Demand Characteristics

Demand Characteristics and Investigator Effects

Psychological research involves a social encounter between researcher and participant. This social dimension introduces sources of bias beyond the manipulation of variables — most importantly, demand characteristics and investigator effects. Both threaten the validity of research by producing changes in participant behaviour that reflect the social dynamics of being studied rather than the effect of the independent variable.

Demand Characteristics

Demand characteristics are any cues in the research situation that allow participants to infer the study's purpose or the researcher's hypothesis, leading them to alter their behaviour accordingly. The term was introduced by Martin Orne (1962), who observed that participants in psychological experiments are not passive responders — they are active interpreters of the research situation who seek to understand what is expected and respond accordingly.

Participants may respond to demand characteristics in several ways:

• Trying to help: participants guess the hypothesis and deliberately confirm it — the 'good participant' effect.
• Trying to look good: participants behave in socially desirable ways regardless of the actual IV — presenting themselves favourably.
• Trying to look bad: some participants deliberately subvert the apparent purpose — the 'screw you' effect.

Demand characteristics undermine internal validity because any change in the DV may reflect the participant's response to perceived expectations rather than the genuine effect of the IV.

Reducing demand characteristics: using deception (not revealing the true purpose until debriefing); single-blind procedure (participants unaware of their condition); conducting the study in naturalistic settings where cues are less salient; using indirect or behavioural measures rather than self-report; using within-experiment controls.

Investigator Effects

Investigator effects (also called experimenter effects) are the ways in which the researcher's own characteristics or behaviour — intentionally or unintentionally — influence participants' responses or the outcome of a study. They include:

• Experimenter bias: the researcher's expectations about the study's outcome subtly influence how they interact with participants, record data, or interpret ambiguous results — producing a self-fulfilling prophecy. Robert Rosenthal's research on experimenter expectancy effects demonstrated this.
• Physical characteristics: the researcher's age, gender, ethnicity, or appearance may affect participant behaviour — for example, participants may respond differently to a young female vs an older male experimenter.
• Verbal and non-verbal cues: unintentional variations in tone, emphasis, or body language when delivering instructions may prime participants differently across conditions.

Reducing investigator effects: using a double-blind procedure — neither the participant nor the experimenter conducting the study is aware of which condition the participant is in. Standardised instructions (delivered by computer or audio recording) reduce verbal and non-verbal variability. Using multiple experimenters and comparing their data can identify experimenter-specific effects.

The Single-Blind and Double-Blind Procedures

In a single-blind study, participants do not know which condition they are in, reducing demand characteristics. In a double-blind study, neither the participant nor the experimenter knows which condition the participant is in — this additionally prevents unconscious experimenter bias from affecting data collection and interpretation. Double-blind designs are the gold standard in clinical trials of drug effectiveness.