Data Presentation and Interpretation โ OCR A-Level Mathematics
Data Presentation and Interpretation explained
An outlier is a value that lies well away from the main body of a data set.
Read the full explanation
You recognise one by comparing it with the rest of the data or with a stated rule, such as lying more than 1.5 times the interquartile range below the lower quartile or above the upper quartile. On a box plot an outlier appears as a separate plotted point beyond a whisker; on a scatter diagram it is a point far from the general trend. Interpretation matters: an outlier may be a recording or measurement error, or a genuine but unusual observation, and it can distort a mean or a line of best fit. Decide whether to keep, correct or exclude it and justify the choice. In a multiple-choice question, select the option that both identifies the value and gives a sound interpretation rather than merely spotting it.
Your focus
- Identify a possible outlier using a stated rule or visual comparison.
- Interpret an outlier as an error or a genuine value and describe its effect.
- Read outliers correctly from statistical diagrams such as box plots.
Data Presentation and Interpretation exam tips
Marking Points
- Identifies the value that lies far from the main cluster of the data.
- Applies a stated rule such as the 1.5 ร interquartile range boundary to justify the label.
- Reads an outlier correctly from a box plot as a point beyond the whisker.
- Interprets the outlier as a possible error or a genuine unusual value and states the effect on summary statistics.
- Chooses the option that combines correct identification with a sound interpretation.
Examiner Tips
- ๐กQuote the rule or boundary you used before naming the outlier.
- ๐กCheck each option for both identification and interpretation before choosing.
- ๐กMention the effect on a summary statistic to strengthen the interpretation.
Common Mistakes
- Labels the largest value as an outlier without checking it against the spread; compare it with a stated rule instead.
- Assumes every outlier is an error; consider that it may be a genuine observation and say so.
- Ignores the effect of an outlier on the mean; note that the mean is pulled towards it while the median is more resistant.