Browsing by Subject "statistics"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Open Access
    A theory and process evaluation of the Learner Engagement Programme (LEP) implemented by Just Grace NPC
    (2022) Kwenda, Geraldine; Boodhoo, Adiilah
    Background The Learner Engagement Programme (LEP) is an after school dropout prevention programme that operates in Langa, a township located in Cape Town, South Africa. Langa is an impoverished township characterised by socio-economic challenges, such as high unemployment, violence, few economic opportunities and poor school infrastructure. The LEP aims to address learner disengagement of at-risk high school learners. It operates within the only five high schools in Langa: Langa High, Khulani High, Isimela High, Zimasa High, and Ikamva High. The programme is implemented by Just Grace, a non-profit organisation whose goal is to uplift the youth and community of Langa through educational, community and youth development programmes. Just Grace is funded by Trusts and Companies such as DGMT, Mineral Loy (Pty) Ltd, Swiss Philanthropic Foundation, Capfin (Pty) Ltd, Enigma Electrical (Pty) Ltd, Lot Emphangeni (Pty) Ltd, Mergon Foundation, Dairycap CC. Aims of the evaluation The evaluation aimed to determine: (a) the extent to which the programme design can realistically bring about the desired outcomes and (b) the extent to which the programme's planned activities are implemented with fidelity. A programme theory evaluation and process evaluation were carried out to address the following evaluation questions: Programme theory evaluation questions 1) What is the theory and logic underlying the LEP? 2) Is the programme theory and logic plausible? Process evaluation questions 1) Is the programme consistently servicing the planned target population? a. To what extent are learners appropriately identified as at risk? b. Which support services are used the most by learners? c. Are the programme services relevant to meet the learners' needs? 2) Are initial home visits being delivered according to planned programme procedures? 3) Are the programme staff adequately trained and equipped to work with at-risk learners and implement the programme's different components? Methodology The choice of methods for this evaluation was informed by the evaluation questions as well as by the practical opportunities and constraints associated with Level 3 lockdown in South Africa in response to the COVID-19 pandemic. Access to programme beneficiaries was restricted during this time and the risks and concerns associated with face- to- face contact compelled the evaluator to capitalise on available secondary data sources and data gathered from programme staff (through a focus group) to address the process evaluation questions The programme theory evaluation was guided by Donaldson's (2007) systematic five-step framework in conjunction with Brouselle and Champagnes (2011) steps for a logic analysis. An initial LEP programme theory was developed using data obtained through a structured engagement with a purposive sample of four programme staff and a review of relevant programme documents. The plausibility of the programme theory was then examined in line with the best practice literature. Brousselle and Champagne's (2011) steps for a logic analysis were applied to guide process, which culminated into a reconstructed programme theory. To answer the process evaluation questions 1-3, programme documents were systematically analysed. A focus group, which gathered programme staff's experiences of and insights into the current programme infrastructure, challenges, and organisational support, was also conducted. The focus group data was analysed using Krueger's (1994) framework for thematic analysis. Key Findings The programme theory evaluation confirmed that the LEP initial programme theory and logic was plausible: the programme does incorporate a multi-level approach to tackling learner disengagement, targeting the individual, family and community. It addresses the psychosocial aspects that contribute to learner disengagement through the provision of one on one counselling, a life skills programme and parental support groups and training. The programme also has elements of an effective after school programme with qualified staff, adequate resources and efficient programme practices. A few shortcomings were identified through the evaluation: the programme lacks academic support, early warning systems, specialised external partners, and behavioural outcome measures, which are crucial in preventing school dropout. While the evaluator cannot conclusively determine whether the initial home visits are being delivered according to planned procedures (given the limitation of the data at hand), the process evaluation confirmed that the programme the criteria used to identify at risk learners are in line with the best practice literature. The process evaluation also revealed factors that compromised the effective implementation of the programme, including lack of commitment from partner schools, lack of trust in programme methods from parents/caregivers and a lack of staff safety when conducting home visits. Recommendations Key recommendations discussed in this evaluation include the following: • Development of an early warning system in tandem with partner schools as data on atrisk learners needs to be collected earlier in their school career and consistently to ensure the learner receives the necessary assistance timeously and suitable interventions are developed. • Provision of an academic component as learners who are provided with academic support in addition to psychosocial support have a higher chance of school completion. • Development of a behavioural monitoring system as effective programmes utilise behavioural outcome measures to assess programme effects on learners' behaviours • Forging partnerships with external agencies to assist the programme in specialised areas as the programme would benefit by being embedded in a broader network of community-based organisations, NGOs, civil organisations, and government agencies trained to provide specialised support and assistance to their beneficiaries.
