work-to-family conflict and that family hierarchy orientation interacted significantly with family involvement in predicting family- to-work conflict. Overall, the results of this study strengthen the argument for the importance of uncovering and examining culturally salient variables in work-family research.