Control of variables

• Directional hypothesis - one-tailed, researcher makes it clear the sort of difference that is anticipated between the 2 conditions
• Non-directional hypothesis - two-tailed, researcher doesn't indicate the direction in the prediction
• Null hypothesis - when you state there will be no difference, so all changes are due to chance
• Aim - a statement of what the researcher is invistigating
• Operationalise variables - make them tastable and measurable
• Extraneous variable - unwanted factors in the study, if not controlled could negatively affect data collected. Do not vary systematically with the IV. e.g. lightning in the room, age of ppt
• Confounding variable - not included in an experiment, yet affects the relationship between the two variables in an experiment. Do vary systematically with the IV. e.g. in energy drink research, ppt's personality is a confounding variable
• Extraneous variable - affect DV, but don't vary with the IV
• Confounding variable - affect DV and change systematically with the IV
• Demand characteristics - in an experiment give ppt a clue of what researcher expects to find. So they often change the outcome by changing behaviour to confirm the expectations.
• Participant reactivity:

  'Please you' effect - ppt may try to please, do what they guessed is expected

  'Skrew you' effect - deliberately try to skew the results, by doing the opposite of what is expected
• Investigator effects - the experimenter unconsciously conveys to ppt how they should behave -> experimenter biasMight do this by unintentional clues:

  • selection of ppt
  • leading questions
  • expectancy effect
• Randomisation - reduces extraneous and confounding variables, and investigator effect. Use of chance whenever is possible.

  Counterbalancing, random group assigning
• Standarisation - ensuring all ppt are subjected to the same environment condition and experience. Read standard instructions to ppt. To ensure we standarise the procedure.
• To be confounding - it must be correlated to the IV (t correlated to ice cream sales), must have a causal relationship with DV (warmer t had direct causal effect on number of shark attacks, warmer t = more people in the ocean = more attacks)
• Issues with confounding variable (cfv):

  • can make it seem the cause-and-effect relationship exists when it doesn't (cf of t made it seem like ice cream sales and number of shark attacks have it, when we know that they don't)
  • can mask the true cause-and-effect relationship between variables (in study -ability of exercise to reduce blood pressure, cfv - starting weight, it's correlated with exercise and has direct causal effect on blood pressure; so while increasing exercise may lead to reduced blood pressure, starting weight of ppt also has big impact on relationship between two variables)