A Neglected Pipeline: Improving Undergrad Access to Conferences

Conferences represent a great opportunity to improve our field through exposure to undergraduate students — but there are so many barriers in their way.

Academic conferences are amplifiers. We don’t often talk about them in those terms, but there’s a reason why I feel like I’m missing out when I don’t attend. I’ve benefited from the people I’ve befriended, the ideas I’ve been inspired by, the companies I’ve talked to, and the presentations I’ve given.

And these benefits stretch far beyond my research. I think it’s fair to say that my career has been advanced just by showing up to conferences… so it makes me carefully reflect on who doesn’t have the same luxuries.

Conference access is a wide-stretching problem impacting many people, but in this post I want to focus on undergraduate students… because they are our future graduate students, our future professors, and our future industry leaders. But they can’t get into our conferences.

This post is a story in three acts:

• ACT 1: Why we should care about undergraduate students.
• ACT 2: Making invisible barriers visible.
• ACT 3: How to improve undergrad participation.

Note: This post focuses primarily on undergrad attendance in CS conferences… but many of the challenges are broadly applicable to other disciplines

ACT 1: Why we should care about undergrads

Because talent comes from more than resource-rich universities.

Most undergrads that attend our conferences come from either research-heavy or resource-rich universities (and often both). I won’t rehash the importance of expanding the quality and capacity of our CS pipeline, but we should recognize the role that conferences have in captivating interest and fortifying connections for future graduate students. Right now, we’re not drawing graduate students from everywhere. We’re drawing from a small, select crowd who have access to these resources.

Who are some of these people we’re missing? Here is a 57-page(!) spreadsheet of primarily undergraduate institutions in the United States alone. There are (many) thousands of students that attend these universities. They are not without motivation or intelligence, they are without faculty researchers to work with or without funds to explore our conferences. While there are notable exceptions (elite liberal arts colleges send a proportionally high number of students to grad schools), too often doors are closed because of institutional factors beyond student control. And it’s not just undergrad institutions. It’s most institutions.

Because we believe that inclusion is important.

If we implicitly prioritize some universities (resource-rich) over others (for example, community colleges), it fundamentally impacts the racial, socioeconomic, and geographical representation at our conferences. We can’t complain about the lack of diversity and inclusion in our Ph.D. programs or our faculty bodies without examining the role that our conferences have in amplifying students towards those programs.

Because we care about early career faculty.

Have you talked to early career CS faculty lately? It’s a recruiting nightmare. And I worry that we do ourselves no favors by throttling undergraduate participation in conferences — a career-altering experience for many of my own students. The fewer undergrads that engage with our field, the fewer Ph.D. students we’re going to see in our field. This is bad for our undergraduates, bad for our early career faculty, and bad for our field.

Computer Science conferences are too often neglected as an amplifier of undergraduate students. They have been career-defining events for the students I’ve sent. But what about the students who don’t go? We need to reflect on the barriers to undergraduate participation in conferences.

ACT 2: Making invisible barriers visible

So if undergraduate participation is important, why is it currently bad?

• Presenting anything is hard, and especially hard for undergraduates.
• Funding is hard, and especially hard for undergraduates.

Why Computer Science conferences make it hard to present.

While most science disciplines have low presentation barriers for their academic conferences, that is not the case for CS. Presentations require full papers, which are highly vetted research projects. Even worse, the prestige model has a trickle-down effect. Poster sessions are now populated by mature research that simply didn’t make the first round of paper cuts.

There are almost no low-barrier entry mechanisms to many CS conferences. Undergraduate presentations at our conferences necessitate extraordinary effort and extraordinary support.

Why institutions make it hard to engage with research.

Since CS conferences require a deeper research involvement, we need to reflect on the processes that prevent even our most motivated undergraduates from engaging with research. Anna Meier recently posted a wonderful thread reflecting on these challenges, and I will mirror many of those here.. .

1. Most faculty have no incentive to meaningfully include undergraduates in their research. Successful undergraduate research requires immense faculty time and mentorship. For most faculty, this investment doesn’t help them keep their jobs. Early career professors are nudged to (wisely) align with the definitions of success of their universities — definitions that usually don’t include undergrads. When faculty do invest time in undergraduate research, our measures of research productivity push attention towards students who come in already prepared to do research (only amplifying the systemic advantages they were given before college).

2. Of course, all of this assumes that students have access to research opportunities in the first place. For a host of reasons, many universities are simply not very active in research.

