<div>This thesis combines human and machine intelligence for consensus decision-making, and it contains four interrelated research areas. Before presenting the four research areas, this thesis presents a literature review on decision-making using two criteria: trust and risk. The analysis involves studying the individual and the multi-stakeholder decision-making. Also, it explores the relationship between trust and risk to provide insight on how to apply them when making any decision. This thesis presents a grouping procedure of the existing trust-based multi-stakeholder decision-making schemes by considering the group decision-making process and models. In the first research area, this thesis presents the foundation of building multi-stakeholder consensus decision-making (MSCDM). This thesis describes trust-based multi-stakeholder decision-making for water allocation to help the participants select a solution that comes from the best model. Several criteria are involved when deciding on a solution such as trust, damage, and benefit. This thesis considers Jain's fairness index as an indicator of reaching balance or equality for the stakeholder's needs. The preferred scenario is when having a high trust, low damages and high benefits. The worst scenario involves having low trust, high damage, and low benefit. The model is dynamic by adapting to the changes over time. The decision to select is the solution that is fair for almost everyone. In the second research area, this thesis presents a MSCDM, which is a generic framework that coordinates the decision-making rounds among stakeholders based on their influence toward each other, as represented by the trust relationship among them. This thesis describes the MSCDM framework that helps to find a decision the stakeholders can agree upon. Reaching a consensus decision might require several rounds where stakeholders negotiate by rating each other. This thesis presents the results of implementing MSCDM and evaluates the effect of trust on the consensus achievement and the reduction in the number of rounds needed to reach the final decision. This thesis presents Rating Convergence in the implemented MSCDM framework, and such convergence is a result of changes in the stakeholders' rating behavior in each round. This thesis evaluates the effect of trust on the rating changes by measuring the distance of the choices made by the stakeholders. Trust is useful in decreasing the distances. In the third research area, this thesis presents Rating Convergence in the implemented MSCDM framework, and such convergence is a result of changes in stakeholders' rating behavior in each round. This thesis evaluates the effect of trust on the rating changes by measuring the perturbation in the rating matrix. Trust is useful in increasing the rating matrix perturbation. Such perturbation helps to decrease the number of rounds. Therefore, trust helps to increase the speed of agreeing upon the same decision through the influence. In the fourth research area, this thesis presents Rating Aggregation operators in the implemented MSCDM framework. This thesis addresses the need for aggregating the stakeholders' ratings while they negotiate on the round of decisions to compute the consensus achievement. This thesis presents four aggregation operators: weighted sum (WS), weighted product (WP), weighted product similarity measure (WPSM), and weighted exponent similarity measure (WESM). This thesis studies the performance of those aggregation operators in terms of consensus achievement and the number of rounds needed. The consensus threshold controls the performance of these operators. The contribution of this thesis lays the foundation for developing a framework for MSCDM that facilitates reaching the consensus decision by accounting for the stakeholders' influences toward one another. Trust represents the influence.</div>

No description available.

[pt] Neste trabalho apresentamos algumas contrubuições ao estudo dos modelos de avaliação estatística usados nas ciências sociais. As contribuições originais são: i ) uma descrição unificada sobre como a teoria da medição evoluiu nas diversas disciplinas científicas; ii ) uma resenha abrangente sobre os métodos de estimação por máxima verossimilhança empregados na medição estatística; iii ) uma formulação geral do métodos da máxima verossimilhan ça tendo em vista a aplicação em modelos não-lineares; e principalmente, iv ) a apresentação do método da aproximação estocástica na estimação dos modelos estatísticos de avaliação e medição. Os modelos não-lineares ocorrem freqüentemente nas ciências sociais onde é importante a modelagem de variáveis de resposta dicotômicas ou ordinais. Em particular, este trabalho trata dos modelos da teoria da resposta ao item, dos modelos de regressão logística e dos modelos de componentes aleatórias em geral. A estimação destes modelos ainda é objeto de intensa pesquisa. Não se pode afirmar que exista um método de estimação inteiramente confiável. Os métodos aproximados produzem estimativas com viés acentuado nas componentes de variância, enquanto os métodos de integração numérica e os métodos bayesianos podem apresentar problemas de convergência em muitos casos. O método da aproximação estocástica se baseia na maximização da verossimilhança e emprega o algoritmo de Robbins- Monro para resolver a equação do escore. Como um método estocástico ele gera um processo de Markov que se aproxima das estimativas desejadas e portanto pode ser considerado um MCMC (Monte Carlo Markov chain) freqüentista. Nas simulações realizadas o método apresentou um bom desempenho, produzindo estimativas com viés pequeno, precisão razoável e raros problemas de convergência. / [en] This work presents a study of statistical models used for assessment and measurement in the social sciences. The main contributions are: i ) a unified description of how evaluation, assessment, and the theory of measurement evolved within several branches of science; ii ) a review of estimation methods currently employed in nonlinear models; iii ) a general formulation of the maximum likelihood estimation method; and particularly, iv the presentation of the stochastic approximation method for estimation of non linear statistical models in measurement and assessment. Non linear models occurs frequently in the social sciences where it is important to model binary or ordinal response variables. This work deals with item response theory models, logistic regression models and general models with random components. The estimation of these models has been the subject of several recent simulation studies. One cannot say there is a best estimation method. The approximate methods are known to produce biased estimates, numerical integration methods and bayesian methods can present convergence problems in many cases. Stochastic approximation method is a maximum likelihood method that uses the Robbins-Monro algorithm to solve the score equation. As a stochastic approximation method it generates a Markov chain that converges to the desired estimates and can be considered a frequentist MCMC. A simulation study and a comparative estimation study show a good performance, the method producing small bias for the estimates, good precision, and very rare convergence problems.

