IB Maths AI 4.8 Notes

Data collection methods

A data collection method should be designed so that it gives useful, relevant, and trustworthy data. When designing a survey or questionnaire, questions should be clear, precise, and easy to understand. Ambiguous wording should be avoided, since different people may interpret the same question differently.

For example, the question 'Do you exercise often?' is vague because the word 'often' can mean different things to different people. A better question would be 'How many times per week do you exercise for at least $30$ minutes?'

Questions should also avoid leading language. A leading question pushes the respondent towards a particular answer.

For example, 'How much do you agree that school lunches should be improved?' suggests that improvement is needed. A less biased version would be 'How would you rate the quality of school lunches?'

There are thus two types of data collection methods: biased and unbiased.

• A biased method is one that systematically favours certain outcomes. For example, surveying only students in a sports club about exercise habits would give biased results if the aim is to represent the whole school.
• An unbiased method gives every relevant individual or response an equal and fair chance of being represented. Bias can come from:
  • Question wording
  • Poor sampling
  • Missing groups
  • Inconsistent answer options