3. Of the institutions that do run research for undergraduates, the pay is below (sometimes very very far below) even modest industry standards. This creates significant barriers for students without a financial safety net.

These forces are largely out of the control of our conferences, but they are critical to consider when we reflect on why it’s important to open new pathways to participation.

And yet, some students overcome all these barriers!

They are independent and motivated and brilliant. And they even manage to co-author publications at prestigious CS conferences. Surely, the hard part must be over, right? Well, not quite…

Conferences are expensive. Funding often ignores undergrads.

There isn’t enough travel funding to begin with, and most conference funding heavily prioritizes graduates students. The rising cost of conferences have created financial hurdles not just for undergraduates, but everyone. As a result, travel grants tend to prioritize the people we perceive who need it most. So does student volunteer selection. With only rare exceptions, undergrads get pushed off the priority queue. Undergraduate sources of funding are few and far between.

Why institutional funding procedures make it hard to attend.

What about other pools of money? Well…

1. Institutions don’t have the money. Again, conferences are expensive. Higher ed budgets are a mess. The money just isn’t there.
2. For many institutions that have the resources, funding is contingent on presenting work. Given the lower presentation barriers in most other academic fields, this makes sense. But it hurts undergraduate CS students.
3. Faculty have little incentive to fund undergrad travel. Speaking generously, faculty need to focus funds on their grad students — not undergrads who will go to some other lab in a year. That being said, I have been incredibly disappointed to see some resource-rich universities balk at funding undergrad travel, even when students co-author a publication.

ACT 3: How to improve undergrad participation

I want to end with good news and ideas. First, the good news:

• Most of us are philosophically aligned: My conversations with other professors suggest that many of us value undergraduate exposure to conferences. Below, Anna hits the nail on the head for why I believe this is an important issue… and I’ve found most conference organizers agree.

• There are opportunities for resources: Students want Ph.D. advisors and jobs. Ph.D. advisors want students. Our corporate sponsors want job applicants. Grad schools and industry sponsors should both be incentivized to financially invest in measures that attract our bright, motivated undergrads who are looking for a job (they might be the most available group in our conferences).

Now some ideas…

For undergrads who engage with research, make sure that conferences have low-barrier presentations.

Presentations are the key to unlocking whatever funding pools do exist.

• Make a separate track for undergraduates. Understand that their access and constraints differ from grad students and professors. Putting undergrads in the same pool as seasoned professors limits access.
• Rethink how we evaluate non-archival content. What are the purpose of posters? Are they to provoke new ideas? Are they to share not-quite-ready research studies? Are they to give completed (but rejected) research studies a platform? Our decisions here open or close doors.

Create opportunities for undergrads who may not have the privilege to engage with research

If we really want to see a more diverse group of students, we need mechanisms that make curiosity, not research, the prerequisite to attend.

• Make a mentoring symposium for undergrads: Venues like ACM CHI often have mentoring symposiums, but they tend to focus on graduates students and early career faculty. There, undergrads could present (unlocking internal funding!) without having to overcome many of the research barriers described above.
• Improve access for local students: Hotels and airfare are big financial hurdles, but I see no reason why we shouldn’t have a stronger presence from local students. Reduce registration costs for local students who otherwise might not have any access to our disciplines.

While there isn’t an undergrad slot here, I think SIGCSE does a great job at (1) keeping conference fees low, and (2) recognizing that different groups have different resources.

• Reflect on who travel funds & volunteer slots should support: We should decide whether we want to support undergraduates or not. If we do, we can’t throw them in the same application pool as graduate students (who naturally have more developed CVs, recommendations, etc.). We also need to promote these funding opportunities to schools beyond the traditional R1s or top liberal arts colleges.

Advocate for local change at our universities

• Just as we advocate for universities to consider conference publications alongside journal publications in CS, our funding mechanisms should also be sensitive to the presentation challenges in CS. Co-authoring a paper should be enough to access internal funding.
• As a mentor, you should commit to supporting undergraduate travel if they co-author a paper. They did the hard work — probably for less money than they deserve.
• Advocate for universities to reward faculty for student-centered research. This is the hard one. It means changing the way we quantify success with undergrads… and it’s an idea I’ve written about at length in a Student-Centered Research Manifesto.

A Student-Centered Research Manifesto

If you have ideas for increasing conference access to undergraduates, please let me know!

I am an Assistant Professor of Computer Science at Bucknell University, a liberal arts university in central Pennsylvania. You can learn more about me on my website or on Twitter. I’d also love to hear what your guiding principles are for research, so please don’t hesitate to contact me.