Browsing by Author "Dunne, Tim"
Now showing 1 - 13 of 13
Results Per Page
Sort Options
- ItemOpen AccessAspects of estimation in the linear model with special reference to collinearity(1994) Thiart, Christien; Dunne, Tim; Troskie, Cas
- ItemOpen AccessAssessment, equity and language of learning : key issues for higher education selection in South Africa(2001) Yeld, Nan; Muller, Johan; Dunne, TimThe central problem investigated by this study arises from the fact that South African Senior Certificate results are not, for the majority of educationally disadvantaged candidates, reliable predictors of academic success in Higher Education. Despite this limitation, however, the Senior Certificate examination plays a vital role in the education system. The aims of the study are thus to investigate procedures that could be used in addition to, rather than instead of, the Senior Certificate, and that would provide useful information about the future academic performance of educationally disadvantaged candidates. The purpose of these procedures is to widen effective access opportunities for such students. It is clear that such procedures need to provide different information from that provided by the Senior Certificate which, like all achievement tests, aims to test learners' understandings in terms of the knowledge and skills covered in a preceding course of instruction. In contexts where great educational disparities exist, as is the case in the South African education system, it is neither fair nor defensible to base key gate-keeping events (such as entry to Higher Education) entirely on performance on such an examination. Apart from issues of fairness, however, for students whose prior opportunities to learn have been grossly inadequate, achievement (curriculum-aligned) tests yield little useful information about candidates' underlying capacities and abilities. The study therefore investigates alternatives to achievement tests, and concludes that non curriculum-aligned testing of core skills and abilities could provide a workable alternative. However, moving from curriculum-aligned to non curriculum-aligned tests can not in itself address the assessment challenge posed in identifying talented students in highly heterogeneous populations, in terms of educational preparation. In such contexts, educationally disadvantaged students will inevitably perform poorly in competition with their more advantaged peers, regardless of the basis of the tests. The study therefore reviews various approaches to what has become known as dynamic assessment, and concludes that non curriculum-aligned, core skills tests developed as far as possible on dynamic lines may represent the most effective and fair approach to assessment in this context. After reviewing major theories of knowing and learning, the roles of language in teaching and learning processes, and the history and possibilities of language testing, a set of specifications (a construct) is developed and proposed as the basis for an academic literacy test designed on dynamic lines. The study then sets out to examine the Placement Tests in English for Educational Purposes (PTEEP), developed by the Alternative Admissions Research Project at the University of Cape Town. These tests aim to provide access opportunities for students whose Senior Certificate results do not necessarily reveal their potential to succeed at UCT. The investigation focuses on the extent to which the tests can be said to be (i) valid in terms of the construct established earlier, and (ii) useful in terms of providing useful, additional information about educationally disadvantaged candidates for selection purposes. In other words, the first part of the study is devoted to developing, on the basis of an extensive literature review, a set of requirements for an academic literacy test for selection to Higher Education in South Africa. The second part of the study assesses the extent to which a series of tests developed by the author and currently being used for selection in this context, can be considered to be valid in terms of the construct established in part one. Given the importance of English Second Language Higher Grade (ESL-HG) as the largest single subject registration in the Senior Certificate, and of English as language of learning, the study includes an investigation of the validity of the ESL-HG examinations, and of the usefulness of ESL-HG results for selection purposes. . The investigation employs both quantitative and qualitative research approaches. in summary, the analysis leads to the following major conclusions: + overall, the PTEEP tests can be considered to be valid in terms of construct and content validity; + the use of scaffolding within a test, for talented educationally disadvantaged candidates, can significantly enhance test performance; + on the basis of survival analysis techniques (Polakow 1999), the PTEEP tests are effective in predicting academic success at UCT. That is, students who score in the top quintile of their candidate pool are significantly less likely to be excluded than are comparable students who are admitted on the basis of their Senior Certificate resits alone. Students who score in the bottom quintile, however, have a very significantly higher risk of exclusion than their peers admitted on the basis of their Senior Certificate results alone; + the PTEEP tests and the ESL-HG examinations exhibit divergent validity (that is, they are not positively associated, but reveal either random or inverse correlations); and + ESL-HG and performance at UCT are not significantly associated. On the basis of these conclusions, the study recommends that Higher Education institutions include, as part of their selection criteria and in addition to Senior Certificate results, a test that is non curriculum-aligned; based on the domain of academic literacy as defined in the study; and developed on the basis of dynamic principles. The study also recommends that the potential contribution of such a test to strengthen quality assurance at the school-leaving/Higher Education interface be investigated by the national Department of Education. Finally, it is recommended that as a matter of urgency, the examining of ESL-HG be investigated, with particular reference to the extent to which the examination targets (and therefore contributes to promoting the development of) cognitive academic language proficiency.
