In research fields such as computer science, academic publication mostly happens at two kinds of venues: peer-reviewed journals, and peer-reviewed conferences. The main difference is that conferences usually require that some author presents the work at the conference and pays a registration fee for the conference, whereas journals do not.

One criticism of conference in this light is that they are not open to all academics: they exclude researchers in poorer countries and institutions which cannot afford the conference registration fee or trip, they discourage geographically remote authors for which attending a conference is complicated, they discourage people who for personal reasons cannot easily travel (caring for a child or family member, having a disability, etc.).

My question is: has there been any systematic study to estimate the magnitude of this effect? E.g., take the dataset of article metadata for articles and journals of various fields, and study if there are systematic differences in terms of the author's institutions, countries, gender, etc.

  • Even authors who are next door to the conference may not be able to go - can depend on funding and advisor, so it may not be distance...
    – Solar Mike
    Apr 17, 2020 at 10:59
  • Many conferences have avenues to support attendance from disadvantaged countries in an attempt to ameliorate their situations. They, and those who have difficulties traveling (times in life many of us have experiences), are also well served by having co-authors who can travel and present the work. But, I have seen no systematic study, and suspect there are many confounding parameters that would make it difficult to determine much beyond the fact that it likely is an issue.
    – Jon Custer
    Apr 17, 2020 at 14:29
  • Sure I guess it's noisy, but already determining that this is an issue (or that there are other factors at play which make some other unexpected difference in the distribution, e.g., age, nationality, etc.) would be interesting. I wonder if someone has done it.
    – a3nm
    Apr 17, 2020 at 18:28

2 Answers 2


My relatively quick Google and Google Advanced searches didn’t identify empirical studies that address the question you ask. However, a search using the terms [scientific conference presentation] and [diversity] yielded this article discussing the need for actively working to increase diversity among conference presenters: https://physicsworld.com/a/fifteen-tips-to-make-scientific-conferences-more-welcoming-for-everyone/

This article does not address your question directly, but it makes me think that framing your question in terms of a comparison between authors of published papers and authors of conference papers may not be the most productive approach to finding relevant literature.

The majority of researchers who have published in peer-reviewed journals have also presented their work (sometimes the same work) at conferences. By making the author/presenter the unit of analysis, one risks substantial overlap in the two groups—-very likely washing out any real differences that exist. “Noisy” data are problematic in their potential to overlook effects as well as to suggest them erroneously.

I think that a reasonable answer to your question would be more likely to come from separate examination of the two avenues (journals vs presentations)for the diversity of their authors.

In 20+ years’ experience with both conference presentations and peer-reviewed publications, I note barriers to diversity in both venues. It is certainly true that attending a conference takes time and money that many researchers lack. However, research of sufficiently high quality to obtain inclusion in peer-reviewed journals typically comes from well-funded labs to which only a small percentage of the academic world has access. Biases in admitting students into graduate programs, biases in hiring academic/research personnel, and biases in the awarding of research grants guarantee that both journal and conference avenues will have less than optimal diversity. I think that separate investigations of the two would be very interesting.


The main challenge for those living in countries with low financial support in academia is not that they cannot pay the conference registration fee.

In most cases they actually can pay such fees by several approaches: Many theory conferences provide financial supports for Ph.D. students or those who are living in the 3rd world countries. Or in a worst case, the participant can ask for payment from the corresponding authority in their country and since this is nothing for a nation they pay for it (if it is a decent conference). Of course for the reasons that I will explain in the following, they usually cannot get into good (computer science) conferences, and therefore they go for (lower level) journals. But even there, their publications are usually way less than their peers that are living in developed countries (even if they have the same country of origin).

The major problem for people living in such countries is that they cannot have a reasonable academic life. For instance, a professor has a very limited budget to support his Ph.D. students (almost nothing), thus, students have to work and study Ph.D. at the same time, which drastically reduces their performance. Even income of many professors is not enough to live a decent life in their own country and they also have to engage with industry or other occupations to make profit out of it, hence after a while it makes them a completely different person, a non-academic person.

Therefore if there is a lack of paper from such countries in good conferences, it is not because they cannot pay for the conference fees, this one or two time payment per year, contributes very little in their scientific outcome, as explained above the main issue is the entire structure of the country that pushs back science.

  • Thanks! This is a reasonable point, but does not really answer my question.
    – a3nm
    Sep 23, 2020 at 10:25
  • I believe it perfectly answers your question. I explained that the assumption you make is wrong in the first place (they can pay to participate); second, to elaborate it further consider this: why there aren't many studies to check the effect of the common cold in the death of patients with brain cancer? Answer: one with brain cancer has a more important reason to die than the common cold, if the system immunity goes down one will die for whatever mild sickness, so, it is not relevant to study that topic. Actually, such questions are more like attempts to downgrade the actual cause. Sep 23, 2020 at 14:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .