UniMelb · MAST20034 · Critical Thinking with Data

MAST20034: pass the exams, not just read the notes

Your complete guide to University of Melbourne's critical thinking with data unit. See where the marks are, work real practice questions, and study with an AI tutor that knows MAST20034.

12.5 credit points Level 2 undergrad Offered S1 / S2 ~60% exams School of Mathematics and Statistics

Sia generates MAST20034 practice questions, walks through frameworks for inference and statistical analysis step by step, and quizzes you on the material the exam weights most heavily.

Which thesis is stronger?

Sharpen your argument

Pick one · the reasoning is revealed after you answer

A student team plans to survey opinions on a new campus policy by handing the questionnaire to the first 40 people they meet outside the library on a Tuesday afternoon. The lecturer flags this as convenience sampling and asks why it is not recommended. Which response is the strongest critique?

Why this one wins

Name the issue: convenience sampling selects whoever is easiest to reach, not a random draw from the target population, so the sample is not designed to be representative.

Identify the mechanism: convenient participants tend to be similar to each other and to the researchers, which induces bias and artificially shrinks the variability you would see in a genuine random sample.
Identify who is missed: people who are not outside the library on a Tuesday afternoon (commuters, online students, people in different faculties or time zones) may differ in exactly the ways that matter for the policy question, and they are systematically excluded.
Match to the options: option C states both the bias-and-low-variability mechanism and the systematic exclusion of non-convenient participants, which is the full critique the marking criteria reward.
Eliminate the rest: A is the misconception that size alone buys representativeness; B repeats it (a bigger convenience sample is still biased, just more precisely biased); D invents a calculation requirement, but this subject has no calculations and the flaw is about how the sample was selected, not arithmetic.

The weaker choice: Believing that a large enough sample fixes a convenience sample. Increasing n improves precision but does nothing about selection bias: a big sample of the wrong people just estimates the wrong quantity more confidently. The fix is how you select (random sampling), not how many you select. watch this!

your whole grade
Where your grade comes from Exams 60% · Assignment 20% · Projects 15% · Quizzes 5%

One exam decides 60% of your grade. Covers the whole subject, emphasising material seen in multiple ways (lectures, tutorials and assessment); marks reward explained reasoning over any single answer. This whole page is built around that.

Overview

What MAST20034 is, and where it sits

MAST20034 Critical Thinking with Data is a second-year School of Mathematics and Statistics subject, but it is unlike most statistics units: there are almost no calculations. Instead it teaches you to read, question and critique how data is produced, presented and used to make claims. The semester moves through objectivity and the framing of data, honest graphics, study design (experimental versus observational), observational studies and confounding, qualitative methods, frameworks for statistical inference, statistical modelling, sampling and WEIRD bias, how research accumulates, and big data and context. The running thread is critical thinking applied to real case studies rather than memorising formulas.

Because the skill being taught is judgement, the subject is built around active learning. Two one-hour lectures a week introduce ideas, two-hour tutorials explore them through discussion and case studies, and the assessment then asks you to apply that thinking. Four short written assignments (each capped at 200 words, APA 7 referencing required), five low-stakes revision quizzes, a group project and a three-hour open-note final exam together reward careful reasoning and concise communication, not arithmetic. The unit is explicit that you may not use generative AI for assessment except to help write graphing code or summarise unrelated articles.

The difficulty here is not technical depth but the craft of argument. Students who are used to statistics meaning equations and calculators sometimes underestimate it, then find the strict 200-word limit and the demand to justify reasoning surprisingly exacting. It pairs naturally with the data-handling and modelling skills from other Mathematics and Statistics subjects and with any degree that involves reading research critically, from science to social science.

How it differs from its first-year siblings. MAST20034 is a statistics subject with the equations removed: it is about critically evaluating how data is generated, displayed and used, not about computing estimates by hand. That makes it complement rather than overlap the quantitative subjects. ECON10004 teaches the economic reasoning and basic data interpretation behind business decisions; BIOL10008 is where many students first meet experimental design and the difference between correlation and causation in a science context; COMP90038 sits on the computing side of working with data. MAST20034 supplies the critical-reading layer that sits on top of all of them.

Official outline: handbook.unimelb.edu.au · MAST20034 outline. Always treat the official outline and the exam timetable as authoritative.

Difficulty & time commitment

Is MAST20034 hard, and how much time does it take?

MAST20034 is manageable if you keep a weekly rhythm and treat the back half as the main event. Across student reviews the pattern is consistent: it starts gently and steepens, and the heaviest assessment is the part that separates grades.

