Previous research has noted that members of research ethics committees are unclear about the extent of their roles. In this study, research amongst members of independent ethics committees (IECs) about how the ‘expert’ and ‘lay’ roles are understood and operationalized offers an explanation for this lack of clarity. IECs were selected for study because they have only addressed one type of research (Phase 1 ‘healthy volunteer’ studies) and this limited remit suggested that it would be in such committees that the member roles would have become most pronounced. Drawing on findings from the sociology of professions and employing a phenomenological approach to understanding, 20 semi-structured interviews with both expert and lay members of these committees revealed that a number of members were not only unclear about the roles, but unclear too whether they, or certain of their colleagues, were in which membership category. Notwithstanding this fact, and paradoxically, the ‘expert’ designation was seen as granting its members a privileged position on the committees. The expert member was seen to be either a medically qualified member or one tightly associated with the medical model. Such a repository of expertise being with the medical model privileges this model in ethics review such that other matters formally to be scrutinized by ethics committees become marginalised. Participant safety was the prime concern of the ethics review for IEC members. This relegated other matters including the adequacy of the insurance arrangements, the readability of the consent forms, the fairness of the inclusion criteria, and so forth, into areas of lesser concern. That this occurs though when the science, the safety and the methodology of the trials are already – separately - subject to an independent analysis by a body of experts, whose statutory role is to concern itself with these issues such that no trial may occur without their sanction, is of significance. IEC members were cognizant of this duplication of role but unable to resolve it. The situation could be accounted for as due to capture by the medical model and a cognitive dissonant process. Members’ training and education were found to have been neglected because under the medical professions’ gaze no other type of knowledge was considered necessary in ethics review. The study revealed that the medical profession’s dominance of such committees accounts for the members’ role uncertainty and as such allies itself to Freidson’s theory of professional dominance. If such a concept has been thought to be an obsolete one, this study suggests such a notion of the status of the theory is premature. The medical model’s status is implicitly accepted such that nothing else need be considered. The research calls for further studies to corroborate such findings in other research ethics settings and for a debate about what society wants its ethics committees to focus upon in their review.

Concept drift refers to a problem that is caused by a change in the data distribution in data mining. This leads to reduction in the accuracy of the current model that is used to examine the underlying data distribution of the concept to be discovered. A number of techniques have been introduced to address this issue, in a supervised learning (or classification) setting. In a classification setting, the target concept (or class) to be learned is known. One of these techniques is called “Ensemble learning”, which refers to using multiple trained classifiers in order to get better predictions by using some voting scheme. In a traditional ensemble, the underlying base classifiers are all of the same type. Recent research extends the idea of ensemble learning to the idea of using committees, where a committee consists of diverse classifiers. This is the main difference between the regular ensemble classifiers and the committee learning algorithms. Committees are able to use diverse learning methods simultaneously and dynamically take advantage of the most accurate classifiers as the data change. In addition, some committees are able to replace their members when they perform poorly. This thesis presents two new algorithms that address concept drifts. The first algorithm has been designed to systematically introduce gradual and sudden concept drift scenarios into datasets. In order to save time and avoid memory consumption, the Concept Drift Introducer (CDI) algorithm divides the number of drift scenarios into phases. The main advantage of using phases is that it allows us to produce a highly scalable concept drift detector that evaluates each phase, instead of evaluating each individual drift scenario. We further designed a novel algorithm to handle concept drift. Our Dynamic Committee for Concept Drift (DCCD) algorithm uses a voted committee of hypotheses that vote on the best base classifier, based on its predictive accuracy. The novelty of DCCD lies in the fact that we employ diverse heterogeneous classifiers in one committee in an attempt to maximize diversity. DCCD detects concept drifts by using the accuracy and by weighing the committee members by adding one point to the most accurate member. The total loss in accuracy for each member is calculated at the end of each point of measurement, or phase. The performance of the committee members are evaluated to decide whether a member needs to be replaced or not. Moreover, DCCD detects the worst member in the committee and then eliminates this member by using a weighting mechanism. Our experimental evaluation centers on evaluating the performance of DCCD on various datasets of different sizes, with different levels of gradual and sudden concept drift. We further compare our algorithm to another state-of-the-art algorithm, namely the MultiScheme approach. The experiments indicate the effectiveness of our DCCD method under a number of diverse circumstances. The DCCD algorithm generally generates high performance results, especially when the number of concept drifts is large in a dataset. For the size of the datasets used, our results showed that DCCD produced a steady improvement in performance when applied to small datasets. Further, in large and medium datasets, our DCCD method has a comparable, and often slightly higher, performance than the MultiScheme technique. The experimental results also show that the DCCD algorithm limits the loss in accuracy over time, regardless of the size of the dataset.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1968. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Includes bibliographical references (leaves 328-340).

