Experimenter's Bias

Experimenter's bias, also called the observer-expectancy effect or experimenter-expectancy effect, is a form of reactivity in which a researcher's own expectations unconsciously influence the outcome of an experiment, either by leading the researcher to interpret ambiguous results in a way that favors their hypothesis or by causing the researcher to subtly communicate cues to participants that prompt them to behave in the expected way. Confirmation bias plays a part in the effect: an experimenter checking results against a hypothesis may notice and register information that supports it while overlooking information that conflicts with it. The classic illustration is the case of Clever Hans, a horse whose owner claimed could perform arithmetic; the philosopher Carl Stumpf and his assistant Oskar Pfungst investigated the claim and found the horse answered correctly 89 percent of the time when its questioner already knew the answer but only 6 percent of the time when the questioner did not, tracing the horse's apparent skill to unconscious changes in the questioner's posture and facial tension that released at the moment of the correct answer. Modern research addresses the effect through double-blind experimental designs, in which neither the experimenter administering a trial nor the participant knows which condition is being tested, along with practices such as preregistering a hypothesis before data collection begins.