In-person conferences exasperate inequalities in academic communities (Sarabipour et al., 2020). Increasing the diversity of research environments is vital for scientific innovation and will increasingly be the focus of research and funding institutions (ERC, 2020; Wellcome, 2020).

Undergraduate trainees, graduate students, postdoctoral researchers, and female academics have fewer economic means and travel less than tenured professors and male colleagues to attend in-person conferences (Wynes and Donner, 2018; Jöns, 2011). Since virtual meetings have lower or no registration fees, they are open to a much wider variety of researchers. Having researchers from diverse backgrounds and life experiences at scientific meetings brings more varied perspectives and creativity towards addressing complex scientific problems.

Enabling presentations in multiple languages with captioning and translation will further encourage scientific exchange across nations. Additionally, providing closed captions and transcripts of virtual talks will further assist attendees with hearing and visual disabilities.

Online conferences have allowed much higher levels of participation of researchers from all scientific sectors internationally – and, in particular, researchers from underrepresented groups, researchers from countries with low-to-middle income economies, and early-career researchers (Figure 1, Figure 1—figure supplement 1; Figure 2; Figure 3; Figure 4). Well-planned virtual conferences can also significantly increase the diversity of speakers, increase representation of historically marginalized researchers, and help to achieve gender parity (De Saá‐Pérez et al., 2015). A higher number of attendees also means presentation of diverse research, viewpoints, discussions and ideas, which will benefit all researchers, since diversity brings excellence to scientific exchanges (De Saá‐Pérez et al., 2015; Nielsen et al., 2018).

Figure 2 with 1 supplement see all

Download asset Open asset

Figure 3 with 2 supplements see all

Download asset Open asset

Figure 4

Download asset Open asset

Unfortunately, conferences in many disciplines are still heavily dominated by speakers who are male and/or from economically well-off nations (Sarabipour et al., 2020). Co-organizing conferences and delivering conference talks, particularly as an invited, featured, plenary or keynote speaker, are major career milestones that enhance the national and international scientific profile of any researcher. As such, passing over women and other underrepresented researchers for speaking and organizing engagements harms their careers. Concerted efforts are required to create equity for attendees in featured speakers, session chairs, and scientific and organizing committee member roles (Sarabipour et al., 2020). If representation is not increased at these positions and opportunities and the current culture persists, parity at funding rates, award nominations, early-career researcher (ECR) hiring patterns or promotion from junior to senior faculty levels and fair treatment of women will be less likely in the long term. Online conferences provide an unparalleled opportunity to increase representation.

Virtual conference platforms can provide better file sharing, data presentation, and interactive visualization formats for the speakers and attendees. For instance, the 2020 International Conference on Learning Representations provided interactive figures and animations (ICLR Organizing Committees, 2020), which are often not supported by the platforms used to host conference abstracts or peer-reviewed articles. The virtual format also allows participants to comment on and exchange computer code, workflows, and other interim research products while the conference is ongoing. Digital libraries of the abstracts, talks, slides, and posters further provide the opportunities to attend and watch the entire conference including parallel oral and poster sessions. This ability to share and present data more efficiently will accelerate scientific progress as researchers can immediately access new information that they can use in their own research.

Further, virtual conferences attract record numbers of researchers from all continents and time zones (Figure 1, Figure 1—figure supplement 1, Figure 3, Figure 3—figure supplements 1 and 2). Researchers who have to cope with socioeconomic, geographic and/or physical limitations, and researchers with family commitments and responsibilities can all attend virtual events. Additionally, low or nil registration costs have meant that more researchers can attend multiple virtual conferences per year. This increases the number of people with access to new research and ideas, which will lead to more innovation in more countries compared to in-person conferences.

The open and global setting of virtual conferences enables wide inclusion of demographics and productive discussions that can lead to long-term collaborations among researchers of different career stages. The virtual setting also allows for better training opportunities for ECRs, by increasing opportunities to present their research.

Low registrations costs enable a higher number of ECRs to attend virtual conferences (Figure 2, Figure 2—figure supplement 1), allowing them to take advantage of interactions with more senior researchers. These interactions can lead to collaborations and/or to learning about potential studentships, internships, postdoctoral opportunities or faculty positions.

Well-planned conferences need a good ratio of early-career researchers to senior scientists acting as speakers or session chairs. In in-person conferences, this often means that not everyone who submits a quality abstract can give a talk, and it is usually ECRs who miss out on valuable speaking time and exposure. Virtual conferences can solve this issue by allowing participants to watch talks asynchronously or by being organized over a period of time, which allows most submitted abstracts to be accepted for oral presentations.

Besides increasing scientists’ opportunities to participate in conferences, online discussion platforms during conferences can also help establish long-term collaborations and mentorship relationships between scientists in different career stages, including employment and sponsorship opportunities for ECRs.

Finally, peer-mentorship and collaborative peer-review of conference abstracts and other research products during virtual conferences will increase discussion and transparency on research, potentially improving the reproducibility of the research presented.

Virtual conferences allow organizers to gather participant feedback and data on demographics, audience engagement and levels of networking, which would be difficult or impossible to track during in-person meetings.

