Postgraduate Diploma in Data Science

In our current data era, more and more jobs are involving working with data, where graduates from different academic backgrounds are challenged to turn overwhelming amounts of data into actionable insights for industries. Consequently, it was identified that industries are in need of an analytical skill enhancement program for such employees. Similarly, the employees themselves may also be interested in enhancing their analytical skills for their own career development, either within their current industry or beyond.

The Postgraduate Diploma (PGDip) in Data Science was thus developed to allow graduates with industry experience and a strong passion for data management and analytics to enhance their skills in the field of Data Science. This PGDip also serves as a "bridge" to the Masters in Data Science for graduates from Engineering, Commerce and Science who are interested in pursing a career in Data Science or who wish to upskill in this field.

About the Program and Curriculum:

This is a two-year program that results in a Postgraduate Diploma (NQF Exit Level 8). The modules were designed to complement one another to form a coherent program with a focus on producing well-rounded professionals in analytical Data Science. The modules cover a range of statistical and machine learning concepts as well as statistical software. Focus is on the application of the methodology rather than the theory. Furthermore, bootcamps are held to expose the students to a range of software.

This program is facilitated by the Discipline of Statistics, which is home to a diverse group of academics specializing in a wide array of Statistics and Data Science teaching and research areas.

How the program runs:

A hybrid approach has been adopted where online lectures take place after hours (from 6pm) on weekdays, with on-campus sessions every few weeks. These contact sessions can be used for teaching, practicals and assessments. All exams take place on campus. Therefore, this program is not fully online.

Cost of the program:

Fees are charged per module. Click on the following for the fees for year 1 and year 2 in 2025. Please note that these fees will be subject to a slight increase each year.

International students are required to pay an additional levy. More information on this can be found here.

IMPORTANT: If accepted into the program, a registration fee of approximately R5950 (amount subject to change each year) must be paid by the deadline (typically mid-February) to complete your registration into the program. This registration fee gets deducted off your fees for the year. See more info about financial clearance here. International students are required to pay full fees as well as an international levy.

Curriculum:

Two modules are taken per semester for the first 3 semesters, followed by an industry project in the last semester. The curriculum is as follows:

Year 1

Semester 1:

Data Mining: Descriptive Analytics (STAT603)
Aim: To introduce advanced exploratory techniques of descriptive data analytics with the goal of discovering hidden facts contained in the data.

Content: Processes of data mining which include data cleaning, explorations, pattern recognition, machine learning techniques; visual analytics; and reporting. Elementary analyses which include, regression, detection of outliers, contingency table analysis as well as specialized analyses of multivariate techniques.

Applied Binary Classification and Matching (STAT606)

Aim: To introduce statistical techniques that will be needed to classify and match variables/observations in big business data and their application in big data classification and market segmentation.

Content: Similarity and Dissimilarity, Missing data mechanism, Binary classifiers, Classification methods, Statistical matching methods.

Semester 2:

Applied Generalized Linear Model Analysis (STAT601)

Aim: To equip students with the methods of analysis using the models belonging to the class generalized linear models and expose students to a range of practical problems.

Content: The general linear model for quantitative responses (including multiple regression, analysis of variance and analysis of covariance), binomial regression models for binary data (including logistic regression and probit models); Poisson regression models for count data (including log-linear models for contingency tables); Quasi likelihood methods and event history analysis or reliability analysis which covers a set of techniques for modelling the time to an event.

Machine Learning and Predictive Modeling Techniques for Business (STAT605)

Aim: To introduce and demonstrate through hands-on practice how to use different tools to build predictive models and machine learning techniques to predict future events based on the data.

Content: A wide array of topics in data exploration; machine learning and predictive techniques to solve challenging business problems.

Year 2

Semester 1:

Time Series and Forecasting Econometrics (STAT602)

Aim: To provide a thorough understanding of the theory and application of the time series and econometric techniques.

Content: Box-Jenkins methodology, ARMA, ARIMA, and SARIMA models, ARCH and GARCH models, Extreme value theory, linear regression, structural equation modelling, state-space modelling

Applied Longitudinal and Geospatial Analysis (STAT604)

Aim: This course introduces students to methods of longitudinal data analysis and issues involved with spatially sparse data.

Content: The course will be based on multilevel models (also referred to as hierarchical models, mixed effects models, and random coefficient models) with a major emphasis on modelling intra individual effects. General Concepts in Spatial Data Analysis, Analysis of Point Patterns and Analysis of Area Data.

Semester 2:

Industry Project in Data Science (STAT6RP)

Aim: To give students the opportunity to conduct real-world analytics projects using data from business and government organizations.

Content: Students work independently to understand the business problem, and then clean and analyze the data—to achieve the necessary balance given the needs of the project. The final product is a research report.

Entrance Requirements:

The minimum requirement to be considered for the PGDip in Data Science is a 3-year or equivalent Bachelors degree in Commerce, Engineering or Science, with at least two years of industry data management/analytics experience. Evidence of this experience will be requested prior to selections. See more information about this under the Portfolio of Evidence and Selection Process section below.

