Self-Report Techniques

• Questionnaires


  Written sets of questions designed to quickly accumulate information from a large number of respondents.
• Open questions

  Questions with no fixed answer/response and respondents can answer in any way they wish.
• Closed questions


  Questions that can usually be answered with yes or no.
• Strengths of questionnaires

  + Cost effective

  + Can gather large amounts of data quickly

  + Can be completed without the researcher being present
• Limitations of questionnaires

  + people can lie due to social desirability bias (type of demand characteristics)

  + response bias- answer in the same way e.g all yes or no etc. due to doing the questionnaire too quickly

  + Acquiescence bias- the tendency to say yes
• Interviews

  A formal or informal approach to elicit information from participants by talking to them directly.
• Structured interviews


  Interview process that asks the same job-relevant questions of all applicants, each of whom is rated on established scales
• Unstructured interviews


  Interviews in which interviewers are free to ask the applicants anything they want
• Semi-structured interviews

  There is a set of core questions or topics that the interviewer will follow, but the interviewer may prompt for more information, ask follow-up questions, or clarify questions as the interviewer deems necessary
• Strengths of structured interviews

  - Easy to administer

  - Do not need to establish a rapport between researcher and respondent
• Limitations of structured interviews


  - interviewer may influence answers

  - take time

  - several interviewers may approach the task differently

  - participants may give socially acceptable answers instead of honest ones
• Strengths of unstructured interviews

  - More flexible as questions can be changed

  - Increased validity

  - Usually in depth and detailed answers
• Limitations of unstructured interviews

  1. researcher may influence response - decreases validity

  2. focus on one person - cannot generalize

  3. lacks representativeness