Designing a Cohort So the Quiet Founder Is Seen by Week Three

Across 18 programs in three countries, AVODA has learned to ask which founders in the room had actually been heard by week three. We cannot answer that for women and men, because AVODA holds no gender-split figure of any kind.

Before each lecture, a handful of biology instructors in the United States sorted their class list at random in Excel and called names in that order. In their rooms, women were 61% of the class and 60% of the voices heard. In the rooms that waited for raised hands, women were 59% of the class and 37% of the voices. Who speaks in a cohort is decided mostly by the rules of the room, and a program lead can change those rules before session one.

Key Takeaways

In 23 large US biology classes, women were about 60% of students and under 40% of the voices answering the instructor. In the four classes that called names from a shuffled list, the gap closed.
Calling on a group instead of a person reopens the gap, because the group chooses its own spokesperson.
In Uganda, 70% of women and 20% of men do unpaid household work between 5 pm and 11 pm (Time Use Survey 2017/18), so an evening session is already uneven before anyone speaks.
In Uganda’s parliament, women in reserved seats went from being named in debate 96% as often as male constituency MPs (2003) to 51% (2008). Recognition by name can fade within a term.
The Airtime Register gives each founder four columns per session (Spoke, Wrote, Named, Met) and a review after session three, so a quiet founder is noticed while the cohort can still respond.
The evidence moves who speaks. Whether equal airtime also raises sales or survival is still unmeasured, so a program should track that for itself.

A shuffled list against a raised hand

Sarah Eddy, Sara Brownell and Mary Pat Wenderoth studied 23 large introductory biology classes, scoring archived class videos for who was heard. When instructors asked for volunteers, women gave 36.7% of the answers in classes where they were 59.2% of those enrolled, and in nine of the 13 classes the gap was statistically clear. When four instructors used random call, women were 61.0% of the class and 60.0% of the students called.

Four classrooms is a small sample, and the study observed classes as they were taught rather than assigning methods at random. Still, the mechanism is plain. A raised hand rewards whoever is fastest and surest, and a list rewards whoever is on it. The instructors also put a small step before the call: students discussed the question in pairs or small groups first, so the person called had something to say.

A later study at a large research university in the northeastern United States found the step that undoes it. Stepfanie Aguillon and colleagues observed an active-learning course for two semesters, with women at 56% and 57% of the class. Men were over-represented in voluntary answers after small-group discussion in both semesters. When instructors called on a whole group at random, the authors noticed that “Group members quickly negotiated who would speak after a group was selected.” In the second semester men again spoke more than their share. A shuffled list of groups hands the choice back to the group, and in that course the groups’ choices tilted toward men.

Two findings from political science point the same way. Christopher Karpowitz, Tali Mendelberg and Lee Shaker ran experiments with many deliberating groups and found a large gender gap in voice and authority that disappeared under unanimous rule when women were few, or under majority rule when women were many. The right rule depends on who is in the room. In South India’s village assemblies, Ramya Parthasarathy, Vijayendra Rao and Nethra Palaniswamy found women less likely to speak, to set the agenda and to get a relevant answer from officials, and found that a woman village president improved the chance that women were heard. The chair shapes who is heard.

Evening sessions, community meetings and a parliament’s memory

A Kampala cohort adds three pressures a US lecture hall lacks. The first is the hour the cohort meets. A cohort that meets after work, from five or six in the evening, suits people with jobs or shops. An EPRC analysis of Uganda’s Time Use Survey data found that between 5 pm and 11 pm, 70% of women and 20% of men were doing unpaid household work, and 31% of women against 6% of men were caring for children, the sick or the elderly. Picture a founder (an invented example) who leaves at 7:15 to reach a child before dark: she misses the closing discussion every week, and with it most of the facilitator’s attention. The journey home after dark carries its own cost, set out in Who a Founder-Friendly City Is Actually Friendly To.

The second is habit. Afrobarometer’s January 2024 survey of 2,400 Ugandan adults found 49% of women and 64% of men had attended a community meeting in the past year, and 38% of women and 49% of men had joined others to raise an issue. Many women therefore arrive at a cohort with less practice at speaking up in mixed public meetings, a gap of opportunity rather than ability.

The third is memory, and it is the closest measure we have to “invisible by week three.” Amanda Clayton, Cecilia Josefsson and Vibeke Wang counted how often each Ugandan MP was referred to by name in plenary debate between 2001 and 2008. Two years into the 7th Parliament, women in reserved seats were named 96% as often as men in constituency seats. Two years into the 8th, they were named 51% as often. The authors report that the groups had similar backgrounds and qualifications, and they point out that “In order for an MP to be referred to by name, it is first necessary that the others know that MP by name.” A cohort has its own version. By session four, the whole room is likely to know the name of the founder who spoke first in session one.

