Hawthorne Effect

The Hawthorne effect refers to a methodological artifact (see ARTIFACTS IN RESEARCH PROCESS) in behavioral FIELD EXPERIMENTS arising from the awareness of research participants that they are being studied. This awareness leads them to respond to the social conditions of the data collection process rather than to the experimental treatment the researcher intended to study. Somewhat akin to what has been called the “guinea pig effect” in laboratory research, the artifact reduces the INTERNAL VALIDITY of experiments. The term Hawthorne effect may also be used in a management context to refer to workers' improved performance that is due to special attention received from their supervisor. Some researchers may confuse these usages, incorrectly interpreting improved performance due to a Hawthorne effect as a desirable outcome for their experiment.

Both effects were observed in a series of studies conducted in the 1920s and 1930s at the Hawthorne Works of the Western Electric Company (Roethlisberger & Dickson, 1939). Investigating the influence of various conditions on work performance, researchers were surprised that each new variable led to heightened levels of productivity, but were even more surprised when increased levels of productivity were still maintained after all improvements to working conditions had been removed. Researchers concluded that this unexpected result had been caused by incidental changes introduced in an attempt to create a controlled experiment.

Although there has been continuing controversy over the exact source of the artifact, and whether there was even evidence for it in the original experiments (Jones, 1992), the Hawthorne studies provided some of the first data suggesting the potential for social artifact to contaminate human research. The Hawthorne effect has been widely considered to be a potential contaminant of field experiments conducted in real-life settings, most notably in education, management, nursing, and social research in other medical settings. Researchers in a wide range of social sciences continue to control for Hawthorne effects, or claim its potential for misinterpretation of their research results.

Rather than procedures preventing its occurrence, the extent of artifact in a study is typically addressed by the inclusion of CONTROL GROUPS to measure the difference in performance on the DEPENDENT VARIABLE between a no-treatment control group and a “Hawthorne” group. The conditions manipulated to create the Hawthorne control group are one of the variables typically thought to generate Hawthorne effects; the others are special attention, a novel (yet meaningless) task, and merely leading participants to feel that they are subjects of an experiment. The EFFECT SIZES associated with each of these Hawthorne control procedures in educational research (Adair, Sharpe, & Huynh, 1989) have been found by meta-analyses (see META-ANALYSIS) to be small and nonsignificant, suggesting that these control groups have been ineffective for assessing artifact.

In spite of this lack of evidence in support of any of its controls and the difficulties in defining the precise nature of the Hawthorne effect, the potential for artifact within field experiments remains. The significance of the Hawthorne studies was to suggest how easily research results could be unwittingly distorted simply by the social conditions within the experiment. In contemporary research, the term Hawthorne effect is often used in this broad sense to refer to nonspecific artifactual effects that may arise from participants' reactions to experimentation.

...