I just did a livestream answering questions about PhD applications and am really concerned about the number of people who think that having published papers is an absolute prerequisite for admission to a PhD program.
And I'll just say it: If as a professor or an admissions committee your major criteria is "author on a published paper in a major journal" then I hope you're looking real hard at the diversity of your student population because I'm guessing you might have a problem.
There's a reply in this thread that's (I'm paraphrasing): profs want applicants who've already published b/c they see PhD students as paper producing factories.
I'm concerned with the number of likes it has :( Are there so many profs like this that this is a dominant perception?
OK, let's talk about this. There are two issues with different problems and solutions: (1) PhD applicants thinking they don't have a chance if they don't already have publications. Since that's not always true, how can we provide encouragement? (2) The extent to which it IS true.
In terms of anxiety over "auto-reject" metrics (of which there are many theories, from no pubs to low GRE scores to "not elite enough undergrad") I'd just like to encourage everyone to remind applicants that strengths make up for weaknesses. There should be no auto-rejects. <3
To the extent to which it is clear that for SOME programs "no pubs" might actually be an auto-reject (🤮), two things: (1) How do we get those programs to think hard about why this is a terrible idea; and (2) How can we help level the playing field as much as possible?
And yes, I am 100% comfortable saying that if your program will not consider accepting students who are not already authors on PUBLISHED PAPERS then you are doing something wrong. Full stop.
1. Not everyone has the opportunity as an undergrad to do the kind of research that results in published papers. Because of e.g. the university they attend, they don't know they can (e.g. first-gen) or there aren't paid RA opportunities & they can't afford to work for free.
So if your criteria for PhD admission is "already has publications" you're weeding people out based on things like where they can afford to go to school for undergrad and not accounting for e.g. work in classes that was REALLY GOOD RESEARCH but doesn't result in pubs.
A thought experiment. What is better: (1) you are 4th author on a 7-author paper because you did data analysis as part of a huge lab; or (2) you led your own research project & wrote an amazing senior honors thesis that of course isn't published yet because you're an undergrad
"Must have published papers" as a PhD admissions criteria suggests that (1) is better, but that is obviously wrong, and not only are programs missing out on awesome people who did (2) instead but it incentivizes doing (1) over (2) b/c it's more likely to result in publications.
It is true that often the strongest PhD applicants do have publications. In part b/c it means not only research experience but doing it long enough that they made it *through the publication cycle* which is not a small amount of time. Also likely an LOR from a prof they helped.
I often tell applicants that an application isn't stronger by virtue of having a masters degree, but that having a masters degree often means they have more research experience or better LORs or other things that might make an application stronger.
Similarly, I don't see why having a publication already makes your application stronger. I don't care that you've published. I care about the project you worked on and why you loved it and how it inspired you to pursue a PhD. If you still have that without a pub, it's equivalent.
This brings us to how we can help level the playing field. Because *publications* as a silly metric aside, it's well established that some experience with what research looks like in your field (whether that's in a lab or with a thesis or in a class) makes a stronger application.
There are a lot of efforts for encouraging research experience for undergrads. I would like to see even more of this that also includes education and opportunities for leading work, not just undergrads as cogs in the paper factory.
I'd also like to see these opportunities advertised more widely and made more widely known to students who don't even know to look for them. And I'd like to see them not silo-ed at universities so that even students who aren't at research universities have these opportunities.
I'd also like to see more efforts to de-mystify PhD admissions, especially in specific fields. The number of messages I got from especially first-gen students telling me that would never have known how to apply were it not for my youtube videos last year was alarming.
I apologize for this tweet-thread-that-probably-should-have-been-a-blog-post. I would really love to hear others' thoughts though. :)
Also if you think that general PhD admissions advice would be useful for someone please pass this on. With the caveat that my experience is only my own and probably most applicable to STEM programs in the U.S. youtube.com/playlist?list=…
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Following four years of empirical work on research ethics for public data, our (@pervade_team) manifesto for trustworthy pervasive data research--foregrounding power dynamics and learning from ethnographers--published in @BigDataSoc. journals.sagepub.com/doi/10.1177/20…
This paper in part details evidence that many data subjects are unaware of the research uses of their digital communications, and often express unhappiness and alarm. We map awareness to this 👇 spectra and recommend researchers reflect on where their data-gathering methods fall.
Importantly, using “public” data does not relieve researchers from considerations of participant awareness, because awareness of creation is not necessarily awareness of research use. And we should reflect on both awareness and the power implications of our research.
I'm "teaching" a highly condensed version of my tech ethics & policy class on TikTok. Here's all the videos I made for the week on traditional ethical theory featuring "should Batman kill the Joker?" and ending with COVID-related moral messaging: instagram.com/tv/CT0epClhQRO/
You can follow along with this experiment to "teach a class on TikTok" here! Links to videos along with readings. I'm keeping pace with the actual @CUBoulder class I'm teaching. Apparently 2.5 hours of in-class time becomes 8 minutes of video. :) bit.ly/caseysclass
(I was going to just post the combined-topic longer videos on Twitter, but apparently Twitter has a 2 minute video length limit! So trying out Instagram TV instead, hopefully that works ok!)
Inspired by this paper (psyarxiv.com/9yqs8/) when covering traditional ethical theory in my class, we discussed what different frameworks might suggest for how to convince someone to do the right thing (e.g., WEAR A MASK!). One insight I had was: utilitarianism won't work. 🧵
A utilitarian moral message (like the one used in that paper) is essentially "think of the consequences for everyone if you don't do this!" but... people have to have an understanding of the consequences for that to be effective, but there's so much misinformation. :(
For the study in the paper (it's a pre-print) they found a modest effect for duty-based deontological messaging ("it is our responsibility to protect people) re: intention to share the message. But we also covered OTHER ethical frameworks in my class...
There has been a very upset/angry reaction to a paper published using tweets about mental health. I'm not RTing because I'd like to talk about this without drawing more attention to the researchers or the community. But it's an important research ethics cautionary tale. [Thread]
The paper is a qualitative analysis of tweets about mental health care. It includes paraphrased quoted tweets that the researchers ensured were not searchable. The study was approved by an ethics review committee in the UK, and the paper cites the AOIR ethics guidelines.
The paper includes an ethics and consent section that includes the above and notes that because tweets are public, consent was not required. The study also included a researcher with mental health lived experience. There do not appear to be any other statements regarding ethics.
The problem with workload (and thus work-life balance) in academia is:
You can always do more.
(A thread based on a recent personal epiphany.)
Unlike many other kinds of jobs, when you are a (research-heavy) professor no one tells you exactly what you need to do. Or even how much you need to do. There are things that are wonderful about this kind of freedom. But also, it means that you can always be doing more.
How many research projects should you be doing at any given time? How many papers should you be writing? You might have a personal sense for this, maybe even a rule-of-thumb, maybe even a mentor giving you advice. But whatever N is, it COULD always be N+1.
I'm often struck by how much the foundation of science relies on individual integrity. And typically I feel it's pretty solid. But this kind of garbage is a result of publish-or-perish, bean counting, and the general incentive structures of academia. cacm.acm.org/magazines/2021…
Also when this situation came to light a year ago I went on a whole lengthy tweet-rant about it so I won't repeat myself but here you go. :)
Also to clarify: "is a result of" above should probably be "exacerbated by" because obviously it's a result of when there is a breakdown of individual integrity. Awful people gonna be awful, but there are incentive structures dictating the particular form of awful.