Observer bias is a type of detection bias where observers' expectations or desires systematically influence their observations and recording of data. This bias can lead to over- or underestimation of findings in scientific studies, thereby compromising their validity. It is especially prevalent when researchers have vested interests or strong preconceptions about the outcomes and can occur even in well-designed studies.
In the early 1900s, psychologist Oskar Pfungst investigated Clever Hans, a horse that seemed to solve arithmetic problems. It was discovered that the horse's owner, Wilhelm von Olsen, unconsciously signaled the correct answers, exemplifying observer bias, as the owner's expectations affected the horse's behavior.
To mitigate observer bias, researchers can employ blind or double-blind study designs to prevent expectations from influencing observations and ensure data collection methods are standardized.