This information, along with data on attendee and speaker career stage and awardee and honoree status is highly valuable (Le et al., 2020) and can be used to improve conferences. A few steps to do this are: (i) conducting pre- and post-conference surveys to receive attendee feedback; (ii) reporting speaker line-ups; (iii) enabling diversity data to be self-reported by attendees; (iv) reporting statistics about the demographics of speakers to assess equity, inclusion and diversity (EDI); and (v) using anonymized data to uncover specific diversity baselines and act to increase EDI.

These steps will help conferences to: (a) better understand research communities in specific disciplines; (b) set and share policies to improve EDI practices across multiple continents; (c) address disparities in gender, country of affiliation, ethnicity and race, language and career stage; (d) assess connections, networking and collaborations; and (e) evaluate the impact of academic gatherings on research and research culture.

Improvements do not have to rely just on self-evaluation: conferences can also influence each other. Accessibility solutions, lessons learned, and improvements in policies at one conference and country can be shared with and adopted by other meetings, making it essential for conference organizers to transparently share policies, measurements, language, and standards.

Science is a global endeavor. Academia must recognize the professional privilege required to attend in-person conferences, and reformulate policies, attitudes, and funds towards organizing virtual conferences with a view toward achieving an equitable system for women, minorities and all researchers worldwide.

The shift to virtual conferencing prior to, and increasingly during, the COVID-19 pandemic has provided an unprecedented opportunity to reform methods of organizing academic conferences in all disciplines. The potential benefits of scientific conferences have been underutilized in the in-person format, and can be further explored in virtual formats. This year, researchers from around the globe and across the full spectrum of research and career stages gained access to innovative research and opportunities to engage and connect. The fully virtual meeting structure enabled unprecedented levels of interactions and exchange of scientific ideas and discussions online.

It is crucial that scientific societies and conference organizers examine the data generated, the actions taken and the outcomes of conferences, and then pool resources from private and public funding organizations to further improve these gatherings and accelerate changes in research culture. To do this, it is particularly important to consider the needs and constraints of meeting participants, and ECRs in particular. These efforts will have both short- and long- term impact on research and research culture. Curating the outputs of conferences (such as code, protocols, workflows, abstracts and manuscripts) in open digital libraries will increase access to scientific research. A culture of open digital communication, group reviews and live national and international discussions will strengthen scientific bonds among researchers in all disciplines. It can also help to improve transparency and reproducibility of research findings to accelerate the rate of scientific and medical discoveries. Broad and speedy access to scientific knowledge worldwide will improve research, health, environmental and economic conditions in all countries for years to come.

The data used for visualization was pulled from multiple sources. Existing data on in-person conferences were sourced from the internet via an in-person conference database curated by the author at https://elifeambassadors.github.io/improving-conferences/, the conference and respective scientific society websites, and Google search. Aggregate number of attendees and representing countries were collected for the years that data was available. Data on country specific and state specific attendance was requested for in-person conferences of 2019 and virtual conferences of 2020 from scientific societies and in a number of cases were provided to the author. Data on the 2019 gross minimum wage (full amount an employer pays before taxes and other deductions are withheld) minimum annual (earned for 12 months in US dollars) wages worldwide was sourced from the world bank database at https://data.worldbank.org/indicator/PA.NUS.PRVT.PP. The latest available data on the number of researchers per million inhabitants worldwide was sourced from the UNESCO statistics institute database (https://en.unesco.org/node/252277). The number of accredited academic institutions in the United States was sourced from (https://catalog.data.gov/dataset/colleges-and-universities-cb8a4). A table with the acronyms for the different conferences discussed in the article is available in Supplementary file 1.

All figures were made with R and the ggplot2 package (Wickham, 2016), with colors from the RcolorBrewer package (Neuwirth, 2014). Code and aggregate anonymous raw data on attendee counts and represented countries and career stage statistics underlying all figures is available on (http://doi.org/10.5281/zenodo.4044404).

No external funding was received for this work.

The author thanks meeting presidents and organizers Drs. Alexander Rush of Cornell University, Lee Campbell of Salk Institute, Elina Zúñiga of University of California San Diego, Carla Rothlin of Yale University, Andrea Armani of University of Southern California, Orad Reshef of University of Ottawa, Riccardo Sapienza of Imperial College London, Rachel Grange of ETH Zürich, Konrad Kording of University of Pennsylvania, Antoni Ribas of the University of California Los Angeles and the American Association for Cancer Research and Warren Ko of University of California San Diego, the Genetics Society of America, American Diabetes Association, American Society for Nutrition, American Society for Mass Spectrometry, American Society of Clinical Oncology, Society for Imaging Informatics in Medicine, American Physical Society, Society for Developmental Biology and the International AIDS Society for their efforts in organizing virtual conferences, providing anonymous aggregate attendee statistics on their meetings in support of this work. The author thanks Drs. Richard Sever of Cold Spring Harbor Laboratory, Natalie Niemi of Washington University in Saint Louis and Alexandre Bisson Filho of Brandeis University, United States and Humberto Debat of National Institute of Agricultural Technology, Argentina, for their valuable comments on the manuscript.

1. Received: September 1, 2020
2. Accepted: October 28, 2020
3. Version of Record published: November 4, 2020 (version 1)

© 2020, Sarabipour

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.