- ItemOpen AccessCalculation of calibration factors from the comparative fishing trial between FRS Africana and RV Dr Fridtjof Nansen(2008) Antony, Luyanda Lennox; Dunne, Tim; Leslie, Rob WIncludes abstract. Includes bibliographical references (leaves 153-157).
- ItemOpen AccessContributions to modern portfolio theory(2001) Raubenheimer, Heidi; Dunne, Tim; Page, MikeFund managers and investors are confronted with the problem of selecting a single investment portfolio from a large number of possible combinations of available assets. In South Africa the set of possible portfolios has become even larger with the gradual relaxing of the constraints on foreign investment from 1995 to the present day, thereby expanding the investment universe for South African investors. Moreover, portfolio selection in South Africa is being transformed increasingly from being the exclusive domain of high net worth individuals, trustees and their investment managers to being the domain and responsibility of the man on the street. The Unit Trust industry started in South Africa in 1965 and gave the lower net worth individual a vehicle with which to invest in a diverse investment portfolio. This industry has proved very popular and has expanded from only 8 funds in 1980 to 338 funds and 136 billion rands under management in November 2000. Moreover the past two years, 1999 and 2000, has seen a change in the pension fund industry from defined benefit (DB) to defined contribution (DC) pension funds, transferring more of the risk and the responsibility of portfolio selection onto pension fund members. With increasing demand for fund management and investment advice by pension fund members and individual investors alike, the financial services industry in South Africa has also expanded. The consequent competition for assets of all descriptions have led, one hopes, to a more efficient market in equity, fixed income and derivative products. Thus modern portfolio theory has come a long way and will have to go further in meeting the demand to assist investors in their decision making.
- ItemOpen AccessEmpirical statistical modelling for crop yields predictions: bayesian and uncertainty approaches(2015) Adeyemi, Rasheed Alani; Guo, Renkuan; Dunne, TimThis thesis explores uncertainty statistics to model agricultural crop yields, in a situation where there are neither sampling observations nor historical record. The Bayesian approach to a linear regression model is useful for predict ion of crop yield when there are quantity data issue s and the model structure uncertainty and the regression model involves a large number of explanatory variables. Data quantity issues might occur when a farmer is cultivating a new crop variety, moving to a new farming location or when introducing a new farming technology, where the situation may warrant a change in the current farming practice. The first part of this thesis involved the collection of data from experts' domain and the elicitation of the probability distributions. Uncertainty statistics, the foundation of uncertainty theory and the data gathering procedures were discussed in detail. We proposed an estimation procedure for the estimation of uncertainty distributions. The procedure was then implemented on agricultural data to fit some uncertainty distributions to five cereal crop yields. A Delphi method was introduced and used to fit uncertainty distributions for multiple experts' data of sesame seed yield. The thesis defined an uncertainty distance and derived a distance for a difference between two uncertainty distributions. We lastly estimated the distance between a hypothesized distribution and an uncertainty normal distribution. Although, the applicability of uncertainty statistics is limited to one sample model, the approach provides a fast approach to establish a standard for process parameters. Where no sampling observation exists or it is very expensive to acquire, the approach provides an opportunity to engage experts and come up with a model for guiding decision making. In the second part, we fitted a full dataset obtained from an agricultural survey of small-scale farmers to a linear regression model using direct Markov Chain Monte Carlo (MCMC), Bayesian estimation (with uniform prior) and maximum likelihood estimation (MLE) method. The results obtained from the three procedures yielded similar mean estimates, but the credible intervals were found to be narrower in Bayesian estimates than confidence intervals in MLE method. The predictive outcome of the estimated model was then assessed using simulated data for a set of covariates. Furthermore, the dataset was then randomly split into two data sets. The informative prior was later estimated from one-half called the "old data" using Ordinary Least Squares (OLS) method. Three models were then fitted onto the second half called the "new data": General Linear Model (GLM) (M1), Bayesian model with a non-informative prior (M2) and Bayesian model with informative prior (M3). A leave-one-outcross validation (LOOCV) method was used to compare the predictive performance of these models. It was found that the Bayesian models showed better predictive performance than M1. M3 (with a prior) had moderate average Cross Validation (CV) error and Cross Validation (CV) standard error. GLM performed worst with least average CV error and highest (CV) standard error among the models. In Model M3 (expert prior), the predictor variables were found to be significant at 95% credible intervals. In contrast, most variables were not significant under models M1 and M2. Also, The model with informative prior had narrower credible intervals compared to the non-information prior and GLM model. The results indicated that variability and uncertainty in the data was reasonably reduced due to the incorporation of expert prior / information prior. We lastly investigated the residual plots of these models to assess their prediction performance. Bayesian Model Average (BMA) was later introduced to address the issue of model structure uncertainty of a single model. BMA allows the computation of weighted average over possible model combinations of predictors. An approximate AIC weight was then proposed for model selection instead of frequentist alternative hypothesis testing (or models comparison in a set of competing candidate models). The method is flexible and easy to interpret instead of raw AIC or Bayesian information criterion (BIC), which approximates the Bayes factor. Zellner's g-prior was considered appropriate as it has widely been used in linear models. It preserves the correlation structure among predictors in its prior covariance. The method also yields closed-form marginal likelihoods which lead to huge computational savings by avoiding sampling in the parameter space as in BMA. We lastly determined a single optimal model from all possible combination of models and also computed the log-likelihood of each model.
- ItemOpen AccessItalian undergraduate students comprehending economics and business texts in English as a foreign language : a case study of language-conceptual transfer strategies used in reading domain-specific texts(2009) Samson, Christina Muriel; Young, Douglas; Paxton, Moragh; Rochford, Kevin; Dunne, TimThis study, within the Vygotskyian constructivist socio-cultural developmental framework, investigates the interdependence of general and domain-specific conceptual knowledge, cognitive and metacognitive strategies, attitude, motivation and context in the process of bidirectional ItalianΓåöEnglish transfer in Italian undergraduates comprehending domain-specific texts in the Faculty of Economics, University of Florence, Italy. The method adopted is primarily qualitative, with quantitative support.
- ItemOpen AccessMathematical, cognitive and didactic elements of the multiplicative conceptual field investigated within a Rasch assessment and measurement framework(2011) Long, Margaret Caroline; Dunne, Tim; Craig, Tracy SHow may the essential elements of a framework including mathematical, cognitive, and didactic elements, and applied in the multiplicative conceptual field, address the challenges in mathematics education, and inform the curriculum and the validity of assessment processes.
- ItemOpen AccessMeeting the requirements of both classroom-based and systemic assessment of mathematics proficiency: the potential of Rasch measurement theory(AOSIS, 2012) Dunne, Tim; Long, Caroline; Craig, Tracy S; Venter, ElsieThe challenges inherent in assessing mathematical proficiency depend on a number of factors, amongst which are an explicit view of what constitutes mathematical proficiency, an understanding of how children learn and the purpose and function of teaching. All of these factors impact on the choice of approach to assessment. In this article we distinguish between two broad types of assessment, classroom-based and systemic assessment. We argue that the process of assessment informed by Rasch measurement theory (RMT) can potentially support the demands of both classroom-based and systemic assessment, particularly if a developmental approach to learning is adopted, and an underlying model of developing mathematical proficiency is explicit in the assessment instruments and their supporting material. An example of a mathematics instrument and its analysis which illustrates this approach, is presented. We note that the role of assessment in the 21st century is potentially powerful. This influential role can only be justified if the assessments are of high quality and can be selected to match suitable moments in learning progress and the teaching process. Users of assessment data must have sufficient knowledge and insight to interpret the resulting numbers validly, and have sufficient discernment to make considered educational inferences from the data for teaching and learning responses.
- ItemMetadata onlyProficiency in the multiplicative conceptual field: using Rasch measurement to identify levels of competence(Taylor & Francis, 2010) Long, Caroline; Dunne, Tim; Craig, Tracy SIn the transition years, Grades 7 to 9, the shift from natural numbers to rational numbers and the associated multiplicative concepts prove challenging for many learners. The new concepts, operations and notation must be mastered if the student is to thereafter rise to meet the challenges of algebra and more advanced and powerful mathematics. The multiplicative conceptual field (MCF) groups together such concepts as fraction, ratio, rate, percentage and proportion, all of which are related yet subtly distinct from one another, each with its own challenges. Rasch analysis allows us to compare the difficulty of mathematical problems located within the MCF while, on the same scale, locating the degree to which individual learners have mastered the necessary skill set. Such location of problems and learners on the same unidimensional scale allows for fine-grained analysis of which aspects of the problems being analysed make one problem more difficult than another. Simultaneously the scale gives the teacher clear evidence of which students have mastered which concepts and skills and which have not, thereby allowing more targeted assistance to the class and individual learners. This paper illustrates the process involved in such analysis by reporting on results located within a larger study. It is suggested that implementing Rasch analysis within the school classroom on appropriately designed assessment instruments would provide clarity for the teacher on the locations of difficulty within the problems used in the assessment and the relative degree to which individual learners are achieving success at mastering the targeted concepts.
- ItemOpen AccessPromoting understanding in mathematical problem-solving through writing : a Piagetian analysis(2007) Craig, Tracy S; Dunne, Tim; Webb, JohnIncludes bibliographical references (leaves 204-221).
- ItemOpen AccessSocial class, pedagogy and achievement in art(2005) Bolton, Heidi; Muller, Johan; Dunne, Tim; Ensor, PaulaThis thesis explores pedagogy associated with high levels of achievement in art in the final year of secondary school, by learners in different social positions. It investigates first, achievement patterns in final-exhibition percentage grades, in relation to learners' social class, race, and gender. Second, based on comparison of high and lower-achieving school classes of socially similar learners, it examines specific pedagogic features linked to success in particular social contexts. The research asks whether or not specific pedagogic features are associated with achievement in art for learners in particular social positions, and if so, whether. These are similar to pedagogic features linked to success in science. It combines different methods to address these questions: a survey to identify art achievement patterns in relation to learners' social class, race and gender, and a multiple case study for detailed exploration of pedagogy and curriculum linked to performance in art, in six school classes. Finding achievement patterned more strongly in relation to social class than race or gender, analysis focuses on social class and achievement.
- ItemOpen AccessStandards of South African Senior Certificate Biology examinations : 1994 to 2007. Volume 1 : Chapters and references.(2012) Crowe, Anna Aletta; Laugksch, Rudiger C; Dunne, TimPublic examinations, such as the South African Senior Certificate (SC) examinations at the end of Grade 12, signal two messages to the society in which they operate: first, the competencies that are valued, that is, its standards; second, the required level of mastery in these competencies that are construed as indicators of success. The SC examinations certified successful students as competent to enter the workforce and, if they obtained a matriculation exemption, qualified them for admission to tertiary study. The SC was not a part of an explicit standards-based curriculum, and there is thus little understanding, but much public speculation, about the relationship between student achievement in the SC examinations, competency and standards. In an attempt to understand this relationship - with a particular focus on the role of standards - in the SC Biology examinations over a period of time, the answer to the following research question was sought: What did the SC Biology examinations in South Africa assess; did their focus change during the period 1994 to 2007; and, if so, what did this change mean?. Both in South Africa and internationally, "standards" is an often-used educational term, the meaning of which has become confused in the literature and by public use. In this study, a methodology to make explicit the standards inherent within the SC Biology examinations - and the relationship between standards and student achievement - was developed, described and applied.
