A research report submitted to the Faculty of Commerce, Law and Management, University of the Witwatersrand, in partial fulfilment of the requirements for the degree of Master of Management in the field of Strategic Marketing, Johannesburg 2019 / There is increased usage of self-service technology in efforts to deliver superior service quality in business schools. This transition from face-to-face interaction has led to both satisfaction and dissatisfaction with service quality delivered. This study sought to ascertain the impact self-service technologies have on the perception of service quality (Servqual) dimensions of reliability, tangibility, assurance, responsiveness and empathy. A survey research instrument based on an adapted Servqual questionnaire that was self-administered was employed. Convenience sampling was utilised on the Wits Business School campus, with 210 students’ views captured. Then, the data was analysed using GAP analysis and descriptive statistics extrapolated from the data. The results of the study indicate that a transition to self-service portals does not lead to customers having a positive view of all the service quality dimensions as most customers were neutral in their perceptions. Students, therefore, need to have both self-service technologies and walk-in centres for face-to-face interaction. The study contributes to the ever-increasing knowledge on the usage of self-service technology in higher education in emerging markets. While there are significant shifts towards usage of the portals in business schools, there is a need to control the pace of change and avoid drastic adoption without other options as it may not lead to positive customer perceptions. The study recommends that transition not be drastic, that focus be on user-friendliness of the portals, that there be increased use of interactive technologies such as chatbots to increase responsiveness as well as a provision of people to engage with when there is need for escalation of issues aiding service recovery. / XL2019

Problem gambling is a growing social concern that results in debt, lost jobs, broken families, and at times, suicide. Slot machines are the most popular and most addictive form of gambling, generating nearly 70% of annual profits for the commercial casino industry in North America. Behavioral researchers have identified variables that establish and maintain problem gambling on slot machines, and the data reveal characteristics that influence preference for specific games, and subsequently, time spent engaged with specific machines. A degree of variability has been reported in the outcomes observed across participants and studies; such variability may be influenced by generic features of the games used by researchers utilizing a “one machine suits all” approach to slot research, i.e., within most studies a single machine is used for all participants rather than chosen according to participant preference for the machine or features therein. The following set of studies aimed to investigate variables related to slot machine selection, gambler preference for structural characteristics, and the reinforcing effectiveness of the machine on gambling behavior. The first of three experiments evaluated participant preference for functional and structural characteristics of popular three-, four- and five-reel electronic Vegas-style slot machines according to (a) a Likert rating scale of attitudes toward the machines, (b) forced ranking of most-preferred machines, and (c) a paired-stimulus preference assessment. Experiment II utilized a series of multiple schedules randomized across participants to examine differences in the rate of play when participants gambled on their most- and least-preferred machines. In Experiment III, the reinforcing effectiveness of the machines was evaluated. Participants engaged in an arbitrary task to gain access to their highest- and least-preferred machines as indicated in Experiment I. Response effort during the work task was evaluated under a progressive ratio schedule of reinforcement to determine the degree to which the individual would work to gain access to the machine. Participant break points were compared across the two conditions. Results of Experiment I yielded significant differences in the degree to which participants rated the machines: three-reel machines and those displaying a win were rated significantly higher in the attitude assessment when compared to four- or five-reel machines, and when compared to those with a loss, respectively. During the forced ranking procedure, participants again selected three-reel machines significantly more frequently in their “top five” favorite machines than four- or five-reel options. Last, when subjected to the paired stimulus preference assessment, participants selected three-reel machines on a significantly higher percentage of trials than five-reel alternatives. Results of Experiment II demonstrated that rate of play on high-preference machines is on average, faster than rate of play on lower-preference machines, and results from Experiment III yielded on average, higher break points in participant responding when offered the opportunity to gamble on high-preference slot machines following a response requirement than when offered a low-preference option; however, the average rate of response on the work task that led to machine access was lower during high-preference conditions than low-preference conditions. Implications of these findings for the gambling literature and for the effective prevention and treatment of disordered slot machine gambling are discussed.

This dissertation explores an unintended consequence of using personalized recommendations, that is, recommendations that are targeted to an individual consumer (e.g., personalized music playlists). I conceptualize that using personalized recommender systems can impede consumers’ learning of their own preferences and tastes from product experiences. Therefore, using these systems can decrease preference clarity, which is defined as certainty about individuals’ own preferences. For example, people may feel less certain about their own music preferences after listening to auto-generated personalized playlists. This reduced preference clarity, in turn, reduces consumer willingness to generate word-of-mouth (WOM) about their consumption experiences, such as their intent to talk about music they listened to with others, or to post social media content on their favorite musicians. Eight studies, using correlational and experimental designs and conducted with consumers who actively use personalization services (in the fashion and music domains), support this theorization. I end with a discussion of the potential theoretical extensions of this novel finding, as well as its practical implications.

