Research methods

• Lab Experiment


  An experiment conducted in a controlled environment where the IV is manipulated
• Quasi Experiment


  An experiment conducted in a controlled environment where the IV is pre-existing
• Field Experiment


  An experiment conducted in the participants' natural environment where the IV is manipulated
• Natural Experiment


  An experiment conducted in the participants' natural environment where the IV is pre-existing
• Lab Experiment: Strengths

  - Cause + Effect can be established

  - Reliable

  - High in internal validity
• Lab Experiment: Weaknesses


  - Low in ecological validity

  - High chance of demand characteristics leading to invalid data
• Quasi Experiment: Strengths


  - High in internal validity
• Quasi Experiment: Weaknesses

  - Cause + Effect cannot be established

  - High chance of demand characteristics leading to invalid data

  - Low in ecological validity
• Field Experiment: Strengths

  - High in ecological validity

  - Cause + effect can be established

  - Low chance of demand characteristics leading to valid data
• Field Experiment: Weaknesses

  - Low in internal validity

  - Unreliable
• Natural experiment: Strengths


  - High in ecological validity

  - Low chance of demand characteristics leading to valid data
• Natural Experiment: Weaknesses

  - Cause + Effect cannot be established

  - Unreliable

  - Low in internal validity
• Reliability


  The extent to which an experiment can be repeated under the same conditions with the same participants
• Internal validity


  The extent to which an experiment measures what it sets out to measure as a result of the levels of control
• Ecological validity


  When the experiment location is new to the participants' leading to the results not reflecting the participants' true behaviour
• Generalisation


  When you take a sample from a target population, study their behaviour, and use the results to explain the behaviour of the whole target population
• Directional hypothesis


  A hypothesis where only one outcome is predicted

  
  

  E.g. An adult with no children will have more hours spent socialising compared to an adult who has children
• Non-directional hypothesis


  A hypothesis where two or more outcomes are predicted

  
  

  E.g. There will be a difference in hours spent socialising between an adult with children and an adult with no children
• Null hypothesis


  A hypothesis predicting no difference

  
  

  E.g. There will be no significant difference in hours spent socialising between an adult who has children and an adult who has no children
• Operationalisation


  The process of specifically defining a variable before measuring it

  
  

  E.g. Aggressive behaviour - kicking, pushing, punching hitting

  
  

  The purpose is to be objective and to allow for replication
• Extraneous variable


  Variables other than the IV that, unless controlled, could affect the DV

  
  

  - Can only be controlled in a lab setting
• Situational variable


  - A type of EV

  
  

  Light, noise levels, temperature
• Participant variables


  - A type of EV

  
  

  IQ, age
• Order effects


  - A type of EV

  
  

  When participants begin to feel tired/ bored, affecting the validity of data
• Demand characteristics


  - A type of EV

  
  

  When participants figure out the aims of the experiment, leading to invalid data
• Investigator effect


  - A type of EV

  
  

  When the researcher consciously/unconsciously influences the participants' responses, leading to invalid data
• Standardisation


  The process of ensuring all situational variables are the same for all participants
• Counter-balancing


  The process of changing the order of tasks between groups to prevent order effects from taking place
• Single blind technique


  When participants' don't know the true aim of the experiment
• Double blind technique


  When neither the researcher or participants know the true aim of the experiment
• Independent groups design


  When participants take part in one condition
• Repeated measures design


  When participants take part in two or more conditions
• Matched pairs design


  When participants take part in one condition but are also matched based on participant variables
• Independent groups design: Strengths + Weaknesses

  + Order effects won't take place, improving the validity of data

  
  

  - Expensive - more participants needed
• Repeated measures design: Strengths + Weaknesses

  + Cost-effective - less participants needed
• Matched pairs design: Strengths + Weaknesses

  + Order effects will not take place

  
  

  - Matching people on certain characteristics is time consuming
• Target population


  The group of individuals the researcher is interested in e.g. all UK teenagers
• Sampling frame


  A list containing all the names of the target population
• Sample


  A small group of people who represent the target population
• Random Sampling


  Everyone in the target population has an equal chance of being chosen