The purpose of this research was to empirically examine the relationship between audit committees and the changing of independent accounting firms, especially those independent auditor changes in which accounting firms are displaced as auditors by accounting firms that are in the larger-sized category. This research involved the testing of two different, but closely related questions. The first research question asks, "Does the existence of an audit committee of a board of directors partially explain a company's decision to change independent accounting firms?" The second research question asks, "Does the existence of an audit committee of a board of directors partially explain a company's decision to change to a larger accounting firm rather than change to an accounting firm of equal or smaller size?" Statistical models were developed for the two research questions using criteria involved in the auditor change decisions as reported in previous research studies. Eight criterion, including the existence of an audit committee, were defined specifically and incorporated into a survey instrument. The survey instrument was circulated to the accounting firms in the Dallas-Fort Worth area. The accounting firms were asked to report certain information regarding their current and previous auditing clients. Data received from the firms were analyzed using multiple regression techniques .

Concept drift refers to a problem that is caused by a change in the data distribution in data mining. This leads to reduction in the accuracy of the current model that is used to examine the underlying data distribution of the concept to be discovered. A number of techniques have been introduced to address this issue, in a supervised learning (or classification) setting. In a classification setting, the target concept (or class) to be learned is known. One of these techniques is called “Ensemble learning”, which refers to using multiple trained classifiers in order to get better predictions by using some voting scheme. In a traditional ensemble, the underlying base classifiers are all of the same type. Recent research extends the idea of ensemble learning to the idea of using committees, where a committee consists of diverse classifiers. This is the main difference between the regular ensemble classifiers and the committee learning algorithms. Committees are able to use diverse learning methods simultaneously and dynamically take advantage of the most accurate classifiers as the data change. In addition, some committees are able to replace their members when they perform poorly. This thesis presents two new algorithms that address concept drifts. The first algorithm has been designed to systematically introduce gradual and sudden concept drift scenarios into datasets. In order to save time and avoid memory consumption, the Concept Drift Introducer (CDI) algorithm divides the number of drift scenarios into phases. The main advantage of using phases is that it allows us to produce a highly scalable concept drift detector that evaluates each phase, instead of evaluating each individual drift scenario. We further designed a novel algorithm to handle concept drift. Our Dynamic Committee for Concept Drift (DCCD) algorithm uses a voted committee of hypotheses that vote on the best base classifier, based on its predictive accuracy. The novelty of DCCD lies in the fact that we employ diverse heterogeneous classifiers in one committee in an attempt to maximize diversity. DCCD detects concept drifts by using the accuracy and by weighing the committee members by adding one point to the most accurate member. The total loss in accuracy for each member is calculated at the end of each point of measurement, or phase. The performance of the committee members are evaluated to decide whether a member needs to be replaced or not. Moreover, DCCD detects the worst member in the committee and then eliminates this member by using a weighting mechanism. Our experimental evaluation centers on evaluating the performance of DCCD on various datasets of different sizes, with different levels of gradual and sudden concept drift. We further compare our algorithm to another state-of-the-art algorithm, namely the MultiScheme approach. The experiments indicate the effectiveness of our DCCD method under a number of diverse circumstances. The DCCD algorithm generally generates high performance results, especially when the number of concept drifts is large in a dataset. For the size of the datasets used, our results showed that DCCD produced a steady improvement in performance when applied to small datasets. Further, in large and medium datasets, our DCCD method has a comparable, and often slightly higher, performance than the MultiScheme technique. The experimental results also show that the DCCD algorithm limits the loss in accuracy over time, regardless of the size of the dataset.

Thesis (MPA.) --University of Limpopo, 2017 / This study was undertaken to evaluate the effectiveness of the oversight role of the Portfolio Committee on Education, in determining their influence on accountability functions by the Department of Education. The overall aim is to provide an understanding into inherent challenges in the oversight role of the Portfolio Committee on Education in the Limpopo Legislature. The study further aimed to collect data on the role played by the Members of the Education Portfolio Committee on how they exercise their oversight role in ensuring that the Department of Basic Education delivers quality service to the public. As such, the study was conducted from a qualitative perspective; using interviews and documentation analysis. The results of this study indicate that although Members of the Limpopo Legislature, particularly Members of the Portfolio Committee on Education, are capacitated to conduct oversight over the Department of Basic Education, there are still gaps and challenges as far as the oversight role of Parliament is concerned. The study further reveals that the Portfolio Committee’s ineffectiveness when performing its oversight may be attributed to the Department’s lack of co-operation, submission of poor quality reports, non-compliance with rules and regulations, Public Finance Management Act (PFMA) (and other policies), deviation from the budget, unwillingness to implement the Committee’s recommendations and failure to submit required documentation to Standing Committee on Public Accounts (SCOPA) and the Auditor General’s (AG) Office. / Limpopo Legislature