  • Thumbnail Image
    Item
    Open Access
    An exploration of alternative features in micro-finance loan default prediction models
    (2020) Stone, Devon; Britz, Stefan
    Despite recent developments financial inclusion remains a large issue for the World's unbanked population. Financial institutions - both larger corporations and micro-finance companies - have begun to provide solutions for financial inclusion. The solutions are delivered using a combination of machine learning and alternative data. This minor dissertation focuses on investigating whether alternative features generated from Short Messaging Service (SMS) data and Android application data contained on borrowers' devices can be used to improve the performance of loan default prediction models. The improvement gained by using alternative features is measured by comparing loan default prediction models trained using only traditional credit scoring data to models developed using a combination of traditional and alternative features. Furthermore, the paper investigates which of 4 machine learning techniques is best suited for loan default prediction. The 4 techniques investigated are logistic regression, random forests, extreme gradient boosting, and neural networks. Finally the paper identifies whether or not accurate loan default prediction models can be trained using only the alternative features developed throughout this minor dissertation. The results of the research show that alternative features improve the performance of loan default prediction across 5 performance indicators, namely overall prediction accuracy, repaid prediction accuracy, default prediction accuracy, F1 score, and AUC. Furthermore, extreme gradient boosting is identified as the most appropriate technique for loan default prediction. Finally, the research identifies that models trained using the alternative features developed throughout this project can accurately predict loan that have been repaid, the models do not accurately predict loans that have not been repaid.
  • Thumbnail Image
    Item
    Open Access
    Complex statistics and diffusion in nonlinear disordered particle chains
    (2014) Antonopoulos, Ch G; Bountis, T; Skokos, Ch; Drossos, L
    We investigate dynamically and statistically diffusive motion in a Klein-Gordon particle chain in the presence of disorder. In particular, we examine a low energy (subdiffusive) and a higher energy (self-trapping) case and verify that subdiffusive spreading is always observed. We then carry out a statistical analysis of the motion in both cases in the sense of the Central Limit Theorem and present evidence of different chaos behaviors, for various groups of particles. Integrating the equations of motion for times as long as $109$, our probability distribution functions always tend to Gaussians and show that the dynamics does not relax onto a quasi-periodic KAM torus and that diffusion continues to spread chaotically for arbitrarily long times.
  • Thumbnail Image
    Item
    Open Access
    INTROSTAT (Statistics textbook)
    (2013) Underhill, Les; Bradfield, Dave
    IntroStat was designed to meet the needs of students, primarily those in business, commerce and management, for a course in applied statistics. IntroSTAT is designed as a lecture-book. One of the aims is to maximize the time spent in explaining concepts and doing examples. The book is commonly used as part of first year courses into Statistics.
  • No Thumbnail Available
    Item
    Open Access
    Teaching fundamental concepts in statistical science
    (2011) Barr, Graham; Scott, Leanne
    These modules are essentially crafted as teaching tools and the experience of first year students would be of the lecturer leading the students through the simulations at an appropriate pace, allowing plenty of opportunity for discussion and clarification. Lab based tutorials also support this process. A suite of VBA simulation programmes used at first year level containing a number of tools for teaching introductory statistics at university level. Note that these are written for MS Excel 2007 (or later versions). The modules roughly follow chapters in the first year statistics textbook, Introstat (LG Underhill) and essentially support and supplement that book. They are to a significant extent self explanatory for those with some knowledge of statistics and simulation.