Rehearsal needs a room built for it

The book under test here is Women in the Technology and Innovation Ecosystem: From Movers to Makers in America (Kaur, Hyunh and Singh, eds., CRC Press). In chapter four, Amy Amann describes returning to work in American tech firms and finding “A gradual low-level compulsion to perform, keep visible, be active, contribute vocally, and demonstrate effort.” Her advice to the returnee is mostly to adjust herself: treat requests for input as a sign that her view is valued, and expect that her “contribution level will rise in concert with greater comfort levels and more team cohesion.” Chapter three’s list of measures for Indian professionals abroad has the same shape: show confidence, work on language skills, socialize more.

That advice fits a salaried professional inside a firm that already expects everyone to speak. It transfers poorly to a Kampala founder whose silence has a timetable and a chairperson behind it. Reframing leaves the cooking hour where it was, and comfort rarely arrives in a room where the same four voices filled the first three evenings. Amann herself points to the better answer: self-advocacy needs rehearsal “in environments where there is no risk and the stakes are relatively low.” A cohort can be that environment, and only its design makes it one.

The Airtime Register

A facilitator makes one decision over and over in every session: whom to call next. The Airtime Register keeps a record of those decisions. Each founder’s name runs down the side, each session gets four columns, and a program assistant or co-facilitator fills them in as the session runs.

Spoke. Did she speak to the whole room at least once? Whole-room questions go to individuals from a shuffled name list, with groups and raised hands kept for other moments. Shuffle a fresh list each session in a spreadsheet and strike names as they are called, so every founder is called before anyone is called twice.
Wrote. Did she write an answer before the discussion? Every whole-room question starts with two minutes of writing on a card or in the cohort’s WhatsApp thread. The facilitator reads the cards during the break.
Named. Was her idea referred to by her name, by the facilitator or a peer? When the facilitator walks between tables during group work, she takes one idea from a quieter table and introduces it to the room with its owner’s name. She might say, for instance, that Grace’s table noticed their customers pay on the 28th (the name and the date are invented for this example).
Met. Did a facilitator speak with her one-to-one, even for three minutes at the door?

After session three, before session four, the lead reads the grid. Any founder with two or more empty columns across those three sessions gets a call within 48 hours. The call asks about the timetable before confidence: when she has to leave, whether the session language suits her, whether someone in her group talks over her. A later start, a session moved to Saturday morning, a mixed group regrouped, or a short summary in Luganda are ordinary adjustments.

Composition changes which rule works. Where women are a small minority of a mixed cohort, use rounds in which every member speaks before any decision, which is the unanimity condition in the Karpowitz study. Where women are most of the room, simple majority discussion is less risky. Put one woman facilitator in the chair for some sessions, because the South Indian evidence suggests the chair changes who is heard.

The fear of being called on

Katelyn Cooper, Virginia Downing and Sara Brownell interviewed 52 students in large active-learning science courses about what made them anxious. Cold call and random call, students said, only raised it. The authors traced this to fear of negative evaluation, the dread of being judged in front of peers, and argued that repeated exposure reduces that fear only when students have a positive experience. They recommend a different practice: instructors walk around during group work, gather ideas and share them with the whole class.

A shuffled list run carelessly can turn a quiet founder’s evening into dread, and a founder who dreads Tuesday is likely to stop coming. The evidence is also American and comes from undergraduates, so it describes perception rather than measured outcomes.

Cooper’s own tools are already in the register. The Wrote column means every founder has thought a question through before her name is called, which is how Eddy’s instructors used random call. The Named column is the walk-around practice the authors recommend, made routine and credited by name. The call after session three gives a founder a private channel before any public one. The list stays, because the raised hand has a measured record of leaving women out of the conversation.

The crowd on the road to Jairus’s house

Three findings from this essay meet on the road to Jairus’s house in Mark 5: attention goes to the nearest and most urgent claim, standing in a room is built by being named, and being called on carries fear. The crowd presses in, and the request in front of Jesus is urgent and legitimate, because a synagogue leader’s daughter is dying. Somewhere behind him a woman who has bled for twelve years touches his cloak and plans to leave without a word. Jesus stops and asks who touched him, and his disciples answer with the efficiency argument every busy facilitator knows: “You see the people crowding against you.” He “kept looking around.” He is in a hurry for a good cause and still spends the time to find the one who stayed quiet. The woman comes forward “trembling with fear” and tells “the whole truth,” and he meets that fear with belonging before anything else. The order on that road runs seen, then named, then heard, and the first word she hears from him is “Daughter.”

