COMP326 — Data Science Foundations

1Course information

2Course description

COMP326 introduces the principles, methods, tools and practices of modern data science. Students learn how data is collected, cleaned, explored, analysed, visualised and communicated. Emphasis is placed on practical data analysis using Python and contemporary data science workflows.

The course develops the foundational competencies needed for machine learning, artificial intelligence, business analytics and evidence-based decision-making. It follows the full data science lifecycle: acquiring data, preparing it for analysis, exploring it statistically and visually, and communicating what it means to a real audience.

COMP326 deliberately does not cover machine learning algorithms. Statistical regression is treated as a descriptive and inferential technique, not a predictive modelling method. Machine learning is introduced in COMP328.

3Learning outcomes

Upon successful completion of COMP326, students will be able to:

4Prerequisites and assumed knowledge

5Teaching and learning approach

COMP326 is built around doing, not watching. Every unit pairs conceptual tutorials with hands-on work in Jupyter notebooks using authentic datasets. You will write code from the first week.

1. Study the concept

   Each tutorial opens with objectives and a plain-language explanation, illustrated with worked examples.

2. Run the code

   Follow along in Jupyter using provided notebooks and real, messy datasets.

3. Practise and apply

   Complete guided lab activities, then apply the same techniques in graded assignments.

4. Check your understanding

   Finish each unit with self-check questions, a unit quiz and a short reflective summary.

6Schedule

The course runs across six modules and eighteen units over fourteen teaching weeks. Weeks containing two units are indicated. The schedule is indicative and may be adjusted during the semester; any changes will be announced on the course site.

Module structure

7Assessment

Assessment is weighted towards practical work. Five assignments build progressively towards the capstone project, with quizzes and exams confirming conceptual understanding.

Assignment details

Assignment 1 — Data Collection Project

Collect and document an original dataset. Deliverables: the dataset itself, a metadata record describing its structure, provenance and limitations, and a collection report explaining your sourcing decisions.

Assignment 2 — Data Cleaning and Preparation

Improve the quality of a supplied dataset. Deliverables: a Python notebook performing the cleaning, a written cleaning report justifying each decision, and a quality assessment documenting what remains unresolved.

Assignment 3 — Exploratory Data Analysis

Discover and interpret patterns in a real-world dataset. Deliverables: an EDA notebook, a set of visualisations, and an interpretation report distinguishing what the data shows from what it does not.

Assignment 4 — Statistical Analysis Project

Conduct hypothesis testing and statistical analysis. Deliverables: an analysis notebook, a statistical report explaining method and assumptions, and evidence-based recommendations.

Assignment 5 — Data Visualisation Dashboard

Build a professional dashboard. Deliverables: the working dashboard, a design rationale explaining your chart and layout choices, and a presentation of the dashboard to an audience.

Midterm examination

A 90-minute invigilated examination covering Modules 1–3: data literacy, data preparation and exploratory data analysis. The examination includes short-answer questions, interpretation of output, and short analytical problems. A formula sheet is provided. Held in Week 8.

Final examination

A 2-hour invigilated examination covering all six modules, with an emphasis on interpretation and judgement rather than recall. You will be asked to read output, identify problems with analyses, and justify methodological choices. A formula sheet is provided. Held in the examination period.

Unit quizzes

Each of the 18 units concludes with a short online quiz. Quizzes are formative in design, have unlimited attempts, and contribute 10% of the final grade. Their purpose is to confirm that you have understood a unit before moving on, not to rank you against your peers.

Capstone project

An end-to-end data science project in which you complete the entire lifecycle on a problem and dataset of your choosing. The capstone carries 25% of the final grade and is the single largest component. Deliverables are a project proposal, dataset documentation, an analysis notebook, a visualisation package, a final written report, and a presentation.

8Grading scale

Grade boundaries are indicative and may be subject to moderation by the examination board. Final grades are released through the official results system, not through the course site.

9Assessment policies

Late submission

Assignments submitted after the published deadline without an approved extension will incur a penalty of 10% of the available mark per calendar day, up to a maximum of five days. Work submitted more than five days late will receive zero unless special consideration has been granted.

Extensions

Extensions of up to seven days may be requested before the deadline via the online extension form. Extensions are granted for documented circumstances such as illness, family emergency or significant personal disruption. Extensions are not granted for workload pressure or competing deadlines.

Special consideration

If you are unable to complete an assessment due to serious circumstances, apply for special consideration through the central student portal. If granted, you may be offered an alternative assessment, an extended deadline or a deferred examination, depending on the component.

Academic integrity

All submitted work must be your own. This means the code you write, the analysis you perform and the conclusions you draw must be your own work. You may discuss concepts with classmates and use published references, but you must acknowledge them. Submitting another person's work, or reusing work submitted for another course, is academic misconduct.

Referencing

Use a consistent referencing style throughout. The default for this course is APA. Cite all datasets, libraries, documentation and external sources you use, including the source of any code adapted from documentation or tutorials.

Resubmission

There is no automatic resubmission of assignments. If you fail the course overall, a supplementary assessment may be offered at the discretion of the examination board, normally taking the form of a single consolidated assessment task covering all learning outcomes.

Return of marks and feedback

Assignments are returned within two teaching weeks of the deadline, with written feedback against the published criteria. Marks for the final examination and the capstone are released through the official results system.

10Use of AI tools

Generative AI tools are part of contemporary data science practice, and this course does not prohibit their use. However, COMP326 assesses your judgement, not your ability to obtain output. The policy below reflects that distinction.

11Learning resources

Required

All required material is provided through the course site: 153 tutorial pages, Jupyter notebooks, datasets, lab exercises and quiz questions. There is no required textbook to purchase.

Recommended reading

Documentation and online references

• pandas documentation — pandas.pydata.org/docs
• NumPy documentation — numpy.org/doc
• Matplotlib documentation — matplotlib.org/stable
• Seaborn documentation — seaborn.pydata.org
• Jupyter documentation — jupyter.org/documentation

Datasets

A curated set of datasets is provided for laboratory work and assignment starting points. For assignments and the capstone you are encouraged to source your own data. Approved sources include national statistics agencies, open government portals and recognised research repositories. All sources must be documented in your submission.

12Software and tools

All software used in COMP326 is free and open source. You can install it on your own machine and continue using it after the course ends.

13Support and accessibility

Academic support

• Office hours — Tuesdays 14:00–16:00 and Thursdays 10:00–11:00. Drop in without an appointment.
• Laboratory demonstrators — available during all laboratory sessions for one-to-one help with code and concepts.
• Discussion forum — monitored daily on the course site. Ask questions publicly so others benefit from the answer.
• Email — for private matters. Allow two working days for a response.

Study skills support

• Academic writing and referencing support from the university learning centre.
• Numeracy and statistics drop-in sessions.
• Peer study groups organised through the course site.

Accessibility and inclusion

The course is designed to be accessible. All tutorial pages use semantic HTML with proper heading structure, keyboard-navigable components and sufficient colour contrast. Figures and diagrams include alternative text. Code examples are provided as text, not images, so they can be read by screen readers and copied.

Wellbeing

If you are struggling, please reach out early. The university counselling service is free and confidential, and academic staff can adjust deadlines where circumstances warrant it. It is always easier to help before a deadline than after it.

14Communication and expectations

Where to go for what

15Version history