Cleaning Data
Master Cleaning Data for GCSE Statistics with this free worksheet and full mark scheme — Foundation and Higher exam-style questions with worked answers for AQA and Edexcel. Cleaning data means dealing with errors, missing values and outliers before analysis.
Free downloads
These worksheets and mark schemes are original, written for Virtus Academy and checked against the current AQA and Edexcel specifications. Every worksheet comes with a full mark scheme.
Topic overview
Raw data often contains errors that must be identified and dealt with before analysis. This process is called cleaning the data.
Three problems must be recognised. Outliers are values far outside the pattern of the rest of the data. Errors include impossible values, such as a negative height, and transcription mistakes where a digit has been mistyped. Missing data occurs where a value was not recorded.
How each is handled matters. An outlier caused by a genuine error should be removed or corrected. An outlier that is a real value should usually be kept, because removing it distorts the data. Missing values may be excluded from that calculation, but the reason for the omission should always be stated.
Revision notes
Outliers
An outlier is a value that lies far outside the pattern of the rest of the data.
It may be a genuine extreme value or the result of an error. Deciding which is the key judgement, and questions frequently ask you to justify your decision.
Errors
Impossible values, such as a negative height or an age of 200.
Transcription errors, where a digit has been mistyped — 175 cm entered as 1750 cm. These should be corrected if the true value is known, or removed if it is not.
Missing data
A value that was not recorded, perhaps because a respondent skipped a question.
It may be excluded from that particular calculation, but the reason should be stated. If many values are missing, the remaining data may no longer be representative.
Key points
- Cleaning removes errors before analysis.
- An outlier lies far outside the pattern.
- Impossible values must be corrected or removed.
- Transcription errors mistype a value.
- Genuine outliers should usually be kept.
- State the reason for excluding any value.
Worked examples
Example 1
A data set of heights in centimetres contains the value 1750. State what this is and how it should be dealt with. [2 marks]
Working
Example 2
Explain why a genuine outlier should usually be kept in the data. [2 marks]
Working
Example 3
Explain why the reason for excluding a value should always be stated. [2 marks]
Working
Common mistakes
Removing every outlier automatically.
Genuine extreme values should usually be kept; only errors should be removed.
Not stating why a value was excluded.
The reason must always be given.
Treating missing data as zero.
Missing means not recorded, which is different from a value of zero.
Failing to spot an impossible value.
Check whether each value could genuinely occur.
Exam tips
- Decide whether an outlier is genuine or an error before acting.
- Always state the reason for excluding a value.
- Look for impossible values as a first check.
- Never treat missing data as zero.
Key terms
- Cleaning
- Identifying and dealing with errors in raw data.
- Outlier
- A value far outside the pattern of the rest.
- Transcription error
- A value mistyped when recorded.
- Missing data
- A value that was not recorded.
Related topics
Written and reviewed against the current AQA and Edexcel specifications. Spotted an error? Let us know.