Frequently Asked Questions

How do I get quiet founders to speak in a cohort session?

Ask every whole-room question to individuals from a shuffled list, give two minutes of writing first, and carry ideas from quieter tables into the room with their owners’ names. Then review who has spoken by session three and call anyone missing.

Is random calling fair to founders who are anxious about speaking?

Random calling raises anxiety when it is a surprise. Interview research found students described cold call as only stressful. Reduce that by letting founders write or discuss in pairs before anyone is called, by warning the room that everyone will be called once, and by treating the first answer as practice.

Why do women founders speak less in mixed cohort sessions?

Evidence points mostly to the room’s rules. Volunteer-based questions favor quick, confident speakers, and in Uganda evening sessions overlap with hours when 70% of women do unpaid household work. Afrobarometer also finds fewer Ugandan women than men attend community meetings, so many arrive with less practice.

How should a facilitator credit an idea to the founder who raised it?

Carry the idea into the room with its owner’s name attached, in her words where possible, for example telling the room what Grace’s table (an invented name) noticed about when customers pay. Do it during group work, from quieter tables first, and record it in the register’s Named column so the credit is tracked across sessions.

What is the Airtime Register?

It is a session grid with four columns per founder: whether she spoke to the whole room, wrote an answer, had her idea named, and met a facilitator one-to-one. After session three the lead reviews it and calls anyone with two or more empty columns before session four.

Next Step

Draw the register for your next intake: names down the side, Spoke, Wrote, Named and Met across the top, and the session-three review date in the corner. Bring that grid and the draft session plan to a strategy call with Joshua to settle the question rules and who makes the review calls. You leave with a sheet your co-facilitator can fill from session one and a named list of founders to phone before session four.

Sources and Evidence

Women in the Technology and Innovation Ecosystem: From Movers to Makers in America (Kaur, Hyunh and Singh, eds., CRC Press, 2026), publisher page: chapter 4 (Amy Amann, pp. 50, 55 and 57) and chapter 3 (p. 38), the three quotations and the adjustment advice. DOI 10.1201/9781003604402.
Eddy, Brownell and Wenderoth, “Gender Gaps in Achievement and Participation in Multiple Introductory Biology Classrooms”, CBE Life Sciences Education 13(3), 2014 (full text read via Europe PMC): 23 classes; volunteer answers 36.7% women against 59.2% enrolled; random call 60.0% against 61.0% in four classes.
Aguillon et al., “Gender Differences in Student Participation in an Active-Learning Classroom”, CBE Life Sciences Education 19(2), 2020 (full text read via Europe PMC): men over-represented after small-group discussion; groups chose their own speaker under group random call.
Cooper, Downing and Brownell, “The influence of active learning practices on student anxiety in large-enrollment college science classrooms”, International Journal of STEM Education 5, 2018 (full text read via Europe PMC): 52 interviews; cold and random call only increased anxiety; fear of negative evaluation; the walk-around alternative.
Karpowitz, Mendelberg and Shaker, “Gender Inequality in Deliberative Participation”, American Political Science Review, 2012, abstract: the gap disappears under unanimous rule with few women or majority rule with many.
Parthasarathy, Rao and Palaniswamy, “Deliberative Democracy in an Unequal World: A Text-As-Data Study of South India’s Village Assemblies”, American Political Science Review, 2019, abstract: women less likely to speak, set the agenda and get a relevant response; women village presidents improve the chance women are heard.
Clayton, Josefsson and Wang, “Present without Presence? Gender, Quotas, and Debate Recognition in the Ugandan Parliament”, Representation 50(3), 2014: name recognition in plenary, 2001 to 2008; 96% in 2003 and 51% in 2008.
Afrobarometer Dispatch AD881, “Ugandan women trail men in education, jobs, asset ownership, and political engagement”, October 2024: 2,400 adults, January 2024; community meetings 49% against 64%; raising issues 38% against 49%.
Mwesigye, “An Empirical Investigation of Gender and Time Use in Uganda”, EPRC Occasional Paper 49, 2019: Time Use Survey 2017 data; between 17:00 and 23:00, unpaid household work 70% of women and 20% of men, unpaid care 31% and 6%.
Mark 5:21 to 34 (NIV): Jairus, the crowd, and the woman who touched the cloak.