observations

• observations

  -involve watching & recording behaviour

  -no IV or DV (would make it an experiment)
• types

  -naturalistic

  -controlled

  -covert

  -overt
• naturalistic

  -observation of behaviour in its natural setting

  -researcher makes no attempt to interfere with / influence behaviour
• controlled

  -some variables in environment are controlled

  -reduces naturalness of environment

  -may be in lab
• overt

  -aware of being observed

  -researches try to be unobtrusive
• covert

  -not aware of being observed (before & during)

  -may be informed after
• non-participant

  -no interaction with Ps

  -watching/listening to behaviour of others
• participant

  -observer part of group being observed

  -some interaction with Ps
• naturalistic strength

  high realism (high ecological validity)
• naturalistic weakness

  lack of control over variables

  (conclusions can't be made about patterns of behaviour, can't inter cause & effect)
• controlled strength

  lack of realism (variables controlled)
• controlled weakness

  high control over variables (can infer cause & effect)
• covert strength

  high internal validity (Ps unaware of observation)
• covert weakness

  ethical issues (no informed consent)
• overt weakness

  lack internal validity (demand characteristics)
• overt strength

  lack of ethical issues (informed consent)
• participant strength

  greater insights into behaviour (as observer part of group)
• participant weakness

  -difficult to be objective observer if in group

  -observer may effect behaviour of group (impact internal validity)
• non-participant strength

  lack of direct involvement ensures greater objectivity (increased internal validity)
• non-participant weakness

  data lacks richness as may not be possible to understand feelings of group (lowering internal validity)
• recording observations - sampling


  -event

  -time

  -point
• event sampling


  record how many times certain behaviour occurs in set amount of time
• time sampling


  record what behaviours are occurring at different time intervals
• point sampling


  record what behaviours just one P in group is doing
• observer checklist


  -clearly operationalise

  -specific categories to tick off

  -produce quantitative data

  -clear descriptions of behaviours
• pilot study


  -small-scale trial run of study before main

  -checks there's no problems

  -check behaviours in list are appropriate

  -see if any behavioural categories are missing

  -ensure observers clear about behaviours they're identifying