Why Dominant People Get Influence: Speaking First Reads as Competence, Even When It Isn’t

By Noam Shemla · The Studies

TL;DR Anderson and Kilduff put strangers into four-person groups and watched who came to lead them. People who described themselves as dominant gained more influence, because their groups saw them as more competent. In a maths task with right and wrong answers, the dominant members were not actually better at maths. What they did was answer first, and groups nearly always went with the first answer offered.

  • Paper: Anderson, C., & Kilduff, G. J. (2009). Why do dominant personalities attain influence in face-to-face groups? The competence-signaling effects of trait dominance. Journal of Personality and Social Psychology, 96(2), 491–503.
  • Participants: Study 1: 68 students in 17 groups. Study 2: 100 students in 25 groups. Each group had four unacquainted members of the same sex.
  • Tasks: Study 1: 45 minutes inventing an organisation and its strategy. Study 2: 30 minutes solving maths problems from the GMAT, the business-school admissions test. The best group each time won $400.
  • Measures: Self-rated dominance before the task; after it, each member’s influence and competence as rated by teammates, by outside observers from video, and by research staff; in Study 2, coded behaviours, answer accuracy and SAT maths scores.

Some people take charge of every group they join, and the groups usually let them. Cameron Anderson and Gavin Kilduff, then at the University of California, Berkeley, asked why. Most theories of groups say influence has to be earned with ability; simply being assertive should not be enough. So why do dominant personalities keep ending up in charge?1

Their answer was that dominant people behave in ways that make them look competent, whether or not they are. The two studies were built to test that and to rule out the obvious alternatives.

Two studies, three sets of judges

Before each group met, members rated how dominant, assertive and forceful they were. After the task, everyone rated everyone else on influence and on competence. So did two further sets of judges who watched the video: students from the same pool and members of the research staff, trained to rate carefully. Using several judges matters because a group might simply decide, after the fact, that whoever ran things must have been the most able. Outsiders watching the tape have no such hierarchy to justify.

In both studies, dominance predicted influence to almost the same degree, even though one task was open-ended and creative and the other had right and wrong answers.

The paper tested each indirect path with a Sobel test, the standard check at the time for whether an effect travels through a middle step.

What dominant people actually did

In Study 2, a coder who did not know the hypotheses counted what each person did in the first ten minutes. The reason for the first ten: earlier research had found that people who contribute early are the ones likely to emerge as leaders.

Speaking first mattered for a simple reason: in 171 of the 182 problems coded, 94%, the group went with the first answer anyone offered. There was little debate. Whoever spoke first effectively decided.

Seen as competent, without being more competent

The decisive test used the maths task’s one great advantage: there was a right answer, so actual ability could be measured. The authors had two measures of it, each person’s SAT maths score and how often their answers in the session were correct.

The authors’ conclusion: dominant individuals gained influence partly because they acted in ways that made them seem more competent, despite not being more competent than their teammates.

Has it held up?

The central idea, that confident behaviour is read as competence and converts into status, has been developed and extended since, including by the same research group. In six studies, Anderson and colleagues found that overconfident people, who believe they are better than they are, gained higher status in groups because their overconfidence produced behaviour that made them look competent.2

Other work has asked whether dominance is the only route. Cheng and colleagues distinguished dominance, gaining rank through force and intimidation, from prestige, gaining it by sharing expertise. Both produced influence over unacquainted groups, as judged by members and outside observers and in a behavioural measure, and both drew more of others’ visual attention. The difference was in how people felt about them: those using prestige were liked, those using dominance were not.3

What this study cannot tell you

  • It is two studies of students, in groups of four strangers working for well under an hour. Groups that last, where ability becomes visible over time, may correct the early impression.
  • Dominance was self-rated, with three adjectives. The behaviour coding supports it, but the trait measure itself is brief.
  • The key behaviours were coded in one task only. Answering first means something clear with maths problems; in an open discussion, the equivalent is harder to count.
  • It is correlational. Nobody was assigned to be dominant, so other differences between dominant and less dominant people could contribute.
  • Some of the behaviour correlations are modest, and one of them (giving relevant information) sits at the edge of significance.

What a speaker can take from it

Groups make up their minds early, and they often take the first clear answer on offer. This study shows how much that rewards whoever speaks first, even when the first answer is no better than the others. That is a lesson in two directions.

If you know the answer, say it early and plainly; waiting to see what others think hands the decision to whoever speaks before you. And if you are running the discussion, protect the room from the first-answer rule: ask for several views before anyone commits, especially from the people who have not spoken, because speaking first is a signal of confidence, not of being right.

This article summarises a published paper, read from a posted copy of the journal article. Every correlation, coefficient, count and sample size is the authors’ own, from the results of both studies and Figure 3. The 95% confidence intervals are ours, computed from the reported correlations and degrees of freedom.

Measured

  • Self-rated dominance before the task
  • Influence and competence as rated by teammates, outside observers and research staff
  • Counts of behaviours in the first ten minutes, answer accuracy, and SAT maths scores (Study 2)

Inferred, not measured

  • That perceived competence carries dominance into influence: mediation analysis, consistent with a cause but not proof of one
  • That dominant people are not more competent: shown for maths ability in Study 2, not for every kind of skill

References

  1. Anderson, C., & Kilduff, G. J. (2009). Why do dominant personalities attain influence in face-to-face groups? The competence-signaling effects of trait dominance. Journal of Personality and Social Psychology, 96(2), 491–503. https://doi.org/10.1037/a0014201
  2. Anderson, C., Brion, S., Moore, D. A., & Kennedy, J. A. (2012). A status-enhancement account of overconfidence. Journal of Personality and Social Psychology, 103(4), 718–735. https://doi.org/10.1037/a0029395
  3. Cheng, J. T., Tracy, J. L., Foulsham, T., Kingstone, A., & Henrich, J. (2013). Two ways to the top: evidence that dominance and prestige are distinct yet viable avenues to social rank and influence. Journal of Personality and Social Psychology, 104(1), 103–125. https://doi.org/10.1037/a0030398