Research Issues & Controls (AQA AS Psychology): Revision Note

Demand characteristics

• During the research process, there are likely to be instances of interaction/inteference between the research process and the participant; this is called a demand characteristic

• These interactions can adversely affect the research findings

• Some examples of demand characteristics include:

  • participants may pick up on cues as to what is expected of them and they may think they have worked out what the aim of the research is

  • the procedure setting; lab settings are different to more natural settings, the unnatural lab setting may cause participants to respond in a certain way

  • any form of communication - implicit or explicit- can affect the outcome of the study

  • participants behaving in different ways towards the researcher:

    • Trying to please, annoy or be nervous

• If participants behave according to what they think the aim of the research is it means that their performance on the task is likely to be artificial/contrived

Controlling demand characteristics

• One way to control for demand characteristics is to use a single-blind procedure

  • This means that the participants do not know which condition they have been assigned to

  • Therefore, they are not able to respond according to what they think is being tested in that condition

Investigator effects

• Investigator effects occur when the researcher's presence/behaviour interferes with the research process and becomes a source of bias

• The researcher's characteristics, such as age, gender and ethnicity, could influence how participants interact with them e.g.

  • the researcher may remind a participant of a schoolmate who bullied them or a schoolmate they secretly had a crush on

    • This sense of familiarity will affect how the participant approaches the task set by the researcher

• The researcher's accent, tone of voice, non-verbal communication and even what they are wearing can impact how participants react to them and the research process generally e.g.

  • a very broad accent may lead to participants stereotyping the researcher which could affect how seriously they take the task (this is not the fault of the researcher of course but it is something to bear in mind)

  • a harsh/bored/over-excited tone of voice will introduce an emotional component to what should be a neutral space

  • using overly dramatic body language is at odds with scientific research and is likely to put participants in the wrong frame of mind

  • wearing clothing that is overly vibrant, patterned or includes slogans is too personal, distracting and not befitting a scientific environment

• The researcher could be biased in the way that they instruct participants or lead a task e.g.

  • if the researcher has hypothesised that X condition will result in higher scores, then they may unconsciously communicate this to the participants

Controlling investigator effects

• One way to control for investigator effects is to use a double-blind procedure

  • This means that the participants and the researcher do not know which condition each participant has been assigned to

    • Therefore, the researcher is not able to exercise any forms of bias during the procedure and when analysing the results

Randomisation

• Randomisation refers to the deliberate avoidance of bias on the part of the researcher to keep the research as objective as possible

• Randomness describes a lack of purpose, cause, order or predictability where outcomes do not follow a predetermined pattern

• Random allocation of participants to conditions is one form of control that researchers use, as it excludes the possibility of bias invalidating the research

  • Participants are randomly allocated or assigned to one condition of the IV by methods such as selecting names one by one out of a hat or by using random name-generator software

• Randomisation also relates to the procedural aspects of the research process, e.g.:

  • if the procedure involves a list of words, digits or images presented to participants, then the list must be decided randomly to avoid researcher bias

Standardisation

• Standardisation is the term used to describe the identical procedure set up in an experiment (or the questions used in self-report measures) across all conditions/participants

• Standardisation means that no participant receives an unfair advantage or is treated any differently than any of the other participants

• Implementing standardisation allows the research to be replicated, which in turn makes it reliable

• Standardised aspects of the procedure are as follows:

  • Standardised instructions are given to all participants (tailored per the condition of the IV)

  • Standardised briefing before the procedure (including the standardised consent form) and standardised debriefing after the procedure has taken place

  • Equal number of participants per condition e.g.

    • in an experiment with a sample size of 40, the researcher would ensure that there were 20 participants per condition

      • Sometimes this is not possible due to participant numbers e.g. 19 in one condition and 20 in the other condition

  • Standardised timings with each condition of the IV running for the same amount of time e.g.

    • 15 minutes for condition A

    • 15 minutes for condition B (unless one of the conditions includes a time delay)

  • Standardised materials with identical materials used, the only exception being if the materials need to change for the IV to be implemented e.g.

    • condition A involves learning a poem underwater and condition B involves learning a different poem on land (a repeated measures design

Control groups

• In an experiment with an independent groups or matched pairs design, a control group is used:

  • A group of participants that receives no treatment at one level of the IV

• The purpose of a control group is to:

  • serve as a baseline for comparison

  • determine whether the treatment of interest has any effect by allowing researchers to compare the results of the treatment group (the experimental group) with the no treatment group (control group)

• Control groups are used in clinical trials, e.g.,

  • In a study testing a new anti-anxiety drug, the control group might receive a placebo while the treatment group receives the actual medication

  • Comparing changes in anxiety levels between the two groups once the trial has ended, allows researchers to assess the drug's efficacy

• Control groups are important because they:

  • establish causality

    • Researchers can determine if the observed changes are truly due to the treatment

  • avoid extraneous variables

    • Control groups help minimise the influence of other factors that could potentially affect the results

  • ensure validity and enhance scientific rigour

    • Control groups increase both internal and external validity

    • They increase reliability as a clearly defined control condition enhances replicability