Methods of collecting non-experimental data

• • There are 4 types of non-experimental research methods:​
  • Observational techniques​
  • Self report techniques including questionnaires and interviews​
  • Correlational analysis ​
  • Case studies ​
• • Observations involve watching and recording behaviour, e.g. children in a playground. ​
  • Most observations are naturalistic (occur in real world settings), but can occur under controlled conditions (Milgram’s obedience study)​
  • 2 main types:​

  1. Participant observation: observers are involved in the situation being studied (Zimbardo- Prison simulation study).​
  2. Non-participant observation: researchers are not actively involved in the research (Ainsworth- Strange Situation study).​
• • Observations can also be:​
  • Overt- participants are aware they are being observed​
  • Covert- where participants remain unaware of the fact they are being observed.​
• STRENGTH OF OBSERVATIONAL TECHNIQUES- Value as a preliminary tool- can lead to the identification of appropriate hypothesis for further investigation. prevent wasting time carrying out unrealistic experiments.​

  Validity- checks whether experimental findings can be applied outside lab conditions. Demand characteristics are minimized because behaviour is not affected by anxiety or inhibitions associated with lab exp. ​
• WEAKNESSES OF OBSERVATIONAL TECHNIQUES- Replication- may be difficult due to lack of control over variables.​

  ​

  Observer effects- possible presence of an observer may change the behaviour of those being observed ​

  Costs- can be high when undertaking naturalistic observation​

  ​

  Generalisability of the findings- may be difficult to generalize to everyday situations​

  ​
• • There are several ways of obtaining data in naturalistic observations including: videos, photographs, audio recordings, ‘on-the-spot’ note taking, coding categories. ​
  • Behaviour categories chosen should reflect what is being studied. ​
• • Biggest issue with observations?​
  • Reliability ​
  • A way to overcome this is inter-observer reliability, this occurs when independent observers code behaviour in the same way. ​
  • This lessens the chance of observer bias (seeing what they want to see).​
  • inter-observer reliability needs to be established before an observation begins, along with clearly defined categories that do not overlap. ​
• Sampling methods​

  ​

  • In structured observations the researcher must use a systematic way of sampling their observations:​
  • Event sampling- involves counting the number of times a particular behaviour (event) occurs in a target individual or group. ​

  e.g. counting the number of times players disagree with the referee​
• • Time sampling- involves recording behaviour within a pre-established time frame.​

  e.g. in a football match we may only be interested in one particular player, so we make a note of what an individual is doing every 30 seconds   ​

  ​
• QUESTIONNAIRES:

  • Enables participants to provide information knowingly about specific things relating to themselves.​
  • Involves gathering data from large groups of people.​
  • Questionnaire surveys can be used in a wide range of research situations. ​
  • Questions can be of 2 broad types: ​
  • Closed questions​
  • Open questions ​
• STRENGTHS OF QUESTIONNAIRES: Simplicity- once developed and piloted, questionnaires can be used with a minimum of training. ​

  • Speed and cost large amounts of information can be gathered quickly and cheaply from a large number of respondents within a fairly short period of time​

  Less influence of interpersonal variables- researchers do not have to be present when the questionnaire is being filled out, reducing the opportunity of influencing responses.​