The differential effects of clothing interest and knowledge on the utilization of price and intrinsic attributes on product evaluation and behavioral intentions were investigated. The impact of clothing interest and knowledge, price, and intrinsic attributes on attitude toward advertisements and perceptions of relevance, typicality, and fashionability were examined. Causal relationships among constructs were verified through path analysis. A conceptual model was developed and tested. To test the proposed model and additional propositions, a survey was conducted with manipulated variables. Data were collected in Clothing and Textiles departments of three universities and Communications and Marketing departments in one university. Respondents evaluated a pair of dress slacks inclusive of manipulated price and intrinsic attributes. A total of 417 questionnaires were subjected to statistical analysis. The items used to measure the theoretical constructs were thoroughly tested to assure the reliability of the measures and manipulation checks were conducted. The data were analyzed by multiple regression to test the significance of the proposed relationships. To test the differences between the high and low clothing interest and knowledge conditions, subjects were divided into two groups based on median scores on clothing interest and clothing knowledge scales. Then the data were subjected to analysis of variance and Duncan's Multiple Range Test. The results indicate that taken together, price, intrinsic attributes, clothing interest and knowledge significantly help to predict perceptions of quality, monetary sacrifice, perceptions of benefits, perceptions of value, willingness to buy, search intention, and attitude toward advertisements. Intrinsic attributes were significant in predicting the perceptions of quality, benefits, monetary sacrifice, value, and subjects' willingness to buy, search intention, attitude toward advertisements, and perceptions of typicality. Clothing interest was significant in predicting perceptions of monetary sacrifice, perceptions of benefits, perceptions of value, and willingness to buy. The proposed effect of clothing knowledge was significant in predicting subjects' perceptions of quality and intentions to search. The model specifying the causal relationships among the proposed variables was analyzed by Path Analysis. The results indicated that the relationships were significant and thus the proposed model was supported. / Ph. D.

Background: Youth consume high volumes of fast food, putting them at risk for poor diet, weight gain and several noncommunicable diseases. Fast food marketing can affect youth's food-related behaviours and has been identified as a determinant of excess weight and obesity. The aim of this research is to examine the relationship between exposure to fast food marketing and the fast food brand preferences and intake amongst youth aged 10-17 across six countries. Methods: The International Food Policy Study youth survey collects data on dietary patterns and behaviours amongst youth aged 10-17 living in six countries. Data on fast food marketing exposure, brand preferences and intake were compared using regression models adjusted for age, sex, income adequacy and ethnicity. Results: Exposure to fast food marketing was positively associated with both brand preferences and fast food intake across most countries. Brand preferences were consistently high across all countries when exposed to brand-specific marketing. Fast food intake was higher amongst ethnic minority respondents than ethnic majority respondents and amongst males compared to females. Conclusion: Fast food marketing is consistently associated with brand preferences and intake across all countries investigated, which points to the need for stringent government regulation to reduce unhealthy food marketing to youth in all 6 countries.

This thesis examines the impact of the COVID-19 pandemic on individual gambling behavior by testing whether ticket sales and revenues in the multi-state parimutuel PowerBall and MegaMillions lotteries decreased or increased between the pre-COVID and COVID periods in Florida, Texas, and Colorado. Two competing hypotheses are investigated. First, gambling activity may have decreased during COVID either due to increased risk-aversion induced by the pandemic, negative changes in household income from the loss of employment, or a combination of the two. Second, gambling activity may have increased during COVID either due to increased willingness to take risks, substitution of income previously spent on other recreational activities to consumption of gambling products, or a combination of the two. To test these hypotheses, I use linear regression methods to model sales of tickets and revenues over the period from May 2016 to March 2022. My findings reveal that ticket sales decreased across all three states and PowerBall revenues declined significantly in Florida during the COVID period. Additionally, the mean marginal effects of the lagged amount of the jackpot and the unemployment rate differ significantly between the periods.

