MAST20034 · Critical Thinking With Data
Objectivity and the Critical Lens
Week 1 sets the founding move of the whole subject: data are never neutral. Someone chose what to measure, who to include, and how to record it — so every dataset arrives with a point of view, and your job is to interrogate it rather than take it at face value. This chapter draws the distinction the markers reward all semester: critique (a structured, even-handed appraisal of what is good and what is wrong) versus criticism (merely finding fault). It introduces the engine you will walk on almost every exam prompt — the PPDAC investigation cycle (Problem → Plan → Data → Analysis → Conclusion) — and the who / why / what / how / when context lens you open every answer with. You also meet the variable-type ladder (nominal, ordinal, discrete, continuous), because the type of a variable silently dictates which graphic, which summary and which analysis is even legal. Get this chapter right and you have the frame that holds the other ten together.
What this chapter covers
- 011.1 What “critical thinking with data” actually is
- 021.2 Critique vs criticism — the founding distinction
- 031.3 The “objectivity” myth and the problem of missing data
- 041.4 The PPDAC investigation cycle (Problem → Plan → Data → Analysis → Conclusion)
- 051.5 Variable types — the nominal / ordinal / discrete / continuous ladder
- 061.6 The who / why / what / how / when context lens
The context lens — opening a critique in short-answer form
- +1Name the move: apply the who / why / what / how / when context lens — a claimed statistic is only as trustworthy as the data behind it, so open by interrogating where the number came from.
- +1Question 1 (who): who is in the 89%? If only users who kept using the app were surveyed, the dissatisfied ones have already dropped out — survivorship / self-selection inflates the figure.
- +1Question 2 (how): how was ‘less stressed’ measured? A leading single-item self-report with no baseline and no control group cannot separate the app's effect from time, placebo or regression to the mean.
- +1State the consequence: until both are answered, the 89% is an uninterpretable marketing number, not evidence the app reduces stress.
Key terms
- Critique vs criticism
- Criticism finds fault; critique is a structured, even-handed appraisal that names what is good AND what is wrong AND how to fix it. MAST20034 marks critique — a one-sided takedown leaves marks on the table.
- The objectivity myth
- The idea that data ‘speak for themselves’. They do not: choices about what to measure, whom to include and how to record encode a viewpoint, so apparent objectivity is always the product of human decisions you must surface.
- PPDAC cycle
- Problem → Plan → Data → Analysis → Conclusion: the investigation cycle that frames almost every critique. Locate which node a prompt is really about and you have your answer's structure.
- Variable type
- Whether a variable is nominal, ordinal, discrete or continuous — the property that silently determines which graphic, summary statistic and analysis are legitimate. Misreading the type is the root of many wrong-tool errors.
- Missing data
- The records that never made it into the dataset — the people not surveyed, the events not logged. Because they are invisible, they are the easiest bias to overlook and the first thing a critical reader asks about.
Objectivity and the Critical Lens FAQ
What does ‘critical thinking with data’ mean if there's no calculation?
It means appraising how data were produced and what they can legitimately support — naming the design, the biases, the confounders and the limits of a conclusion. The skill is structured judgement in words, which is exactly why the exam has no calculator.
What's the difference between critique and criticism in the exam?
Criticism just says ‘this is bad’. Critique names a specific strength, a specific flaw, and a specific fix, each justified — the four-part shape the markers reward. Aim to be even-handed, not negative.
Why does the variable type matter so much?
Because it dictates the legal tools: a bar chart suits nominal data, a histogram suits continuous data; a mean suits symmetric continuous data, a mode suits nominal. Pick a tool that mismatches the type and you lose marks before you even reason.
Exam move
Make the context lens and the PPDAC cycle the first two things on your notes sheet — they open and structure almost every answer. Practise turning any claim into two or three sharp context questions (who/why/what/how/when), each tied to the bias it guards against. Drill the critique-not-criticism habit: for every example, force yourself to name one genuine strength before the flaw. Memorise the variable-type ladder as a quick legality check on whatever graphic or summary a prompt shows you.
