Controls for Experimental Methods

• One problem with experimental methods is demand characteristics:

  • Any clues given off within the research could lead to participants changing their natural behaviour to help or hinder the research, reducing the internal validity
• To control for demand characteristics, a single blind procedure should be used:

  • This keeps the aim of the research hidden from the participant so they cannot show demand characteristics
• One problem with experimental methods is investigator effects

  • When a researcher unintentionally influences the outcome of their research
• To control for investigator effects, a double blind procedure should be used:

  • Get another researcher who does not know the aims of the research to carry out the experiment - the aim of the research is hidden from both the participants and researcher
• One problem with experimental methods is bias in research:

  • When the researcher is designing the resources/experiment or collecting results
• To control for bias in research, randomisation should be used:

  • WHY: Where possible aspects of the research should be random and not decided by the researcher which may be open to bias - this could be present to an extraneous variable
  • HOW: When deciding the order of conditions, this should be randomly decided through a computer generator or pulling pieces of paper out a hat, and not by the researcher
• To control for bias in research, standardisation should be used:

  • WHY: All participants should experience the same environment and instructions to avoid extraneous variables affecting measurements
  • HOW: Use the exactly same formalised instructions and procedures for all participants in a research study, this includes standardised instructions that are read out/given to all participants
• ISSUE + CONTROL

  • Demand characteristics -> Single and double blind procedures
  • Investigator effects -> Double blind procedures
  • Bias -> Randomisation and standardisation