reliability and validity

• A test is valid when it measures what it claims to be measuring and the results aren't affected by extraneous or confounding variables.
• face validity is where the researcher looks at the test to see if it appears to measure what it claims to measure.
• When assessing concurrent validity we compare the results obtained to a previous test of a similar nature that is known to be valid
• Studies are reliable when the results are consistent every time it is repeated
• A study can lack reliability if:

  • extraneous variables affect measurements
  • the test used to measure the DV is unreliable
• external reliability is the extent to which a test measures consistently over time
• internal reliability is the extent to which something is consistent within itself
• the split half method assesses internal reliability by splitting the test into two and comparing results to see if results are consistent across the two halves
• the test-retest method assesses external reliability by having the same participants take the same questionnaire or test on two separate occasions and compare the results obtained both times.
• If a study is repeated using the same method, design and measurements and the same results are produced, the results are said to be reliable.
• inter-observer reliability assesses whether different observers are viewing and rating behaviour in the same way by conducting a correlation of all observers' scores, with high correlation indicating they are observing and categorising behaviour consistently.
• internal validity is whether the study is measuring what it claims to measure or whether the findings are affected by extraneous variables
• High internal validity is where the effect is due to the IV.
• Low internal validity is where the effect is due to a confounding variable.
• Internal validity can be affected by confounding variables, investigator effects, demand characteristics, social desirability bias and poor operationalisation of the DV or behavioural categories.
• Face validity is an intuitive, common sense judgement of whether a self-report measure appears to measure what it claims to.
• Internal validity can be improved by reducing investigator effects, minimising demand characteristics and by the use of standardised instructions and a random sample. These factors ensure a study is highly controlled, leaving less doubt that observed effects are due to poor methodology.
• External validity refers to the extent results generalise beyond the experimental setting, other people and over time.
• there are 3 types of external validity:

  • ecological validity
  • population validity
  • temporal validity
• ecological validity is the extent to which findings of a research study are able to be generalised to different settings
• population validity is the extent to which research findings can be generalised to other population groups
• temporal validity is the extent to which findings are era-dependent or if the findings endure over time