Dynamic assessment, as manner of alternative process-based assessment, is currently at a cross-roads chiefly characterised by, at times, vague conceptualisation of terminology, blurred demarcation as to its model and theory status and at times ill-defined fundamental philosophy. As a movement in modern psychological assessment within the broader field of intelligence, dynamic assessment does not present with a coherent unifying theory as such and due to its lack of clarity in a number of key areas its eventual disuse might well be the final outcome of this method and its unique history and methodology. In pursuit of this study’s main goal, dynamic assessment models and theories are critically explored by means of a meta-theory largely inspired by the work K.B. Madsen, a Danish meta-theorist and pioneer in theoretical psychology. Madsen’s meta-theory is attenuated in order to suit the nature and purposes of this study; so as to better analyse dynamic assessment within intelligence research and assessment. In its primary aim, this study builds on a foundation of epistemological and ontological considerations within science in general, the social sciences and psychology in particular. In keeping with Madsen’s method of meta-theory analysis, the author’s predilections are stated at the outset in order to place the progression of analyses of the various models and theories within dynamic assessment. Dynamic assessment and intelligence are discussed and a brief digression into the history of Soviet psychology is offered as it is pertinent to the work of Lev Vygotsky and its subsequent influence within process-based assessment. Theory and model development within science and the social sciences are described from a philosophy-of-science vantage point. Psychological assessment’s prime considerations are critically explored and the discussion highlights the role played by the philosophical aspects of mathematics and statistical foundations as leveraging measurement within assessment. Particular attention is paid to the perennial controversy surrounding null hypothesis significance testing and the possible future directions that can be explored by and within dynamic assessment which lends itself to approaches less restrictive than those offered by mainstream statistics. The obvious and not so obvious aspects within the mathematical, statistical and measurement foundations are critically explored in terms of how best dynamic assessment can manoeuvre within the current mainstream psychological assessment system and how new models of item response theory suited to change-based assessment can be explored as possible manner of handling the gain score issue; itself a paradoxical state of affairs within classical and modern test theory. Dynamic assessment’s past has in large part been dictated by mainstream considerations in the areas mentioned and in order to place itself on an alternative path these considerations are critically assessed in terms of dynamic assessment’s future path. Dynamic assessment and its place within the broader intelligence assessment field is then investigated by means of the meta-theory developed. It is envisaged that the intuitive appeal of dynamic assessment will continue to garner support from practitioners across the globe, specifically those trained in countries outside the traditional stronghold of Western psychological theory. However, the position taken in this argument is that in order to ensure its survival it will need to make a decision in terms of its future progress: either to branch off from mainstream assessment altogether or to become fused within mainstream assessment. The “best of both worlds” scenario has obviously not worked out as it was originally hoped. The study concludes with the meta-theoretical exploration of dynamic assessment within intelligence by utilising a small selection of current models. The application of the attenuated Madsenian framework seeks to explore, place and ascertain the nature of each model regarding the ontological and philosophical status of the approach; the nature of the hypothetical terminology, scientific hypotheses and hypothesis system utilised and lastly the nature of the abstract data, concrete data and prime considerations as implicit concerns within the varied approaches. An HQ score is calculated for each such model and is a partial indicator of the testability (verifiability or falsifiability) of the model in question. The models are thus couched in meta, hypothetical and data strata and can be positioned on a continuum of sorts according to which tentative claims can be made regarding the veracity of the approach behind each model. The study concludes with two appendices; a meta-analysis which was conducted on South African research in the field of dynamic assessment (1961-2002) and which cumulated in a significant effect size evidencing an overall positive effect that dynamic assessment has had as an alternative intervention technique in comparison to conventional or static based assessment models. In order to encourage replication of this study, all details pertaining to the studies included for consideration in the meta-analyses are attached in section 2 of this appendix. Secondly, an informal content analysis was conducted on eleven responses to questionnaires that were originally delivered to one hundred dynamic assessment practitioners and researchers across the globe. The purpose of the questionnaire was to ascertain information on core issues within dynamic assessment, as these fundamental issues were considered as pivotal in the future of this approaches’ eventual development or stagnation. The analysis concluded that dynamic assessment is indeed perceived to be at a crossroads of sorts and thus supported the initial hypothesis stated above. It is hoped that this theoretical study will aid in aligning dynamic assessment in a manner such that its eventual place in psychological assessment will be solidly grounded, theoretically defensible and viable as alternative manner of assessment. / Thesis (PhD (Psychology))--University of Pretoria, 2007. / Psychology / PhD / PhD / unrestricted