ABS 2024 dataset: Rating 3 — government statistical authority, up to date, transparent methods ▶ View Solution
Social media comment: Rating 1 — no source cited, no evidence of research, anyone can post anything ▶ View Solution
Peer-reviewed journal article: Rating 3 — reviewed by other experts before publishing; researchers have accountability ▶ View Solution
Company’s own website: Rating 1 — financial interest in claims; no independent verification ▶ View Solution
Government health report 2019: Rating 2 — generally credible but may not be current; check for newer data ▶ View Solution
Does the Data Support the Claim?
Art highest score: Supports — Art (84) is the highest value in the table ▶ View Solution
Year 7 across Australia: Goes beyond the data — the table covers only one school; cannot generalise to all of Australia ▶ View Solution
Maths higher than History: Supports — Maths (72) > History (61) ▶ View Solution
Students dislike History because boring: Does not support — data shows scores, not attitudes; low scores don’t tell us why ▶ View Solution
More than half above 70: Supports — 4 of 6 subjects have averages above 70 ▶ View Solution
Statistical Inquiry with Secondary Data
Statistical question: “How has the median household income in Australia changed over the past 20 years?” ▶ View Solution
ABS dataset: ABS Survey of Income and Housing — records household income across Australia; conducted regularly so trend data is available ▶ View Solution
Two limitations: 1. Data may not match the exact year or population being studied. 2. ABS definition of “household” may differ from the measure required ▶ View Solution
Confounding factor: Inflation — if income grew 10% but prices grew 15%, people are worse off; comparing dollar amounts without adjusting for inflation is misleading ▶ View Solution
Which half had greater rainfall: Jan–Jun (390 mm) vs Jul–Dec (306 mm) — Jan–Jun had greater rainfall ▶ View Solution
Seasonal trend: Wet summer/autumn (Jan–Mar); dry winter (Jun–Aug); rainfall increases again through spring into summer ▶ View Solution
Correlation or Causation?
Breakfast and test scores: Correlation (possibly causation) — relationship exists but other factors (sleep, home environment) may be involved ▶ View Solution
Ice cream and shark attacks: Spurious correlation — both caused by hot weather; ice cream does not cause shark attacks ▶ View Solution
Libraries and literacy: Correlation — cities may have higher education investment overall; libraries not the sole cause ▶ View Solution
Light switch and light: Causation — switch directly completes the electrical circuit ▶ View Solution
Chocolate and Nobel Prizes: Spurious correlation — wealthier countries tend to have both; chocolate doesn’t cause genius ▶ View Solution
Exercise and heart disease: Likely causation — supported by medical research; diet and other habits also play a role ▶ View Solution
Limitations of Secondary Data
2010 census for today: Data is 15+ years old; population and living conditions have changed significantly ▶ View Solution
Mining industry wage data: Mining workers not representative; industry pays above-average wages, overestimating typical wages ▶ View Solution
US study for Australian teenagers: Cultural, social, and economic differences mean results may not transfer to Australia ▶ View Solution
City weather stations for rural Australia: Urban heat island effect; city data may not reflect rural or remote conditions ▶ View Solution
Limitation vs bias: Limitation restricts scope of conclusions; bias is a systematic error that makes conclusions inaccurate ▶ View Solution
Screen Time Table
Highest screen time: 16–18 age group — 41 hours per week ▶ View Solution
Lowest screen time: 6–9 age group — 14 hours per week ▶ View Solution
Increase from 6–9 to 13–15: 21 hours per week ▶ View Solution
Average across all groups: 30 hours per week ▶ View Solution
Does it always increase with age?: No — drops from 41 to 38 for the 19–25 age group; claim is not fully supported ▶ View Solution
One limitation: Single study in one location; sample size and methodology not given ▶ View Solution
Extended Analysis — Height and Shoe Size
Trend: As height increases, shoe size increases — from 5.2 (140–149 cm) to 9.1 (170–179 cm) ▶ View Solution
Correlation, causation, or both?: Correlation; likely causal (overall body growth drives both) but this table alone doesn’t prove causation ▶ View Solution
Two limitations: 1. Small sample (50 students). 2. Grouped data hides individual variation within each height range ▶ View Solution
Evaluate the conclusion: “Proves” overstates it; better: “suggests a positive relationship consistent with overall growth patterns” ▶ View Solution
Additional information needed: Whether both genders were included; whether students were randomly selected from multiple schools ▶ View Solution