- ItemOpen AccessA variance shift model for outlier detection and estimation in linear and linear mixed models(2009) Gumedze, Freedom Nkhululeko; Dunne, TimOutliers are data observations that fall outside the usual conditional ranges of the response data.They are common in experimental research data, for example, due to transcription errors or faulty experimental equipment. Often outliers are quickly identified and addressed, that is, corrected, removed from the data, or retained for subsequent analysis. However, in many cases they are completely anomalous and it is unclear how to treat them. Case deletion techniques are established methods in detecting outliers in linear fixed effects analysis. The extension of these methods to detecting outliers in linear mixed models has not been entirely successful, in the literature. This thesis focuses on a variance shift outlier model as an approach to detecting and assessing outliers in both linear fixed effects and linear mixed effects analysis. A variance shift outlier model assumes a variance shift parameter, !i, for the ith observation, where !i is unknown and estimated from the data. Estimated values of !i indicate observations with possibly inflated variances relative to the remainder of the observations in the data set and hence outliers. When outliers lurk within anomalous elements in the data set, a variance shift outlier model offers an opportunity to include anomalies in the analysis, but down-weighted using the variance shift estimate Ë!i. This down-weighting might be considered preferable to omitting data points (as in case-deletion methods). For very large values of !i a variance shift outlier model is approximately equivalent to the case deletion approach. We commence with a detailed review of parameter estimation and inferential procedures for the linear mixed model. The review is necessary for the development of the variance shift outlier model as a method for detecting outliers in linear fixed and linear mixed models. This review is followed by a discussion of the status of current research into linear mixed model diagnostics. Different types of residuals in the linear mixed model are defined. A decomposition of the leverage matrix for the linear mixed model leads to interpretable leverage measures. ii A detailed review of a variance shift outlier model in linear fixed effects analysis is given. The purpose of this review is firstly, to gain insight into the general case (the linear mixed model) and secondly, to develop the model further in linear fixed effects analysis. A variance shift outlier model can be formulated as a linear mixed model so that the calculations required to estimate the parameters of the model are those associated with fitting a linear mixed model, and hence the model can be fitted using standard software packages. Likelihood ratio and score test statistics are developed as objective measures for the variance shift estimates. The proposed test statistics initially assume balanced longitudinal data with a Gaussian distributed response variable. The dependence of the proposed test statistics on the second derivatives of the log-likelihood function is also examined. For the single-case outlier in linear fixed effects analysis, analytical expressions for the proposed test statistics are obtained. A resampling algorithm is proposed for assessing the significance of the proposed test statistics and for handling the problem of multiple testing. A variance shift outlier model is then adapted to detect a group of outliers in a fixed effects model. Properties and performance of the likelihood ratio and score test statistics are also investigated. A variance shift outlier model for detecting single-case outliers is also extended to linear mixed effects analysis under Gaussian assumptions for the random effects and the random errors. The variance parameters are estimated using the residual maximum likelihood method. Likelihood ratio and score tests are also constructed for this extended model. Two distinct computing algorithms which constrain the variance parameter estimates to be positive, are given. Properties of the resulting variance parameter estimates from each computing algorithm are also investigated. A variance shift outlier model for detecting single-case outliers in linear mixed effects analysis is extended to detect groups of outliers or subjects having outlying profiles with random intercepts and random slopes that are inconsistent with the corresponding model elements for the remaining subjects in the data set. The issue of influence on the fixed effects under a variance shift outlier model is also discussed.