What is the Hawthorne Effect?
The Hawthorne effect occurs when experimental participants change their behavior because they know researchers are watching them. Typically, this effect refers to cases where subjects improve their performance levels. However, these are short-term improvements that vanish when the observation stops. Consequently, the study results are deceptive because they do not reflect a natural response to the experimental factors.
The Hawthorne effect is widely known in psychology, especially industrial and organizational psychology. The term is named after the location of the original experiments in the 1920s, Western Electric’s Hawthorne Plant. Researchers wanted to determine whether changes in the workplace environment affect worker productivity. The most well-known portion of the study involved assessing how lighting levels affect productivity.
Every time the researchers brightened the lights during the study, the subjects increased their productivity. At first, it seemed like brightening the lights caused the performance to improve. However, when researchers dimmed the lighting, productivity also increased. Eventually, the researchers realized that productivity increased following any lighting change because the workers knew they were being closely observed at those times. After the experiment ended, productivity decreased to normal levels.
In an experiment, the Hawthorne effect can threaten both internal and external validity. The knowledge of being observed can improve the participants’ performance levels rather than the experimental intervention. Hence, this bias is potentially a confounding variable because it provides an alternative explanation for improved outcomes. In this manner, it reduces internal validity by weakening the case for a causal relationship between the experimental factors and the results. However, it can also reduce external validity because the unnatural behavior it causes does not generalize to unobserved populations.
Is the Hawthorne Effect Real?
Later investigation found that methodological shortcomings of Western Electric’s study might have exaggerated the original Hawthorne effect to some degree. However, more recent studies using better methods continue to support its existence.
For example, a systematic review of 19 studies by McCambridge et al. in the Journal of Clinical Epidemiology (2014) found that most of the studies showed behavior changes linked to research participation. However, findings varied widely by context, study design, and the behaviors studied, suggesting that no single mechanism or effect size defines the Hawthorne effect. The review notes a need for more consistency in defining it across studies and for new frameworks to understand how and when research participation impacts behavior.
Examples of the Hawthorne Effect
The following examples highlight how the Hawthorne effect can influence behavior in various contexts. When people know researchers are watching them, they might change their actions to align with perceived expectations or social norms. These changes are short-term and revert to normal when the observations stop, reducing the validity of the research. Here are some examples documented in the literature.
Healthcare Hand Hygiene Compliance
Studies have shown that healthcare workers improve their hand hygiene practices when they know they are being observed. This knowledge inflates compliance rates in studies assessing hygiene interventions.
Student Performance in Testing Environments
Research has found that students perform better on tests or assignments when they are aware evaluators are closely monitoring them. Consequently, these tests can overestimate typical performance.
Customer Service in Retail
Employees in retail settings provide more attentive and polite customer service when they know a supervisor or mystery shopper is observing them. This observation can skew assessments of usual service quality.
How to Reduce the Hawthorne Effect
Researchers must minimize all biases, such as the Hawthorne effect, to trust experiment results. While there are uncertainties around the mechanisms and conditions that trigger these behavioral changes, the literature offers practical ways to reduce their impact.
By applying the following strategies to minimize the Hawthorne effect, researchers can gain more accurate insights into genuine behaviors and outcomes:
- Use Blinding: Ensure participants are unaware of the specific behaviors or outcomes the study is assessing to minimize their reaction to observation.
- Minimize Observer Interaction: Limit contact between observers and participants to reduce the influence of perceived researcher expectations.
- Implement Longitudinal Observations in Natural Settings: Conduct observations over an extended period or in a way that integrates naturally into participants’ routines so they become less aware of being studied. Consider Naturalistic Observation.
- Use Control Groups Not Exposed to Observation: Compare outcomes with a control group that is not exposed to observation or research participation to identify any effects attributable solely to the Hawthorne effect.
- Automate Data Collection: Rely on automated or indirect measurement tools (e.g., sensors or digital tracking) to observe behaviors without active researcher presence.
Learn more about Experimental Designs: Definition and Types.
Comments and Questions