The "Queen Bee" phenomenon - which suggests that women who succeed in male-dominated settings mistreat subordinate female professionals - may be a myth, a study has found. Portrayals in the media and academic research suggest that females act like queen bees, when they succeed in their careers - contributing to gender inequality in the work place. The phenomenon is widely documented in the world's press. For example, Margaret Thatcher, the UK's first female prime minister, received the "queen bee" label for not promoting the careers of other women in her cabinet. "Previous research on the queen bee phenomenon stems from illustrative case studies that are not representative or surveys that do not establish the true causal effects of appointing women to power," said Paulo Arvate, professor at the Sao Paulo School of Business Administration in Brazil. "These studies have reinforced the stereotype that women do not make good leaders," said Arvate, lead author of the study published in The Leadership Quarterly. Researchers said that the "Queen Bee" phenomenon may be a myth. In environments where top leaders are afforded more power and discretion, female leaders act benevolently over their subordinate women. They chose more subordinate females at high managerial levels, which also reduced pay inequality relative to men in similar roles. The researchers' study took into account 8.3 million organisations distributed over 5,600 Brazilian municipalities. From this sample men and women who had won an election race with a very small margin of victory were compared to mimic a randomised experiment. The study took account of time, to allow for leaders to solidify their power and impose their will, and finally whether the leader was from a public or private organisation based on the assumption that public leaders command more authority. The results showed that when a woman was elected leader there was in fact an increase in the number of women occupying top- and middle-management positions in public organisations. "Our research has many methodological advantages relative to previous research and presents an entirely different picture," Arvate said.