Self Reports

• A semantic differential scale is when the participant is given two opposite adjectives and asked to rate on a scale where their opinion falls.

  
• The Likert scale is when as participant is asked to rate how far they agree with a statement.
• A numbered rating scale is when is when the participant gives a rating for each item. (You must always say what 1 and 10 represent.)
• Evaluation of closed questions

  :) Allows for patterns and comparisons to be made as it collects quantitative data which increases external reliability.

  :) Allows for objective analysis of the answers which is more scientific.

  :( Answers are limited which reduces insight and internal validity.

  :( Participants may not feel like their true response is not represented so their findings may not be a true reflection.
• Evaluation of open questions

  :) Collects qualitative data which increases insight and the internal validity of the questions.

  :) Can be more truthful and representative as they do not have to fit their answer into pre-selected options.

  :( Information gained is difficult to analyse as it is more subjective. This makes it less scientific and reduces external reliability.
• In a structured interview the questions are pre-planned before the interview and the researcher does not deviate from these questions.

  :) High internal reliability as the questions are standardised this means we can replicate the interviews to gain more info on the behaviour.

  :( Doesn't allow for spontaneous questions as they have all been planned which reduces insight and internal validity.
• In an unstructured interview it runs as a free flowing conversation where questions are spontaneous and thought of based on the participants previous questions.

  In a semi-structured interview it is a compromise where some questions are pre-planned and others are based on previous answers.

  :) Increases insight as questions can be used to delve deeper into topics brought up in participant responses this increases internal validity.

  :( Low internal reliability as the procedure is not standardised so we cannot easily replicate the interview for many people - less scientific.