Objectivity, reliability, validity and experimenter effects

Reliability and objectivity

• Reliability means consistency: a study is reliable if it gives the same results when repeated with the same procedure. Standardised procedures make replication possible. Test-retest and inter-rater checks assess reliability.
• Objectivity means data are not influenced by the researcher's opinions or expectations, for example counting words recalled. Subjectivity means data depend on interpretation (for example, judging how a story has changed).
• Generalisability: how far results apply to people beyond the sample (depends on a representative sample).
• Credibility: how far findings can be believed, for example because they are supported by other studies.

Types of validity

Validity means the study measures what it claims to and reflects real life.

• Internal validity: the IV really caused the change in the DV, with no confounding variables, experimenter effects or demand characteristics.
• Ecological validity: findings apply to real-life settings and tasks. Learning lists of trigrams in a lab has low ecological validity.
• Predictive validity: a measure correctly predicts future behaviour or performance (for example, a memory test score predicting exam success).

Experimenter effects and demand characteristics

• Experimenter effects: the researcher's expectations or behaviour (tone, smiles, cues, errors in recording) influence the results. Control: double-blind procedures (neither participant nor researcher who runs the study knows the condition), standardised instructions.
• Demand characteristics: cues that let participants guess the aim, so they change their behaviour (to help or to spoil the study). Control: single-blind procedures, deception about the aim (with debrief), independent groups.
• Control issues: the more control, the higher the internal validity but often the lower the ecological validity.

Worked example

A researcher who expects music to harm recall marks the recall sheets herself and knows which condition each sheet came from. Identify the problem and suggest a solution.

1. Problem: experimenter bias. She might mark leniently in the silence condition.
2. This reduces objectivity and internal validity.
3. Solution: double-blind marking; a second person marks sheets without knowing the condition.

Answer: Experimenter effects: use a blind marker who does not know the condition (double-blind).

Check yourself

1. A study is repeated with the same procedure and gives similar results. This shows
   • ecological validity
   • predictive validity
   • reliability
   • subjectivity
   Show answer

   reliability. Consistency on replication is reliability.

2. Which study is most likely to have low ecological validity?
   • Recalling lists of nonsense trigrams in a laboratory
   • Observing children in their own playground
   • A field experiment in a real shop
   • Recording real conversations in a cafe
   Show answer

   Recalling lists of nonsense trigrams in a laboratory. Artificial tasks do not reflect everyday memory.

3. A memory test given to 16-year-olds correctly predicts their exam results two years later. This shows
   • inter-rater reliability
   • ecological validity
   • predictive validity
   • demand characteristics
   Show answer

   predictive validity. The measure predicts future performance.

4. Participants work out the hypothesis and try to help the researcher. This is an example of
   • a situational variable
   • demand characteristics
   • an order effect
   • counterbalancing
   Show answer

   demand characteristics. Cues reveal the aim and change behaviour.

5. What does internal validity mean?
   • The findings apply to everyday life
   • The results are the same when repeated
   • The sample represents the population
   • The change in the DV was caused by the IV and not by other variables
   Show answer

   The change in the DV was caused by the IV and not by other variables. Internal validity is about cause and effect within the study.

6. Explain how a double-blind procedure reduces experimenter effects.
   Show answer

   The researcher running the study does not know which condition participants are in, so their expectations cannot influence how they behave towards participants or record results.. If the researcher does not know the condition, their expectations cannot bias the results.

Practise with instant feedback Print this lesson

Sources

• Pearson Edexcel International Advanced Subsidiary/Advanced Level in Psychology specification (issue January 2026)
• Pearson International A Level Psychology component guide: WPS01
• IAL Psychology revision site: Validity and reliability

Last updated 8 November 2026. Spotted a mistake? Tell us.