Validity (AQA A Level Psychology): Revision Note

Types of validity in A level psychology

• Validity is the extent to which the findings of a study are representative i.e. 'real' e.g.

  • Do the findings reflect how people really think/feel/behave/act/perform in real life?

  • Do the findings reflect what the researcher has set out to measure?

  • Could anything have gotten in the way of what the research set out to investigate or how real the experience was for the participants?

  • Can the findings be generalised to the wider population and beyond the scope of the research setting?

• Internal validity measures the extent to which the results are due to the manipulation of the IV rather than the influence of extraneous variables, e.g.,

  • Loftus & Palmer's (1974) investigation into the use of leading questions used the same critical question (changing only the verb in each case) across five conditions of the IV

  • Asch's (1951) research on conformity used the same line-length stimuli, with the participant seated at the same place at the table, on each of the critical and non-critical trials

• External validity measures the extent to which the results can be generalised beyond the research setting, e.g.,

  • Ecological validity is high when the task participants are given is more aligned to a real, everyday experience rather than a task that is artificial or contrived or when their engagement with the task is real, even if the task itself is artificial (lacks mundane realism), e.g.,.

    • Milgram's (1963) study on obedience saw participants suffer genuine distress and anguish when they thought they were administering electric shocks to a stranger

    • Rutter et al.'s (2011) Romanian orphans study investigated the real impact of deprivation due to having been raised in an institution

• Temporal validity measures the extent to which research findings are still relevant in the current age, e.g.,

  • Asch's research has been called a 'child of its time' as it reflects the higher rates of conformity that were characteristic of a population that had just come out of the Second World War: the same levels of conformity would be unlikely today

  • Bowlby's concept of the mother as the primary caregiver is outdated in terms of modern-day living with the many different permutations of a family which abound today e.g. single-parent families; blended families; stay-at-home dads; same-sex parents

Measurement of validity

• There are various ways of measuring validity, which include:

  • Face validity

    • This measures whether a test looks like it measures what it set out to measure, e.g., does this digit-span test look like it measures the capacity of short-term memory?

  • Predictive validity

    • This measures validity by measuring how well a test or study can predict future behaviour, e.g., the results of social skills tests on 5-year-olds may predict future educational attainment

  • Concurrent validity

    • This measures how closely two different tests of the same behaviour/skill agree with each other, e.g., participants are given a new IQ test, the scores of which are then compared to their scores on a previously established IQ test - if the scores show a strong positive correlation (+0.8), this is evidence of concurrent validity

Improving validity

• The validity of lab experiments can be improved by:

  • ensuring that controlled conditions and a standardised procedure are in place so that the effects of the IV on the dependent variable (DV) can be observed

    • this can be further strengthened by using a control group as a comparison to the experimental group

  • using single-blind or double-blind procedures to ensure that participants and/or researchers do not know which condition each participant has been allocated to

• Investigator effects occur when a researcher (usually without them being aware of it) influences the outcome of the research e.g.

  • by being overly friendly, by wearing eye-catching clothing

  • by using body language that suggests what they expect to find per condition

  • by bringing their expectations when analysing findings

    • The use of double-blind procedures help to reduce investigator effects

• Demand characteristics occur when participants feel overly self-conscious or too aware of the research procedure e.g.

  • participants in a lab experiment on memory may try too hard which does not reflect how they would use memory in real life

  • participants may try to guess the aim of the study and by doing so their behaviour is not natural

    • Researchers should try to disguise the aim as much as ethical considerations will allow

    • The use of single-blind procedures can also help to reduce demand characteristics

• The validity of observations can be improved by:

  • using covert methods in naturalistic observation, both of which mean that participants' behaviour is likely to be unforced and natural

  • ensuring that behavioural categories are clear, unambiguous, and observable with no overlapping

• The validity of questionnaires can be improved by:

  • using a lie scale to show up inconsistencies in the responses

  • using reverse scoring to ensure that participants are answering all questions with the same direction of response, and not just ticking the same number every time e.g.,

    • 'On a scale of 1-10, how competitive are you? (10 being 'extremely competitive')

    • 'On a scale of 1-10, how do you feel when you lose? (1 being 'I hate to lose')