The John Henry effect is a research bias that occurs when people in a control group work harder than they normally would because they know they’re being compared to a new method or treatment. This extra effort from the control group can mask the true benefits of whatever is being tested, making an effective innovation look like it doesn’t work.
The name comes from the American folk hero John Henry, a formerly enslaved steel driver who, in the 1870s, faced replacement by a steam-powered drill. According to the legend, Henry swung a heavy hammer in each hand and outperformed the machine, but died soon after from the effort. In research terms, Henry represents the control condition, threatened by an innovation, pushing himself to produce results that were anything but typical.
How the Effect Was First Identified
The term was coined in 1970 by education researcher Robert Heinich, who noticed something strange in studies comparing new teaching technologies to traditional classroom instruction. The innovative methods kept failing to show better results, which seemed unlikely given the resources poured into them. When Heinich looked more closely, he realized the traditional classroom teachers were performing well above their normal level. They knew their work was being compared to alternative education methods, and they rose to the challenge.
Gary Saretsky investigated further and found that classroom teachers instructing the control groups were actively trying to outperform the “outsider” performance contractors working with the experimental groups. Their students scored higher than would be expected under normal circumstances, not because traditional teaching was superior, but because the teachers were putting in extraordinary effort they wouldn’t sustain long-term. This change invalidated results that could have been used to promote genuine reform in educational practices.
Why Control Groups Change Their Behavior
The psychological driver behind the John Henry effect is compensatory rivalry. When people assigned to a control group realize they’re not receiving the new treatment or method, they can feel disadvantaged, disappointed, or left out. In response, they decide to compensate on their own, sometimes by working harder, sometimes by seeking out similar resources independently.
This rivalry doesn’t always involve direct competition with the experimental group. Sometimes it’s subtler: a teacher who hears about a new curriculum being tested down the hall might simply prepare more thoroughly for lessons, stay later, or pay closer attention to struggling students. The key ingredient is awareness. Once control group participants know they’re the “status quo” being measured against something new, their behavior can shift in ways that have nothing to do with the study’s actual variables.
How It Distorts Research Results
The John Henry effect is a threat to internal validity, which is a study’s ability to accurately determine cause and effect. In a well-designed experiment, the control group is supposed to represent normal conditions. If control participants perform at an artificially high level, the gap between the two groups shrinks or disappears entirely. A treatment that genuinely works can appear to have no effect at all.
This is especially problematic in fields like education, public health, and workplace interventions, where it’s often impossible to hide from participants which group they’re in. A school knows whether it received a new reading program. A hospital ward knows whether it’s piloting a new workflow. That awareness is all it takes for the John Henry effect to kick in and quietly undermine the findings.
John Henry Effect vs. Hawthorne Effect
These two biases are often confused because both involve people changing their behavior simply because they’re part of a study. The difference comes down to which group is affected and why.
- Hawthorne effect: People in the treatment group change their behavior because they know they’re being observed or receiving special attention. They might try harder, report better outcomes, or behave differently than they would outside a study, regardless of whether the treatment itself does anything.
- John Henry effect: People in the control group change their behavior because they feel they’re in competition with the treatment group or feel threatened by potential replacement. Their motivation comes from rivalry or a desire to prove the status quo works just fine.
Both effects can operate in the same study simultaneously. The treatment group performs better because of the attention (Hawthorne), while the control group performs better because of the competition (John Henry). Together, they can flatten the difference between groups and make a study’s conclusions unreliable.
How Researchers Minimize the Effect
The most effective strategy is blinding: keeping participants unaware of which group they’ve been assigned to. If people don’t know they’re in the control condition, they have no reason to feel threatened or competitive. In medical trials, this is accomplished through placebos. In behavioral or educational research, blinding is harder but not impossible. Some studies use “active controls,” where both groups receive an intervention but only one gets the specific ingredient being tested.
Reducing direct contact between the researcher and participants also helps. Written instructions, automated data collection, and online experiments all limit the social cues that might tip someone off about their group assignment. When participants interact with a researcher face to face, subtle signals (tone of voice, level of enthusiasm) can reveal which condition they’re in, even unintentionally.
Study designers can also use unobtrusive measures, collecting data through records or observations that participants don’t know are being tracked. If a teacher doesn’t realize student test scores are the outcome measure, there’s less incentive to teach to the test. The goal across all these strategies is the same: remove the awareness that triggers compensatory behavior in the first place.