Interpreting statistical data
In a minute
Tables and charts only help if you read them carefully and say what they mean. You'll read values from tables and diagrams, compare two sets of data using an average and a , and spot when a conclusion goes further than the data allow.
You'll be able to
- Read values from tables, bar charts, pictograms and stem-and-leaf diagrams, using the key and the scale.
- Draw a sensible conclusion from a table or diagram, using proportions when groups are different sizes.
- Compare two sets of data with one comment on an average and one on the range, in context.
- Compare two sets of data using the median and the interquartile range.
- Explain why a conclusion may not be valid: small or biased samples, misleading scales and hidden individuals.
Before you start2 quick questions on what this builds on
Revise first: 1.13 Percentages and 9.1 Classifying statistical data
Reading tables and diagrams
Before you read any number, check what is being counted. Read the title, the labels on each axis or row, and any key.
On a chart's scale, work out what one small square is worth. Divide the gap between two labels by the number of squares between them.
In a , each number belongs to one row and one column. Row and column totals help you check your reading.
Try it first, then reveal one step at a time.
- Total for Year 11: students.
- Sandwich as a percentage of Year 11: .
- For hot meals, the year groups are different sizes, so compare proportions, not counts. Year 10 has students.
- Year 10: . Year 11: .
- So a Year 10 student is more likely to choose a hot meal.
Try it first, then reveal one step at a time.
- Find the scale. The labels go up in s, with small squares between them. So one small square is books.
- Read each bar: Monday , Tuesday , Wednesday , Thursday , Friday .
- Total for the week: books.
- Friday's fraction: . Divide top and bottom by to get .
- An inference you can draw: Friday was the busiest day this week. The chart alone can't tell you whether that happens every week.
A splits each value into a stem (the first digits) and a leaf (the last digit). The key tells you how to put them back together.
Comparing two sets of data
Two sets of data can differ in two ways. One set can be bigger in general, or one set can be more spread out.
An , such as the or the , tells you which set is bigger in general. A , such as the , tells you which set is more consistent.
A smaller range means the values are closer together, so they are more consistent. You learn to calculate these measures in 9.3.
A good comment does three things. It names the sets, compares the two numbers, and says what that means in the question's context.
Try it first, then reveal one step at a time.
- Means: minutes. Route A has the higher mean.
- Comment 1, on the average: Route A's buses are later on average, by minutes.
- Ranges: minutes. Route B has the larger range.
- Comment 2, on the spread: Route B's lateness varies more, so it is less predictable than Route A.
- Notice that each route wins on one comment. The mean and the range answer different questions.
Try it first, then reveal one step at a time.
- Put feed Q's heights in order: .
- With values, the middle one is the th: cm.
- Range cm.
- Average: feed Q's median is higher ( against ), so feed Q's seedlings are taller on average.
- Spread: feed P's range is smaller ( against ), so feed P's seedlings grow more evenly. Median cm, range cm.
Comparing with the median and interquartile range Extended only
The range uses only the two most extreme values. One unusual value can make it huge.
The (IQR) is the spread of the middle half of the data. It is the upper minus the lower quartile, so one extreme value hardly affects it. You learn to find quartiles in 9.3.
So the Extended way to compare is a median comment and an IQR comment. As before, say what each means in context.
Try it first, then reveal one step at a time.
- Range .
- The lower half is . Its median is the lower quartile: .
- The upper half is . Its median is the upper quartile: .
- IQR .
- Seven of the eight values lie between and . The range is large only because of the , so the IQR describes most of the data better.
Try it first, then reveal one step at a time.
- Group A: IQR ms.
- Group B: IQR ms.
- Median: Group A's is lower ( against ). A lower time is faster, so Group A reacted faster on average.
- IQR: Group A's is smaller ( against ), so Group A's times were more consistent.
What the data can't tell you
Data describe only the people or things that were measured, at the time they were measured. Before you trust a conclusion, ask who was asked, how many, and how the results are shown.
1. A small sample may not be typical. A few people can give very different results from a large group.
2. A sample from one place or one type of person may be biased. Asking only gym members about exercise won't tell you about a whole town.
3. An average describes the group, not each member. A class mean of doesn't tell you that any student actually scored .
4. A chart can mislead. If the vertical axis doesn't start at , small differences look big.
Try it first, then reveal one step at a time.
- Percentage: .
- The sample is small: only people.
- The sample is biased: she asked only her own customers, who already like the café. People who never visit weren't asked.
- So describes these customers. It can't tell you what most people in the town want.
Try it first, then reveal one step at a time.
- Read the bars: Shop X sold and Shop Y sold .
- The axis starts at , so the bars drawn are and units tall. Y's bar looks times as tall.
- Actual comparison:
- Shop Y sold only about times as many phones. The cut axis makes a small difference look large.
Common mistakes
- Subtracting stems or leaves on their own in a stem-and-leaf diagram. Use the key to write each value in full, then calculate.
- Using the mean to say which set varies more. The mean compares averages. Use the range to compare spread.
- Writing only "the mean of A is bigger". Name the set and say what it means, such as "type A is heavier on average".
- Making two comments of the same kind. Make one about an average and one about the spread.
- Giving the height of the tallest bar as the mode. The mode is the category under the tallest bar.
Exam tips
- For "make two comments comparing", write one about an average and one about the spread. Quote both values each time.
- When asked to give a reason, name the statistic and both values, such as "Route B, because its range is larger ( against )".
- Say what a number means in context: for times, a lower average is faster. For marks, a higher average is better.
Key takeaways
Practise this
18 past-paper questions on 9.2 in the 0580 Extended bank, each with a mark scheme and a worked explanation.