Difficulty
2.8 / 5
Moderate. Gentle early, demanding back half. Hard to fail with steady work; an HD takes consistent practice.
Exam load
60%
The exams decide most of the grade. The heaviest single component is 60%.
Weekly time
~8 hrs
The standard load for a 12.5-credit-point unit, around 1.5 hours per credit point per week including class.
Weeks 1 to 5 (objectivity, graphics, study design, observational studies, claims)gentler
Weeks 6 to 12 (qualitative methods, inference, modelling, sampling, accumulating research, big data)steady but conceptual

The difficulty curve and the assessment weighting point the same way: the back half is harder and worth more. Front-loading effort there is the highest-return decision in the unit.

Is this unit for you

Who tends to do well, and who tends to struggle

You will likely do well if

  • You enjoy arguing from evidence and can hold the line between what data shows and what it merely suggests, especially the difference between correlation and causation.
  • You write tightly: you can make a clear, well-justified point inside a strict 200-word limit without padding.
  • You engage with the weekly case studies and tutorials actively, since the subject is built around discussion and the exam draws on cases seen in multiple ways.
  • You read research and graphics with healthy suspicion, asking who was sampled, how the study was designed and whether the claim is actually supported.

You may struggle if

  • You expect a normal calculation-heavy statistics subject and are uncomfortable when the marks depend on reasoning and writing rather than getting a number right.
  • You write long and find it hard to cut: the 200-word cap is strictly enforced, with zero awarded at 241 words or more.
  • You skip the case-study discussion in tutorials, since much of the exam asks you to critique examples worked through in class.
  • You lean on generative AI to draft answers, which is not permitted for assessment here except for graphing code or summarising unrelated articles, and is easy to detect in critique-style work.
do this ↘
What HD students do differently
  • Treat every claim, study and graph you meet as something to interrogate: name the design, the sample, the confounders and the framing, and practise saying why a claim is or is not supported.
  • Rehearse writing 200-word critiques to the marking criteria, leading with your reasoning, because explained reasoning is worth more than the answer.
  • Build your two A4 note pages around frameworks (study-design checklist, what makes a graphic honest, sampling-bias and confounding prompts, the inference and modelling assumptions), not facts to memorise.
  • Use the practice and previous exam (released late in semester) for timing and style, and act on the feedback from each short assignment before the next one.

Syllabus

The 12 topics, week by week

The exam-weight marker on each topic shows where the marks concentrate. The amber topics carry the highest exam weight.

W1

T1 · Objectivity and data

Module 1

Why data is never fully objective: how framing, choices about what to measure and the context of collection shape what data can and cannot tell us, introduced through a critical lens.

Lower exam weight
W2

T2 · Good graphics

Module 2

What makes a graph honest and effective: titles, axis labels, legends, accessible use of colour and symbols, common ways graphics mislead, and how to critique and improve a figure.

Lower exam weight
W3

T3 · Study design

Module 3

Experimental versus observational designs, randomisation and control, and why design determines which conclusions a study can legitimately support.

Lower exam weight
W4

T4 · Observational studies

Module 4

Confounding and the correlation-is-not-causation problem: how lurking variables distort conclusions in observational data and what can and cannot be inferred without an experiment.

Lower exam weight
W5

T5 · Reporting and critically assessing data-based claims

Module 5

How data-based claims are reported in research and the media, and a toolkit for assessing whether a claim is actually supported by the data behind it.

Lower exam weight
W6

T6 · Qualitative methods

Module 6

What qualitative data and methods offer, how they differ from quantitative approaches, and how to read qualitative evidence critically rather than dismissing it as merely anecdotal.

Lower exam weight
W7

T7 · Frameworks for inference

Module 7

Statistical inference conceptually: estimation, confidence intervals, hypothesis testing and p-values, parametric methods and their assumptions, taught for interpretation rather than hand calculation.

W8

T8 · Statistical analysis and modelling

Module 8

What a statistical model is doing, checking model assumptions (for example through residual and normal-probability plots), and reading software output critically rather than trusting it blindly.

W9

T9 · Sampling

Module 9

How samples are drawn and why it matters: convenience versus random sampling, sampling bias, representativeness and WEIRD bias in who research actually studies.

High exam weightQuiz me on sampling →
W10

T10 · Accumulating scientific knowledge

Module 10

How knowledge builds across studies: replication, meta-analysis, publication bias and the reproducibility crisis, and what it takes for a body of research to be trustworthy.

Lower exam weight
W11

T11 · Big data

Module 11

The promises and pitfalls of large-scale data: why bigger is not automatically better, algorithmic and measurement bias, and the new critical-thinking problems big data creates.