<p> The study seeks to derive a spatial preference model for urban places, based on a farm population's spatial choices of urban places for retail expenditure. Specifically, the study tests for the similarity in households' preference orderings of urban places, and finds a high degree of similarity. This is achieved using a model with only two simple variables, namely town population and distance to town. Tests indicate no major variable is omitted in the model. The information on households' preference orderings enables the aggregate preference order to be defined. </p> <p> Tests are inconclusive as to whether households also assign worths to urban places which indicate their awareness of the amount by which different places are preferred. </p> <p> The second part of the analysis seeks to determine if different types of households have different spatial preferences. Differences are revealed in the value different types attach to the locational convenience of places. Farm households with members working off the farm reveal a lower than average preference for convenient shopping places. Smaller, more affluent households display a stronger preference for convenient locations and a lower preference for large towns than less affluent households with young children. </p> / Thesis / Doctor of Philosophy (PhD)

In this applied microeconomics dissertation, we study the effect of religiosity and life events on risk preferences and how happiness affect altruism. We begin with the first chapter by examining the relationship between high-risk health behavior and religiosity. Religious beliefs can impact an individual's behavior, including their future health. The 2021 CDC analyses shows that sexually transmitted infections (STIs) are still common, with an estimated 1 in 5 people in the US having an STI, and 13% of persons in the US with age 12 years old and above consuming any illicit drugs. We estimate the effect of religiosity on high-risk health behavior using panel data from the General Social Survey and construct a high-risk health behavior index using the CDC high-risk behaviors. The religiosity index was developed by combining religiosity dimensions such as religious service attendance, prayer frequency, and religious affiliation. Ordered probit was performed to test the relationship between high-risk behavior and religiosity. The result indicates that religiosity is negatively associated with high-risk health behavior and is statistically significant. We confirmed that religious people are less likely to be involved in risky behaviors, especially for Catholics. We also find that individuals who switch religion are more likely to engage in high-risk health behaviors. In the second chapter, we examined the relationship between health and happiness and how happiness impacts altruism. The previous economics literature has shown that altruism can create a warm glow or cause happiness; we tested instead whether happy people are more altruistic. Using data from the General Social Survey (GSS) covering the period of 2002, 2004, 2012, and 2014, we employed a two-stage least square approach and performed OLS and ordered probit regression. We use health as an instrumental variable for happiness. Overall, the results indicate that happiness is associated with volunteer work and giving to charity and provide a basis for policy development to focus on promoting factors that contribute to happiness and wellbeing in order to foster pro-social behavior such as volunteering and donation to charity. Lastly, on the third chapter, we investigate how life events affect risk aversion using the German Socio-economic Panel (SOEP) Data. Our fixed effects estimation suggests that experiencing childbirth and losing parents decrease individuals' risk-taking propensity, while getting separated from a spouse or partner increases the willingness to take risks. We also find that changing jobs increases the willingness to take risks, and individuals who become self-employed tend to take more risks. Furthermore, we examine the average treatment effect on the treated (ATET) and find that for family-related events that are relatively beyond control, such as experiencing childbirth or the death of a child, parent, or spouse or partner, people tend to become more risk-averse. On the other hand, people tend to become less risk-averse for circumstances that are relatively within control, such getting married, separated, and divorced.

This dissertation studies two-person signaling games where the players are assumed to be Choquet expected utility maximizers a la Schmeidler (1989). The sender sends an ambiguous message to the receiver who updates his non-additive belief according to a f-Bayesian updating rule of Gilboa and Schmeidler (1993). When the types are unambiguous in the sense of Nehring (1999), the receiver's conditional preferences after updating on an ambiguous message are always of the subjective expected utility form. This property may serious limit the descriptive power of solution concepts under non-additive beliefs, and it is scrutinized with two extreme f-Bayesian updating rules, the Dempster-Shafer and the Bayes' rule. In chapter 3, the Dempster-Shafer equilibrium proposed by Eichberger and Kelsey (2004) is reappraised. Under the assumption of unambiguous types, it is shown that the Dempster-Shafer equilibrium may give rise to a separating behavior that is never supported by perfect Bayesian equilibrium. However, it does not support any additional pooling equilibrium outcome. Since the Dempster-Shafer equilibrium may support implausible behaviors as exemplified in Ryan (2002), a refinement based on coherent beliefs is suggested. In chapter 4, a variant of perfect Bayesian equilibrium, the quasi perfect Bayesian equilibrium, is proposed, and its descriptive power is investigated. It is shown that the quasi perfect Bayesian equilibrium does not support any additional separating behavior compared to perfect Bayesian equilibrium. It may support additional pooling behavior only if the receiver perceives a correlation between the types and messages. / Ph. D.

No description available.