Computer Science Optional Modules: How to Choose

Most CS students pick optional modules based on what sounds interesting. That's not a bad starting point, but it leaves a lot on the table.

Optional module selection in computer science is one of the biggest levers you have on your final degree class — and unlike your raw ability, it's something you can actually control. But the information you need to make that decision well is almost nowhere to be found.

Why optional modules matter more in CS than most degrees

Computer science degrees at UK universities are modular by design. In most cases, your second and third years are heavily weighted towards electives — sometimes 60-80% of your credits are from optional modules. That means your module choices aren't just shaping what you learn. They're shaping your degree classification.

The other thing that makes CS unusual: grade distributions vary enormously across modules. Machine learning modules at some universities have very high First rates. Systems programming modules at others are notorious for pulling down averages. This isn't random — it reflects assessment style, cohort size, and marking conventions. None of it is published anywhere you can easily find.

The variables worth weighing

When you're choosing CS optional modules, you're really making several decisions at once.

Assessment format. Coursework-heavy modules give you more control over your output. Exam-only modules introduce more variance. If you're a strong coder but struggle under timed pressure, that's worth knowing before you pick a module that's 100% exam.

Module size. Bigger cohorts produce more predictable grade distributions. Small specialist modules (under 30 students) can swing either way — extremely high or extremely low First rates — partly because small sample sizes amplify the effect of a single difficult exam year.

Overlap with your dissertation. Final-year CS students doing a software project or research dissertation often benefit from taking modules that share foundational material. This creates compounding returns: the module teaches the theory, the dissertation lets you apply it, and the marker for each sees evidence of genuine depth.

Career relevance vs. grade optimisation. These aren't always in conflict, but they're not always aligned either. Machine learning, security, and distributed systems modules are in high demand from employers. But if the assessment structure doesn't suit you, forcing a high-demand module can hurt your overall classification. The answer isn't to avoid difficult modules — it's to make an informed trade-off rather than an uninformed one.

What the FOI data shows about CS grades

The grade distribution data GradeHack holds on computer science modules — sourced via FOI requests to UK universities — shows several consistent patterns.

Modules built around coursework and group projects tend to have above-average First rates across institutions. Modules that rely on unseen exams, particularly in theoretical areas like formal methods or computational complexity, tend to have tighter grade distributions and more students in the 2:1 band.

This doesn't mean you should avoid theory modules. It means you should go in with realistic expectations and a revision strategy that matches the assessment format. A student who picks a module because "machine learning is cool" without preparing for its particular exam style is at a structural disadvantage before they've written a word.

A practical framework for CS module selection

Here's how to approach optional module selection systematically, rather than by gut feel:

Step 1: Map your current classification trajectory. Work out where you stand based on marks so far, using how to predict your degree classification as a starting point. This tells you whether you need to protect your grade, chase points, or are comfortable enough to pick for interest.

Step 2: Get the assessment breakdown. Every module specification lists the percentage split between coursework and exams. Read it. If 70% of the marks come from one three-hour exam, decide whether that suits you before committing.

Step 3: Talk to students one year ahead. Third-year students are an underused resource. They've just sat the modules you're considering. Ask them what the marking was actually like, whether the coursework was reasonable, and whether the module matched its description. This is informal grade distribution data — and it's more actionable than a module handbook.

Step 4: Consider your dissertation alignment. If you're a final-year student, choosing your final-year modules covers this in more depth. The short version: modules that share concepts with your dissertation topic are almost always a good bet, because the effort compounds.

Step 5: Look up FOI data where it exists. Historical grade distributions for specific modules at specific universities — where they're publicly available — are the single most useful input you can have. GradeHack makes this data searchable. Access the GradeHack data to see module-level distributions before you commit.

Common mistakes CS students make

Picking every "popular" module. Machine learning, NLP, and computer vision modules are all legitimately useful. They're also picked by everyone, marked competitively, and increasingly difficult to differentiate yourself in. Balance demand with your actual aptitude for the assessment style.

Ignoring systems modules because they're considered "dull". Operating systems, networks, and distributed systems modules often have surprisingly good grade distributions — partly because fewer students pick them for interest, which means the cohort is self-selected for people who actually want to be there.

Stacking all your hardest modules in the same semester. This is a time management problem as much as a selection problem. If you're picking four intensive coursework modules simultaneously, you're creating a scheduling crisis. Spread the load.

Not reading the module specification before the selection deadline. The specification tells you who delivers the module, what the assessment looks like, and what reading is expected. Picking a module based on its title alone is a mistake that's easily avoided.

What to do right now

If you're approaching module selection — whether for second year or final year — the most useful thing you can do is get hold of historical grade data for the specific modules you're considering. Not averages across the degree, not national statistics. The actual distribution for that module at your university.

That's what GradeHack is built for. The data comes from FOI requests filed directly with UK universities — the same information universities hold internally, made searchable.

See what modules you should take at university for the broader decision framework, or how module choice affects your degree class for the full picture on classification impact.

For the data itself — get access here.