Data science is a new, rapidly expanding field. There is an unprecedented demand from technology companies, financial services, government and not-for-profits for graduates who can effectively analyse data. This subject will help students gain a critical understanding of the strengths and weaknesses of quantitative research, and acquire practical skills using different methods and tools to answer relevant social science questions.
This subject will offer a nuanced combination of real-world applications to data science methodology, bringing an awareness of how to solve actual social problems to the Master of Data Science. We cover topics including elections, criminology, economics and the media. You will clean, process, model and make meaningful visualisations using data from these fields, and test hypotheses to draw inferences about the social world.
Techniques covered range from descriptive statistics and linear and logistic regression, the analysis of data from randomised experiments, model selection for prediction and classification tasks, to the analysis of unstructured text as data, multilevel and geospatial modelling, all using the open source program R. In doing this, not only will we build on the skills you have already mastered through this degree, but explore different ways to use them once you graduate.
Learning outcomes are the key abilities and knowledge that will be assessed in this unit. They are listed according to the course goal supported by each.
1. Students will be well-versed in the various ethical issues and professional standards around the gathering of data.
2. During the unit, students will be required to deliver a small scale group project. Students will be proficient in the delivery of a small-scale project, and the management of the project from initial conception to delivery to evaluation.
3. Upon completion, students will be able to present data and reports of a high standard.
4. Students will be trained in the autonomous collection, collation, assessment and comparison of data from multiple sources, such as the Australian Bureau of Statistics and the Australian Data Archive. Students will be able to discern the quality of data to a minute level, and be able to draw a broad range of insights from data of various degrees of statistical significance.
5. Students will be trained in the sophisticated application of established data analytical methodology. Students will be expected to have a medium degree of proficiency in methodological procedures, and will be tasked with complex problems specifically related to the social sciences.
6. Students will be utilising industry-leading concepts and frameworks in their pedagogy. Students will be directing formidable amounts of data for protracted, complex insights into areas such as polling data and demography.
7. Students will be expected to apply their theoretical understanding of statistical methods to practical problems around data gathering methodology, statistical significance and sample sizing. Students will be expected to autonomously create basic design frameworks for statistical modelling problems.
Taught in Semester 1, 2019
1 x 1 hour lecture per week
1 x 2 hour lab per week
Taught in December Intensive, 2019
3 days per week x 2 weeks
2 x 1 hour lecture per day
2 x 2 hour tutorial per day
10% 3 x class test
30% group work
10% research plan
30% research project
Visit the University of Sydney website for information about fees, cross-institutional and non-award study, and more.