Master of Science in Biostatistics

Purpose

The purpose of this qualification is to

• Develop highly qualified and skilled Biostatisticians with technical competence to contribute significantly to addressing public health problems and challenges of their communities.
• Prepare learners who aspire to move to a higher level of academic research work for Doctoral Studies and to promote an approach based on academic integrity and professional ethics.
• Contribute to the pool of academics and professionals with the competence and critical intellectual abilities to ensure advancement in the field of Biostatistics and to make provision for the country's needs in a skilled scientific workforce of the highest calibre.
• Prepare learners who can apply principles of statistical reasoning in addressing critical questions in public health and Biomedical Sciences.
• Educate and train individuals who are independent thinkers and lead the data management and statistical analyses of research studies to advance knowledge and guide policy.

Rationale

The growing emphasis on evidence-based public health and the collection of vast amounts of increasingly complex health data have resulted in a greater need for high-level Biostatistical skills within the public health workforce. The Master in Biostatistics trains learners in areas of probabilistic and statistical theory, Biostatistical and Bioinformatics methods in planning studies, conducting analyses, and writing reports, the interpretation of numeric data for scientific inference in studies in Medicine and Public Health, and the ability to communicate effectively with Scientists in related disciplines. Application areas include observational studies, clinical trials, statistical genetics, and medical and public health research, among other areas. This qualification will satisfy the unmet demand for Researchers with these skills from academia, other research institutions, government and private sector in South Africa, the sub-Saharan Africa region and internationally as well. The qualification caters for two types of cadres: learners with a strong Statistical/Mathematical background at the Honours level who require more practical training; and learners with a biomedical background who need the theoretical foundations of Biostatistics. The Master of Science (MSc) Biostatistics qualification prepares learners for further Postgraduate studies at Doctoral level.

Recognition of Prior Learning (RPL)

The Master of Science in Biostatistics qualification conforms to the Institution's Policy for the Assessment and Recognition of Prior Learning (ARPL) as well as the ARPL policy of the Faculty of Health Sciences.

The policy defines the process that must be followed in the assessment of an application for Recognition of Prior Learning (RPL) at postgraduate level and provides information pertaining to assessment tools that can be used in the assessment of RPL applications. The APRL process is subjected to the Faculty's quality assurance process. Unsuccessful applicants may appeal according to the relevant appeal procedures as stated in the University's General Calendar Part 1.

Entry Requirements

• An Honours Degree in the following areas: Mathematics, Statistics, Biomedical Sciences, Medical Degree (Bachelor of Medicine, Bachelor of Surgery) or a cognate qualification.

Compulsory Modules at NQF Level 9, 156 Credits

• Mathematical Statistics (MS), 6 Credits.
• Principles of Statistical Inference (PSI), 12 Credits.
• Linear Models (LM), 6 Credits.
• Fundamentals of Epidemiology (FoE), 6 Credits.
• Data management and Statistical computing with R and Stata (DMC), 6 Credits.
• Categorical Data and Generalized Linear Models (CDGLM), 12 Credits.
• Analysis of Survival Data (ASD), 12 Credits.
• Analysis of Observational Data: Causal Inference (AODCI), 12 Credits.
• Longitudinal Data Analysis (LDA), 12 Credits.
• Biostatistical consulting (BC), 12 Credits.
• Research Assignment (A), 60 Credits.

Optional Modules at NQF Level 9 (Learners to choose 2 electives)

• Bayesian Data Analysis (BDA), 12 Credits.
• Clinical Biostatistics (CB), 12 Credits.
• Design and Analysis of Clinical Trials (DACT), 12 Credits.
• Multivariate Statistics (MS), 12 Credits.
• Bio-informatics, 12 Credits.

1. Demonstrate Principles of Statistical Inference using a range of Statistical tools.
2. Demonstrate skills in data collection and data management, including quality control procedures and the ethical handling of data.
3. Conduct research in Biostatics field.

Associated Assessment Criteria for Exit Level Outcome 1

• Fundamental concepts in statistical inference, their practical interpretation and importance in Biostatistical contexts are explained.
• The theoretical basis for Frequentists and Bayesian approaches to statistical inference are applied.
• Parametric methods of inference, with particular reference to problems of relevance in bio statistical contexts are developed and applied.
• An understanding of the theoretical basis to justify more complex statistical procedures is demonstrated.
• Basic alternatives to standard likelihood-based methods are explained, and situations in which these methods are useful are identified.
• Theoretical concepts of day to day problems in a workplace environment are applied.
• Guidance in study design and statistical analysis plans is provided.
• A solution to a statistical problem arising from a workplace environment is proposed.

Associated Assessment Criteria for Exit Level Outcome 2

• Data using two major statistical software packages (Stata and R) is manipulated and managed.
• Data is displayed and summarised using statistical software.
• Fundamental programming skills for efficient use of software packages are demonstrated.
• The statistical methods utilised to analyse longitudinal data in a variety of settings and types of outcome variables are described.
• A scientific problem that requires repeated measurements is analysed, appropriate design is identified, and the statistical methods required to analyse the data are identified.
• Stata and/or R procedures to perform analysis of longitudinal data are utilised.
• Modern methods for the analysis of longitudinal data to a range of settings encountered in biomedical and public health research are applied.
• The clinical/scientific mean of the results from a longitudinal data analysis is interpreted and communicated.
• Generalised linear models (GLMs) and other methods to analyse categorical data with proper attention to the underlying assumptions are used.
• The key principles regarding confidentiality and privacy in data storage, management and analysis are understood.

Associated Assessment Criteria for Exit Level Outcome 3

• Research questions are framed.
• A wide range of basic and complex statistical tools for solving research problems are critically evaluated and utilised.
• Biostatistical consulting sessions with researchers are held.
• Practical Bayesian analysis relating to health research problems is performed.
• Results are interpreted and communicated to researchers.
• Biomedical researchers are consulted and advice on study design and statistical analysis plans is provided.
• Leadership in research studies is provided.

Integrated Assessment

Formative assessments are performed during each module as determined by the course convenor. This is done via the monitoring procedures for the qualification.

The summative assessment is composed of 50% by continuous assessment during the module, and 50% by a final examination in each of the modules. Learners have to achieve a 45% pass in the continuous assessments before being allowed to participate in the final examination.

A portfolio of project reports composed of summarises of work done detailing contributions to study design, statistical methodology, data analysis and report writing will be used to assess the development of the learner's consulting skills. The learners have to complete one research project at the appropriate level - to the satisfaction of the programme committee - following the Institution regulations for research projects of structured Master's qualifications.