The purpose of this research was to empirically examine the effect of the establishment of outside audit committees on the objectivity of the management results-reporting practices of those companies that established such committees in response to the New York Stock Exchange mandate effective June 30, 1978. Management income smoothing behavior is taken as a measurable surrogate for the objectivity of the management results-reporting practices. This research involved the testing of one research problem. The research question asks, "Will the establishment of outside audit committees by companies that had no such committees prior to the New York Stock Exchange mandate effective June 30, 1978, be associated with a decrease in the degree of smoothing in the net income series for the period after that date relative to the degree of smoothing prior to that date?" The answer to this question required the selection of an experimental and a control group. Each group was composed of fifty New York Stock Exchange listed firms. Linear and semi-log regression models were used to measure each firm's degree of income smoothing (defined as reducing the variability of a net income series about its trend line). The change in mean square errors of the experimental and control groups was compared using the chisquare and median tests. Neither the chi-square or the median test found a statistically significant increase in the objectivity of the management results-reporting function for the firms that established outside audit committees in response to the NYSE mandate effective June 30, 1978.

Magister Legum - LLM / South Africa

The thesis explores the relationship between Audit Committees and External Auditors’ fees of a sample of FTSE 350 companies in the UK for the period of 2005-2006. This is achieved by providing answers to three main research questions. First, what are the determinants of Audit Committee activity? Second, what is the relationship between Audit Committee activity and external auditors’ fees? Third, what is the relationship between audit and non-audit fees and how does the Audit Committee affect these? Starting out with an Agency Theoretical background, the study found evidence consistent with the views that a higher proportion of Independent Non-Executive Directors on the board enhances Audit Committees’ activity, but the presence of financial expertise on the committee was not found to be statistically significant in explaining its activity. The thesis also documented evidence that shows that Audit Committee activity is inversely related to managerial ownership of shares in companies. In line with the economic theory of auditing, the researcher used fees paid to the external auditor to proxy for the level of economic bonding between auditors and their clients. Higher fees are interpreted to indicate compromised independence. Five alternative measures of economic bonding were used. The researcher found a stable and statistically significant positive relationship between measures of economic bonding and Audit Committee activity. This finding is consistent with the view that Audit Committees buy more services from the auditors in order to enhance auditing and reporting quality. Strong positive relationships between audit and non-audit services and vice versa were found using a single equation fees model but these relationships were not consistent when the researcher controlled for endogeneity between audit and non audit fees using Simultaneous Equation Models (SEM). Audit Committee activity was not statistically important in these relationships. This evidence taken together supports the proposition that economies of scope exist in the joint provision of both audit and non-auditing services to the same client. Finally the thesis also documents evidence that suggests that knowledge spill-over flows from non-audit services to auditing services and that auditor do not use audit as a loss leader.

The Local Drug Action Committee (LDAC) is the closest body to the community which is made up of individuals from different stakeholders at a local level to coordinate the activities aimed at the prevention of substance abuse and illicit drug trafficking within the communities. Since the implementation of the committees, the sustainability of these committees has been a challenge in the Free State Province. The study endeavoured to explore factors which influence the sustainability of these committees in the Free State Province. A qualitative research approach that was explorative and descriptive in nature was employed. A case study design was used. Purposive sampling was applied to select the participants from two LDACs in Mangaung Municipality. A semi-structured interview schedule was used to collect the data during individual in-depth interviews with five participants. Thematic content analysis was used to analyse the data. The main findings indicated that there was a lack of knowledge among the committee members about their roles, responsibilities and mandates. Their inability to coordinate the stakeholder activities as well as the limited knowledge about the relationship between the Free State Mini Drug Master Plan (FSMDMP) and the LDACs also appeared to be challenging. The main conclusions drawn from the study are that despite the excellent structure of the LDACs to coordinate the implementation of the FSMDMP, the members of the LDACs are neither knowledgeable about nor capacitated to fulfil their roles and responsibilities to meet the intended requirements and implement the activities of the FSMDMP, factors which hamper sustainability. The lack of financial support from Government to fund the activities of the LDACs in the Province is another major impediment in terms of sustainability.