A rumor is an unsubstantiated story that spreads person to person in a community. These rumors may be true or false and can affect crowd behavior, sometimes for good and other times for ill, such as by prompting riots. Researchers have studied rumors in a wide variety of settings, including disasters, social harm, and the dynamics of information flow.
Rumors often have a low credibility and are spread largely by word of mouth, but they can also be written or broadcasted. Rumors can be very specific, such as “A woman was slashed on the ankles at Jonestown Mall by a thief” or very general, such as “The rumor is that a girl was killed in an apartment building.”
The accuracy of a rumor depends on the motivation and environment of the spreaders. Individuals who spread a rumor with the intention of fact-finding tend to be more accurate. Similarly, individuals who spread a rumor for relationship- or self-enhancement tend to be less accurate. However, rumors that circulate in an environment with high group mechanisms and cultural axioms will tend to be more accurate because they are based on shared beliefs and ideas.
People discuss rumors to make sense of an ambiguous situation or to manage a physical or psychological threat. These rumor-spreading functions usually involve derogating groups with whom one is not associated or making oneself feel more powerful. The ambiguity of a rumor’s veracity may be reduced by the use of evidence, but this is seldom successful in proving the rumor to be true.
The social and psychological dimensions of a rumor are complex, and many theories exist to explain how it works. Some scholars have described three different kinds of rumors: pipe dream rumors, which are reflective of public desires and wished-for outcomes (e.g., the Japanese oil reserves were low and World War II would soon end); reality checks rumors, which are attempts to verify a rumor by citing facts (e.g., a teacher was fired for making off-color comments to students); and speculative rumors, which are based on imagination or wishful thinking.
Other scholars have described a model of the dynamics of rumor spreading that includes the role of an individual’s personality in determining their ability to distinguish between fact and fiction and to determine the credibility of a rumor. This model has been applied to the retweeting patterns of various types of tweets. These z-score values represent the number of standard deviations that each type of tweet is above or below the average value for all retweets. Other models have used a crowd classification approach to determine the effect of rumor spreading, in which the crowd is divided into two categories: radicals and conformers. Radicals are more likely to believe a rumor and share it with others; conformers are less skeptical and slower to abandon a rumor. In addition, these models have examined the effect of gender on rumor-spreading behavior. Women seem to be more prone to believing rumors than men.