Victim Derogation

What it is

The term describes something observers do, not something victims are. In the experiments, participants watch or read about a person who suffers through no fault of their own, then rate that person on measures such as attractiveness or likeability. Derogation is the result when the ratings fall below what a comparison condition produces.

The founding experiment is Lerner and Simmons (1966), published in the Journal of Personality and Social Psychology. Seventy-five female undergraduates volunteered, three were excluded because they knew the true purpose, and seventy-two watched a ten-minute video, presented as a live feed, of another woman who appeared to receive painful shocks for wrong answers on a learning task. She was a confederate and was not shocked. The observers were then placed in different conditions. Some expected her suffering to continue in a second session and could do nothing about it. Some voted for her to be rewarded and were told the vote succeeded. Some were told the experiment was over. In one condition the victim agreed to continue so that the observers could earn course credit.

Observers who expected the suffering to continue rated her lower than observers whose vote to reward her had succeeded. Ratings were lowest when she seemed to suffer for the observers' sake. Lerner and Simmons read the pattern as evidence of a need to believe that the world is just. The finding is the empirical root of the belief in a just world and the reason the belief is so often described through its least attractive consequence.

In effect

The 2020 meta-analysis by Rael Dawtry, Mitchell Callan, Annelie Harvey and Ana Gheorghiu is the best single guide to how large and how general the effect is. It was the first meta-analysis of derogation experiments and covered 55 studies published between 1966 and 2016, with 5,947 participants in total. The average effect was small. The differences between studies were larger than chance would produce, and the authors note that since 1980 some studies have found derogation, some have found none, and a few have found the opposite.

Two features of the scene predicted which studies found the effect. Studies with more emotionally impactful material, meaning vivid and close to the observer, found larger effects. Whether the event was real or invented made no clear difference. Older studies found larger effects than newer ones, but emotional impact and year of publication are tangled together, so the authors could not fully separate them. Published studies showed larger effects than unpublished ones, and the authors' sensitivity analyses suggest the true average is probably below the pooled figure. Two further experiments in the same paper found that more impactful contexts increased derogation.

Carolyn Hafer and Laurent Bègue's 2005 review in Psychological Bulletin covered the experimental literature since 1980 and described the research as unsystematic in places. It also set out two ways of understanding the belief that the experiments are meant to test. Read together, the review and the meta-analysis describe a finding that is real in some conditions and weak in others, and a literature that did not always keep track of which conditions it was in.

In the founding experiment itself, the pattern also depended on what the observers could do. Observers who could change the victim's fate rated her more kindly than those who expected the suffering to go on without recourse. The effect tracked the situation the observers were in.

What it does not say

It does not say that people generally blame victims. The pooled effect is small, and the variation between studies is large enough that some experiments found nothing. The effect depends on features of the scene, and the strongest results come from vivid, close and emotionally impactful material.

It does not say that belief in a just world is a harsh belief. The same belief is described in the literature as linked to pursuing long-term goals with confidence, and Dawtry and colleagues treat derogation as one way of protecting it. The blaming and the confidence are two outputs of one belief, which is why this page and the page on Belief in a Just World are written to be read together.

It does not say anything about real victims of crime or accident. The experiments use staged or described victims, and the meta-analysis found no clear difference between real and invented events, which is a finding about the experiments and not an account of how people judge actual cases.

It does not rest on a measure of individual belief. The experiments manipulate the scene. The question of how strongly a person holds the belief is a separate research line, and the Personal Belief in a Just World page covers the distinction that Claudia Dalbert drew within it.

It does not carry any health or clinical claim. This page describes ratings of a staged victim in laboratory experiments.

---