Lower exam weight
W12

T12 · Context revisited

Module 12

Returning to context with the full toolkit: synthesising the semester's ideas to evaluate data and data-based claims in their real-world setting.

Lower exam weight

How it's assessed

Assessment structure

ComponentWeightFormat & timing
Revision quizzes (5)5%Five short Canvas revision quizzes spread across the semester to check understanding of recent material. Quizzes in roughly Weeks 2, 4, 6, 9 and 12 (dates subject to change). Low stakes, designed to keep you current rather than to differentiate marks.
Short assignments (4)20%Four short written tasks, each with a strict 200-word limit, written in full sentences with an in-text reference and an APA 7 reference list; assessed on reasoning and concise communication. Assignments due across the semester (roughly Weeks 3, 5, 8 and 12; dates subject to change). Word-limit penalties apply (up to 220 words marked as usual; 221 to 240 lose 2 communication marks; 241 or more receive zero). Late penalty 1% per hour.
Group project15%Group case-study project including a peer-reviewed presentation and a group contributions component, applying the semester's critical-thinking toolkit to a chosen case study. Group formation around Week 7, presentations around Week 11 (dates subject to change). Includes a group contributions survey and peer review; AI is not permitted except for graphing code or summarising unrelated articles.
Final exam60%Three-hour written exam, short-answer only (no multiple choice, no essay-style questions). Open note: up to two A4 double-sided pages (four sides). No calculator, no calculations required. May include images or statistical output to interpret. Formal University exam period. Covers the whole subject, emphasising material seen in multiple ways (lectures, tutorials and assessment); marks reward explained reasoning over any single answer.
Revision quizzes (5)5%
Five short Canvas revision quizzes spread across the semester to check understanding of recent material.
Short assignments (4)20%
Four short written tasks, each with a strict 200-word limit, written in full sentences with an in-text reference and an APA 7 reference list; assessed on reasoning and concise communication.
Group project15%
Group case-study project including a peer-reviewed presentation and a group contributions component, applying the semester's critical-thinking toolkit to a chosen case study.
Final exam60%
Three-hour written exam, short-answer only (no multiple choice, no essay-style questions). Open note: up to two A4 double-sided pages (four sides). No calculator, no calculations required. May include images or statistical output to interpret.
  • Pass on a weighted average of at least 50%. No single-component hurdle is stated in the unit materials reviewed.
  • Short-answer questions only, in two flavours: explaining reasoning and applying critical thinking to a context, and applying critical thinking to a specific case or graphic covered in a whole-class activity. Explaining your reasoning is typically worth more than the bare answer.
  • Calculator policy: No calculator is permitted in the final exam, and the unit states there are no questions requiring any calculations anywhere in the assessment.
read this! If you read nothing else

This is an exam-cram unit. With the exams at 60% of the grade and the final exam alone at 60%, your result is overwhelmingly decided by how well you perform under time pressure. Covers the whole subject, emphasising material seen in multiple ways (lectures, tutorials and assessment); marks reward explained reasoning over any single answer.

Final exam timing: approx mid-Nov 2026 (estimated S2 2026 window; confirm against the official University of Melbourne exam timetable). Confirm the exact date and venue on the official exam timetable.

How to actually pass it

A weekly rhythm, two checklists, and the traps to avoid

The unit rewards consistency over cramming, and practice over re-reading. Here is the loop that works, then what to have nailed before each exam.

The weekly loop

Before the lectures
Skim the week's module and any case-study materials so the two lectures build on context rather than introducing everything cold.
During the lectures
Engage with the interactive and PollEverywhere parts and note the critical-thinking moves being modelled, not just the conclusions.
In the tutorial
Join the discussion of the previous week's topic and case study fully; the tutorial runs one week behind the lecture, and these whole-class activities are exactly what the exam draws on.
End of each topic
Add a compact framework for that topic (a checklist or prompt set) to your running two-page exam note sheet, and read your assignment feedback before the next task.

Before the mid-semester checklist

  • Keep the five revision quizzes current; they are low stakes but a good early calibration of where you stand.
  • Submit each 200-word short assignment well under the limit, with an in-text reference and an APA 7 reference list, and read the returned feedback carefully.
  • Practise critiquing graphics: identify good features, name a specific weakness and propose a concrete improvement, as in the sample exam question.
  • Lock in the early-semester frameworks (objectivity and framing, honest graphics, experimental versus observational design, confounding) so the later conceptual material has a base.

