UC Berkeley

Men historically and presently hold more power and privilege than women in academia. Men are paid more than women, and their research contributions are valued more than work by women. This power differential also translates to the question-and-answer (Q&A) space, with men participating more than women. The purpose of this dissertation is to investigate gender differences with greater granularity than past research by testing when conference attendees participate and how they use their time. I also investigate psychological correlates (anxiety and fears of backlash) that could explain the gendered patterns of participation. The dissertation ends with an investigation of the influence of structural factors of how Q&A sessions are formatted on gendered participation and as potential mitigating factors for disproportionate participation. Chapter 1 includes an initial investigation into gender differences in participation in question-and-answer sessions and feelings about participating in Q&A sessions. After an exploratory investigation with a small live conference, I found men disproportionately initiate Q&A interactions relative to the gender base rates in confirmatory analyses with a larger conference. However, there were no observed gender differences in other observed behaviors like the amount of time per question, number of points made, or challenging the speaker. In self-reports, women reported feeling less comfortable asking questions, making comments, or approaching speakers and greater fears of backlash. Men were more likely to say that they hold back asking questions to make space for others, whereas women were more likely to say that they hold back asking questions due to anxiety. Reasons why participants held back questions were replicated in a second sample. Chapter 2 tested the explanatory role of fears of backlash in a series of vignette studies and experiments. Men expressed greater question-asking intentions, and women expressed greater fears of backlash across all studies. Fears of backlash significantly explained variance in the relationship between gender and question-asking intentions. Gender differences in fears of backlash were diminished when the audience was described as friendly. However, friendly audiences emboldened men to feel even more comfortable asking questions, maintaining the gender gap in question-asking intentions. In a sample of academics, women’s fears of backlash were more impacted by the framing of the questions they would ask as critical or constructive than by the audience as critical or friendly. An ongoing study testing the effects of the format of Q&A sessions (e.g., chats vs. live) has not found gender effects on question-asking intentions or fears of backlash by format in the data collected thus far. Chapter 3 evaluated the impact of various virtual Q&A formats on disproportionate participation by gender using field data from four conferences. In the three conferences where I had estimates for gender base rates, men took disproportionately more of the total time and word space compared to what would be expected by the gender base rates. This effect persisted when women were in the numeric majority of the conference, when the Q&A was only live video, and when the Q&A was only through chats. Disproportionate participation did not significantly change between live and virtual formats within the same conference one year apart. In an all-chat format, gender disparities in chat text length decreased at higher status levels. In a mixed chat and video virtual conference, there was some evidence that women participated more in chats than in videos. However this cannot be compared to the gender base rates at the conference. Overall, there was greater support that divisional characteristics and particular dynamics within the sessions impacted gender disparities in participation as opposed to reliable effects due to differences in the structures of the Q&A sessions. Overall, I replicated the effect that men disproportionately participate in live Q&A sessions, but that that is where the gender differences in participation seem to end. Other behavioral differences may be driven by the base rates of men participating more than by men being more likely to do them. Women reported feeling less comfortable and having greater fears of backlash, which were largely influenced by how they perceived the community around them. Virtual conferences continued to show disproportionate participation in both video and chat formats. The option for chat messages could potentially be a mitigating element, but the evidence supporting it is weak. The influence of the perception of the immediate audience seems to have a greater influence on gendered participation in Q&A sessions. Across studies and modalities, men’s disproportionate participation and women’s discomfort in Q&A situations was pervasive and stable. More work is necessary to determine how to improve the inclusivity of academic spaces for women.