Experiments

• What are experiments 


  the only research method which can establish C+E link between 2 variables (the IV + DV). only method with a DV + IV. want the IV to be the only variable that causes an effect on the DV - any other extraneous variables must be controlled so they don't confound the DV
• What is the independent variable 


  the variable manipulated by the researcher
• What is the dependent variable


  the variable being measured by the researcher
• What are extraneous variables


  any variable that we don't intend to study - they need to be controlled so that they don't become confounding
• Features of a lab experiment 


  • highly controlled, artificial setting
  • manipulation of the IV
  • measures the DV
• Features of a field experiment 


  • natural setting
  • manipulation of the IV
  • measures a DV
• Features of a quasi experiment 


  • can be in an artificial or natural setting (usually artificial)
  • naturally occurring IV, no manipulation of IV
  • measures a DV
• Strengths of a lab experiment 


  • high levels of control over possible EVs due to artificial setting= cause and effect can be established, high IV
  • highly standardised procedure due to artificial setting = allows replication of the study/increases replicability = high IR
• Weaknesses of a lab experiment 


  • low ecological validity due to the artificial setting = hard to generalise to real life
  • high risk of demand characteristics due to artificial setting (catch onto aim)= lack of internal validity
• Strengths of field experiments


  • high ecological validity due to natural setting = can generalise behaviour to real life
  • due to participants usually being unaware they are taking part, behaviour is more natural than in lab experiment so reduced demand characteristics = increased internal validity
• Weaknesses of a field experiment 


  • due to natural environment there is lack of control of EVs = lack of validity and so harder to establish cause + effect
  • due to natural environment there will be less standardisation = less able to replicate lowering internal reliability
• Strengths of quasi experiment 


  • researcher can use a naturally occurring IV so enables them to study behaviour that would be unethical or not practical to manipulate e.g phobias = more insight in behaviour we couldn't study otherwise, high IV
  • due to artificial setting, high levels of control over EVs = able to establish cause + effect, high IV
• Weaknesses of quasi experiment 


  • due to naturally occurring IV, cannot randomly allocate Ps to conditions which may increase the risk of individual differences = lack of internal validity
  • quasi experiments are often conducted in labs so if the task is unnatural OR artificial setting then the study could lack ecological validity = can't generalise to real life
• Features of independent measures


  • 2 or more experimental conditions + Ps only take part in one condition
  • Ps are randomly allocated to one of the experimental conditions
  • (different Ps take part in each condition)
• Strengths of independent measures


  + no risk of order effects as Ps only take part in one condition = increased internal validity

  + reduced risk of demand characteristics as Ps only take part in one condition = increased internal validity
• Weaknesses of independent measures 


  high risk of individual differences as Ps only take part in one condition = low internal validity
• Features of repeated measures 


  all Ps take part in all/both experimental conditions
• Strengths of repeated measures


  no risk of individual differences as Ps take part in all conditions = increased internal validity
• Weaknesses of repeated measures


  high risk of order effects (fatigue, boredom, practice) as Ps take part in all conditions = lack of internal validity

  high risk of demand characteristics as Ps take part in all conditions = low internal validity

  (use counterbalancing to reduce these effects)
• Features of matched participants

  Ps are matched on key characteristics that could affect the results and then the each matched P in the pair is randomly allocated to each condition (of the IV)
• Strengths of matched participants


  + no risk of order effects as Ps only take part in one condition= increased internal validity

  + due to matching Ps on key characteristics, controls for individual differences = high internal validity
• Weakness of matched pairs


  very difficult and time consuming with large sample sizes = less economical

  still risk of individual diffs = low IV
• What are order effects


  when the participant experiences more than one experimental condition, their performance in the secondary condition may be affected/change their behaviour due to taking part in another condition previously:

  fatigue (decrease score)

  boredom (decreases score)

  practice (improve/increase score)

  these confound results
• What is counterbalancing

  technique used to control for order effects when using a RMD

  the sample is divided in half with one 1/2 completing the 2 conditions in one order and the other 1/2 completing them in the opposite order

  Another way to control for order effects: use IMD
• What are situational variables and how to control them

  Factors in the environment which can affect Ps behaviour/results e.g. noise, time of day, crowded/isolated places, temperature etc

  CONTROL: use lab experiment with a highly controlled artificial setting OR use a standardised procedure
• What are individual differences and how to control them

  Any differences between participants that may impact their behaviour not accounted for in the IV e.g. age, mood, background, ethnicity, IQ etc

  CONTROL: use matched pairs or repeated measures design
• What are researcher effects and how to control them

  When the researcher behaves differently to diff Ps/conditions (e.g. the way instructions are delivered), it can impact participants behaviour. they shouldn't give any cues to Ps about what is expected of them that could encourage certain behaviours (e.g. DCs)

  CONTROL: use a double blind procedure + standardised instructions
• What are demand characteristics and how to control them

  When participants figure out the aim so change their behaviour to fit in line with it

  CONTROL: IMD, deception, single blind procedure, field experiments
• What is social desirability and how to control it 


  When participants change their natural behaviour to appear socially desirable

  Have participants complete task in a room on their own
• What is a single blind procedure


  Participants aren't which condition they are in/the true aim in an experiment - helps control for DCs = high IV
• What is a double blind procedure 


  The researcher and the participants are both unaware of what condition participants are in/when Ps aren't aware of the true aims + researchers aren't aware of which condition each P is in - controls for DCs, researcher effects + order effects = high IV
• Type of experiments and validity table