section 2

• what is a closed question in a questionnaire? 


  they have a fixed number of answers
• what are pros of closed questions?


  the data can be easily processed

  no ambiguous answers
• what are the cons of closed questions?


  the answers are limited
• what are open questions in questionnaires?


  they allow any answers
• what are pros of open questions?


  people can say exactly what they think
• what are cons of open questions?


  the data may be hard to process as there may be a wide range of answers
• how to design questionnaires carefully

  1. questions are relevant
  2. clear, brief, and easy to understand questions
  3. allow all possible answers
  4. no leading or biased questions
  5. unambiguous questions
  6. people may not answer truthfully
• what are pros of questionnaires?


  quick and cheap to make

  shouldn't be biased

  responses are likely to be truthful

  lots of ways to distribute to large numbers
• what are cons of questionnaires?


  distribution can lead to bias (depending on method)

  easy to ignore so lots of non-responses

  may not be understood
• what are methods for distributing questionnaires?

  1. hand it out - target population gets the questionnaire but time consuming
  2. put it online - data recorded and collected easily but people may not have access
  3. send it by post/email - wide reaching but not sure who responds
  4. ask people to collect it - easy but people with stronger views are more likely to take one
• how to improve response rate


  use clear and easy to understand questions

  follow up people who don't respond

  provide an incentive to respond

  collect the questionnaires yourself
• how to deal with incomplete responses

  1. discard all responses that are incomplete - ok when there are lots of responses but may lead to bias
  2. follow up incomplete answers
  3. replace missing answers with guesses
  4. analyse to find the cause
• what is a pilot study?


  when the questionnaire is tested on a small group of people first
• what are pros of a pilot study?


  spot any unclear or ambiguous questions

  may be an unexpected outcome

  easier and cheaper to fix problems

  check that the methods of distribution and collection work
• how to get accurate data from sensitive questions 


  use a random response technique - this uses a random event so people who answer truthfully remain anonymous

  e.g. toss a coin, if it lands on heads tick yes, if it lands on tails tick no