Reliability of Qualitative & Quantitative Methods (OCR GCSE Psychology): Revision Note

Reliability

• Reliability refers to how consistent or replicable the results of a study are

• If the same procedure is repeated and produces similar results, the findings are considered reliable

• Psychologists aim to make their research as reliable as possible so that results can be trusted and generalised

Reliability in psychological research

• Reliability can apply to any research method — experiments, questionnaires, interviews, and observations

  • Lab experiments are the most reliable method, as they use controlled conditions and standardised procedures, making replication easier

  • Questionnaires and structured interviews are reliable, as the questions are standardised

• Field and natural experiments are less reliable because of extraneous variables that the researcher cannot control

• When findings are replicated under similar conditions and produce consistent results, reliability is established

Types of Reliability

Internal reliability

• Internal reliability refers to whether a test or measure is consistent within itself

  • E.g. if an IQ test contains easy questions at the start and much harder ones later, it may lack internal reliability

• However, if all items on a questionnaire measure the same construct (e.g., depression) consistently, it has good internal reliability

Testing internal reliability

• Internal reliability can be tested using the split-half method

• This is where the researcher divides the test into two halves (e.g. odd vs. even questions) and compares scores on both halves

• If both halves produce similar results, the test has high internal reliability

External reliability

• External reliability refers to how consistent the results are over time or across situations

• If a test or study is repeated with the same participants (or under the same conditions) and produces similar results, it has high external reliability

Testing external reliability

• External reliability can be tested using the test-retest method

• This is where the same participants complete the same test on two separate occasions (e.g. completing the same depression questionnaire several weeks apart)

• If the results are similar, the measure is externally reliable

Inter-rater (or inter-observer) reliability

• Inter-rater reliability measures how consistent two or more observers or researchers are when recording or interpreting behaviour

• This is especially important in observations, where researcher bias could affect what is recorded

Testing inter-rater reliability

• Observers agree on clear behavioural categories before the observation

• Each observer records data independently (e.g. using a tally chart)

• Afterwards, their data sets are compared by testing for a correlation

• A strong positive correlation shows high inter-rater reliability, meaning the observation is consistent and unbiased

Improving reliability

• Researchers can improve reliability by:

  • using clear and standardised instructions and procedures

  • training observers or interviewers carefully

  • conducting pilot studies to identify inconsistencies

  • repeating studies under the same conditions to check for consistent outcomes