CONTROL OF VARIABLES

• Aim – purpose of the study  
• Hypothesis – clear, precise, testable statement showing the relationship between the variables – two types (directional and non-directional) 
• Directional hypothesis – precise statement predicting the outcome of the experiment and making the difference between the two conditions clear. 

  Example – those who drink more SpeedUpp are chattier than those who drink less SpeedUpp 
• Non-directional hypothesis – direction of the outcome is not predicted – the results could go either way. 

  Example – there will be a difference in chattiness and alertness based on SpeedUpp intake 
• Steps of a research

  1. Aim – purpose of the study  
  2. Hypothesis – clear, precise, testable statement showing the relationship between the variables – two types (directional and non-directional) 
  3. Research method – experiment, observation, questionnaire, correlation, survey etc 
  4. Data collection 
  5. Data analysis – mean/mode/median/range into a graph  
  6. Results/report findings – writing up the experiment 
• Variables - quality or quantity of something OR anything that can vary or change within an investigation.
• Experimental method has two types of variables: 

  Independent variable – the cause of something. Can either be manipulated by the researcher or change naturally 

  Dependant variable – effect of the IV – measured by researcher 
• Operationalisation of variables – the researcher must describe the process by which each variable will be measured/assessed/implemented in the study 

  Energy drink: Litres 

  Talkativeness: words per minute 
• Extraneous variables – variables that do not affect the experiment too much if controlled or removed – DO NOT systematically vary with the IV  
• Confounding variables – a kind of extraneous variable that varies systematically with the IV. Can often not be controlled and therefore we cannot be sure whether it affects the DV at all e.g. fatigue, mood, health, temperature – unable to be controlled. 
• Demand characteristics – cues that may lead the participants to work out the aim of the study and thus change their behaviour to what they think the researcher wants them to do 

  • Please-U effect – overperformance to please the investigator
  • Screw-U effect – underperformance to sabotage the experiment 
• Investigator effects/researcher bias – any unwanted influence from the investigator that may affect the outcome of the experiment. 
• How to reduce investigator effects

  • Can fix it by training the researcher and be sure the researcher doesn’t know any of the participants 
  • Randomisation can reduce investigator bias as it uses chance methods to decide how to run the experiment OR standardisation which makes all the conditions the same for all the participants