Interpreting Data
Section: Statistics | Syllabus: Cambridge Lower Secondary Checkpoint Mathematics (0862)
Identifying Patterns, Trends, and Relationships
- Trend: an overall direction in data over time (increasing, decreasing, staying constant).
- Pattern: a repeating or notable structure in the data, e.g. seasonal peaks.
- Relationship: how two variables relate to each other, e.g. correlation between two data sets.
-
Question: A shop's monthly ice cream sales are highest in June, July, and August, and lowest in December, January, and February. Identify the pattern.
- Answer: A seasonal pattern - sales are higher in summer months and lower in winter months, likely linked to temperature
Making Informal Inferences and Generalisations
An inference is a reasonable conclusion drawn from data, without claiming certainty. Generalisations should be made cautiously, considering sample size and how representative the data is.
-
Question: A survey of 20 students at one school found 15 prefer online learning. Is it reasonable to conclude "most students everywhere prefer online learning"?
- Answer: Not fully reasonable - the sample is small (20 students) and from only one school, so it may not represent students everywhere. A more careful conclusion: "In this sample, most students preferred online learning, but a larger, broader survey would be needed to generalise this."
Spotting Misleading Graphs and Statistics
The same data can look very different depending on where the axis starts
- Truncated axis: a vertical axis that doesn't start at zero, exaggerating small differences.
- Inconsistent scales: unevenly spaced axis labels, distorting how big changes appear.
- Cherry-picked data: showing only a favourable time period or subset to support a conclusion.
- Misleading pictograms: using differently-sized images instead of a consistent icon count.
-
Question: A company's bar chart shows sales rising from 98,000 to 102,000, but the y-axis only goes from 95,000 to 105,000. Explain why this could be misleading.
- The actual increase is about 4% (98,000 → 102,000).
- Answer: Because the y-axis doesn't start at zero, this modest increase looks like a huge jump, exaggerating the growth visually.
Answering Statistical Questions Using Data
Support a conclusion with specific evidence from the data - numbers, trends, comparisons - rather than a vague statement.
-
Question: "Did Team A or Team B score more consistently across the season?" Team A: mean 2.1 goals, range 4. Team B: mean 2.0 goals, range 1.
- The means are very similar (2.1 vs 2.0 goals).
- Team B's range (1) is much smaller than Team A's (4), showing far less variation game-to-game.
- Answer: Team B scored more consistently
Real-World Applications
Interpreting data critically matters across many fields:
- News and media literacy: critically evaluating graphs and statistics in articles.
- Advertising: recognising misleading marketing statistics.
- Scientific reporting: making appropriately cautious conclusions from study data.
- Business decision-making: using data trends to inform strategy.
- Public policy: interpreting statistics responsibly to avoid misleading the public.
Exam Tips
Common Mistakes
MistakeOvergeneralising from a small or unrepresentative sample to a much larger population
Fixgeneralise cautiously, and note the limitations of sample size and representativeness
MistakeNot noticing a truncated (non-zero) axis that exaggerates a small difference
Fixalways check where the axis starts before judging how big a difference looks
MistakeTreating a correlation as proof of cause and effect
Fixa relationship between two variables doesn't prove one causes the other - correlation is not causation
MistakeIgnoring the range or spread when comparing averages
Fixalways check the spread (range) alongside the average when comparing two data sets
MistakeAccepting a statistic without checking if it's based on a fair, representative sample
Fixask about sample size, sampling method, and possible bias before trusting a claim
For Exams
- Use specific numbers from the data to support your conclusions, not vague statements.
- Check for misleading features: truncated axes, inconsistent scales, cherry-picked time periods.
- Be cautious with generalisations - note sample size and representativeness.
- Remember: correlation does not imply causation.
- Compare BOTH an average and a measure of spread when interpreting two data sets.
Interactive revision notes, videos and practice questions load below.