About this unit
How do the choices made in representing and modelling data affect whether a claim drawn from it is valid?
Sections in this unit
Factual
- What does a box plot show that a mean does not?
- How is the correlation coefficient found, and what does its sign tell you?
- Which measure of centre is most affected by an outlier?
Conceptual
- What is lost when many values are summarised as one?
- Why can two data sets with identical statistics look completely different?
- When does a line of best fit stop being a reasonable basis for a prediction?
Debatable
- Can a statistic be true and misleading at the same time?
- Who is responsible for a misleading claim — the person who made the graph, or the person who shared it?
Skills
This is a joint interdisciplinary unit with English, assessed by a social media task, so the first three skills below are shared with that subject.
Students should be able to:
- Evaluate a statistical claim encountered on social media by reconstructing what was done to the data before judging what is asserted about it.
- Distinguish what a data set shows from what a post claims it shows, including where a correlation is presented as a cause or a prediction is pushed beyond the data.
- Make a data-based claim of one’s own that is defensible both statistically and in the way it is worded.
- Move between representations: read the same data as a table, a graph, a summary statistic and a sentence.
- Judge the reasonableness of an answer, including its accuracy and its units.
- Interpret a mathematical result in the context it came from, in words a non-mathematician would follow.
- Use technology deliberately — and know what the machine has done, not only which keys produced the number.
How you are assessed
- Social media task with English, MYP interdisciplinary criteria
- Mathematics task on a real data set
Conventions
- Population standard deviation, Tukey quartiles, outlier fences at one and a half interquartile ranges.
Before you start
Students arrive with:
- Mean, median, mode and range for ungrouped data, and the mean from a simple frequency table.
- Bar charts, pie charts and simple scatter graphs, with correlation described informally.
- Percentages and proportion, and reading values from a graph.
- From Unit 4, the equation of a straight line and the meaning of gradient and intercept, which the regression line depends on entirely.
Check what you remember
Mode, median, range from listed data