Before the final heaviest topics

  • Build your two A4 double-sided note pages as a set of reusable frameworks, not crammed facts, since the exam is open note and reasoning-based.
  • Work the released practice exam and the single previous exam under timed conditions late in your preparation, focusing on explaining reasoning to the marking criteria.
  • Be ready to interpret an unfamiliar graphic or piece of statistical output, since the exam may give you one to critique with no calculation required.
  • Revise the sampling, confounding and inference-assumption ideas as concepts you can argue about, including convenience versus random sampling and WEIRD bias.
  • Rehearse referring to whole-class case studies the way the exam allows (for example 'the study about the cash transfers') without needing exact references or recalled details.

The mistakes that cost marks

01

Assuming a bigger sample fixes a biased one. Increasing the sample size improves precision but not representativeness. A large convenience sample just estimates the wrong quantity more confidently. The fix is random selection, not more of the same convenient people, and this confusion is a classic sampling error.

02

Treating it like a calculation subject. There are no calculations and no calculator is allowed. Students who revise formulas instead of frameworks for critique waste their preparation; the marks come from reasoning about how data was produced and whether claims hold.

03

Going over the 200-word limit. The limit is strict: 221 to 240 words loses two communication marks and 241 or more scores zero. Headings, captions and in-text references all count. Padding an answer is directly penalised, so cutting hard is part of the task.

04

Skipping the tutorial case-study discussion. The exam asks you to critique examples covered as whole-class activities in lectures and tutorials. Students who treat tutorials as optional lose access to exactly the material the exam is built on.

Teaching team

Who teaches MAST20034

The bios below are factual. The star ratings are not ours: they are impressions from students who have taken the unit, so you can hear from people who sat in the lectures.

Subject coordinator and lecturer

Paul Fijn

Lecturer in statistics in the School of Mathematics and Statistics, University of Melbourne, who researches statistics education and has taught statistics across agriculture, ecology, biomedicine, psychology and education.

Student ratingNo student ratings yet

Teaching team as listed in the unit materials reviewed. AskSia does not rate lecturers; star ratings are submitted by students who have taken MAST20034.

Where it fits

Prerequisites, related units & why it matters

A second-year (Level 2) Mathematics and Statistics subject. Check the University of Melbourne Handbook entry for the exact prerequisite and prohibition list for your offering, as these can change between years.

Why it matters beyond the grade. MAST20034 installs a transferable skill that almost no employer dislikes: the ability to read research and data-based claims critically, spot weak study designs and misleading graphics, and communicate a judgement concisely. It underpins work in research, science communication, data analysis, policy, journalism and any role where you must decide whether the data actually supports the claim.

FAQ

Frequently asked questions

Is MAST20034 hard?

It is moderate, but in an unusual way. There is almost no maths, no calculations and a calculator is not even allowed, so the technical difficulty is low. The challenge is the craft of critical reasoning and concise writing: the four short assignments cap you at 200 words, and the marks reward the quality of your argument rather than a right answer. Students expecting a normal calculation-heavy statistics subject are often caught out by how exacting the writing is.

How is MAST20034 assessed?

Five low-stakes revision quizzes worth 5% combined, four short written assignments worth 20% combined (each capped at 200 words with APA 7 referencing), a group project worth 15%, and a three-hour open-note final exam worth 60%. You pass on a weighted average of at least 50%, with no single-component hurdle stated in the materials reviewed.

What is the final exam like?

A three-hour written exam of short-answer questions only, with no multiple choice and no essay-style questions. It is open note: you may bring up to two A4 double-sided pages (four sides). No calculator is allowed and no question requires calculations. Some questions give you an image or statistical output to interpret. The time is deliberately generous, and explaining your reasoning matters more than the bare answer.

How much maths is in this subject?

Very little. The subject is about critically evaluating how data is produced, displayed and used, not about computing statistics by hand. You will meet inference, modelling and sampling conceptually (confidence intervals, p-values, assumptions, bias) so you can interpret and critique them, but the exam has no calculations and a calculator is not permitted.

Can I use AI for the assignments?

Only in two narrow cases: to help write code for graphics (for example ggplot in R) and to summarise articles that are not directly related to the subject material. Any other use of AI for assessment is not permitted, and if you do use AI for an allowed purpose you must declare which AI and model you used and every prompt you entered, after your reference list.

Why is there a 200-word limit on the assignments?

Deciding what matters and communicating it concisely is part of the skill being assessed, so each short assignment is capped at 200 words. The count includes headings, captions, in-text references and text inside figures; your name, student number, reference list and any AI declaration do not count. Going over carries real penalties: 221 to 240 words loses two communication marks, and 241 or more is not assessed and scores zero.

Study MAST20034 with Sia

Work through frameworks for inference, statistical analysis, sampling and the rest of the unit with a tutor that knows it and quizzes you on the topics the assessments weight most heavily.

Start studying with Sia