How to Apply:

Apply online via this link (click on the red ‘CLICK HERE TO APPLY” button at the bottom of the page). Applications usually open around mid-year and close in September. Please remember to upload all documentation that is requested and pay the required application fee. International students can view which documents are required here. The portfolio of evidence is not required to be submitted at this point. Read more about this under the Portfolio of Evidence and Selection Process section below.

Course Code: PGD-DS

PLEASE NOTE: For your application to be considered, the application fee must be paid upon applying. See here for more information on the application fee and banking details. Please take note of the payment reference to be used. Please submit proof of payment of the application fee by submitting a ticket here.

Portfolio of Evidence and Selection Process:

In November, applicants are emailed requesting them to submit their CV, Proof of Qualification and Portfolio of Evidence. They are supplied with a template to outline their industry experience in data management/analytics (this forms the Portfolio of Evidence).

The selection committee reviews the applications during the course of December. Spaces are very limited and a careful selection process is followed. Priority is given to applicants who are able to demonstrate sufficient data management/analytics experience in their job (at least two years of experience is required). Online courses and certificates will not count towards this required industry experience.

Applicants are informed of the outcome in early January. If you applied for the program by the cut-off in September but do not receive the email requesting your documentation by the beginning of December, please contact us. Lectures usually start in the second week of February.

Student Testimonials:

This program has had a significant impact on many of its students. See some of the testimonials from our alumni below.

Ronewa Mabodi

Qualifications:
BSc Eng - Electrical Engineering (Wits): 2018
PGDip Data Science (UKZN): 2023

"I am currently employed as a Data analyst in the banking space. I got the job on the 9th month of enrolling on the PGDip program . I believe it added a lot of weight to my CV as there are only few people with this type of qualification in the country. The value added from this qualification is priceless; from the approaches and considerations leant on why and how to clean data to knowing how to build ML models and properly evaluate them. The program taught me to do my 9-5 job smarter and not harder and yet achieve greater results.

For sure when I was a student, I felt like some modules were unnecessarily harder than others but now I fully understand why it had to be that way because the program laid a great foundation for the Masters of Applied Data Science program that I am doing at University Johannesburg. I do have an added advantage over my peers because of this program.

My research paper that I did through this program was accepted for publication by the South African Institution of Electrical Engineers (SAIEE) which also shows the quality of the UKZN lecturers and the materials they lecture."

Ronewa's publication can be found here (a first for a PGDip student since its inception in 2021).

Narov Pursotham

Qualifications:
Bachelor of Business Science (UKZN): 2017
PGDip Data Science (UKZN): 2023

"I was a product specialist in the financial services industry before I entered into the field of data science. However, I have always had a passion for analytics and jumped at the opportunity to enroll in the PGDIP-DS programme offered at UKZN. I started this programme in February 2022 and quickly noted that the syllabus covered in the programme would assist in achieving my goal of becoming a data scientist. The programme not only covers the basics of the data science field but also goes into more advanced topics e.g. understanding various machine learning models both from a theoretical and practical perspective. These are topics that senior data scientists deal with in the working environment and learning this through the programme is definitely advantageous. In addition to this, the programme exposed me to softwares such as SAS and R studio (which I had no prior experience in) and assisted in developing my skills at an advanced level.

My goal came to fruition when, within less than a year of being enrolled in the programme, I obtained employment as a data scientist. In my current role, I have worked on forecasting and predictive modeling projects. I was able to land the projects because I was equipped with the knowledge and skills from the programme.

It seemed impossible to complete this programme while working a full time job but I managed to do so at the end of 2023 and was recently awarded the top student award for the class of 2022-2023. If you truly have a passion for the data science space, like me, I would highly recommend this programme as it has changed my life for the better."

Graeguy Kalombo

Qualifications:
BTECH Electronic Engineering (UNISA): 2020
PGDip Data Science (UKZN): 2023

"I worked at a FinTech firm as a Software Quality Engineer where part of my duties were to analyze data and provide insights to inform business decisions. However, there was a feeling that a lot more could be done with the vast amount of data that was being stored, hence why I enrolled in the PGD program in order to upskill in the latest data methodologies and tools. The program exceeded my expectations; by the end of the first year my data collection, preparation, analysis and visualization skills improved considerably, and throughout the second year I was able to use Software like SAS and R-Studio to build good Machine Learning models that tapped into the vast amount of unused data stored at the firm and provided even better insights to the business.

Since then, I have landed a similar role with one of the biggest i-gaming content suppliers in the world where the diploma and my practical data driven approach to assessing and improving Software Quality were significant factors in separating me from other candidates.

I would encourage and recommend anyone who works or is interested in working with data to enroll in this program. The part time schedule was perfect for my situation; I was able to complete the program in the minimum amount of time required despite having a full time job and being a parent. It is not easy, but it can be done."

Nontuthuko Ndlovu

Qualifications:
BSc Hons Applied Mathematics, NUST, Zimbabwe: 2002
PGDip Data Science (UKZN): 2022

"As a retail planner, this course has equipped me with updated analytic techniques that are useful in understanding customer purchasing behaviour from retail data and drawing insights to unlock valuable sales opportunities. The course was well paced and the online classes assisted me to efficiently manage my time between lessons and work. I have benefitted greatly and revived my passion for statistics."

Page updated

Google Sites

Report abuse