Comment by BrenBarn
8 months ago
I see that pressure as well. I find that a lot of the problems we have with AI are in fact AI exposing problems in other aspects of our society. In this case, one problem is that the people who do the teaching and know what needs to be learned are the faculty, but the decisions about how to teach are made by administrators. And another problem is that colleges are treating "make money" as a goal. These problems existed before AI, but AI is exacerbating them (and there are many, many more such cases).
I think things are going to have to get a lot worse before they get better. If we're lucky, things will get so bad that we finally fix some shaky foundations that our society has been trying to ignore for decades (or even centuries). If we're not lucky, things will still get that bad but we won't fix them.
Instructors and professors are required to be subject matter experts but many are not required to have a teaching certification or education-related degree.
So they know what students should be taught but I don't know that they necessarily know how any better than the administrators.
I've always found it weird that you need teaching certification to teach basic concepts to kindergartners but not to teach calculus to adults.
I attended two universities to get my computer science degree. The first was somewhat famous/prestigious, and I found most of the professors very unapproachable and cared little about "teaching well". The second was a no-name second tier public uni, but I found the professors much more approachable, and they made more effort to teach well. I am still very conflicted about that experience. Sadly, the students were way smarter at the first uni, so the intellectual rigor of discussions was much higher than my second uni. My final thoughts: "You win some; you lose some."
This is universal. I’ve had largely the same experience. There’s several reasons for this.
1. Stupider people are better teachers. Smart people are too smart to have any empathic experience on what it’s like to not get something. They assume the world is smart like them so they glaze over topics they found trivial but most people found confusing.
2. They don’t need to teach. If the student body is so smart then the students themselves can learn without teaching.
3. Since students learn so well there’s no way to differentiate. So institutions make the material harder. They do this to differentiate students and give rankings. Inevitably this makes education worse.
8 replies →
That's been my experience too, and I think it actually makes perfect sense from an evolutionary perspective - if the students are smart enough to learn well regardless of the level of the instruction, then the professors don't face any pressure to improve.
Taking this to the extreme, I think that a top-tier university could do very well for itself by only providing a highly selective admission system, good facilities and a rigorous assessment process, while leaving the actual learning to the students.
6 replies →
>I've always found it weird that you need teaching certification to teach basic concepts to kindergartners but not to teach calculus to adults.
There is a lot more on the plate when you are kindergarten teacher - as the kids needs a lot of supervision and teaching outside the "subject" matters, basic life skills, learning to socialize.
Conversely, at a university the students should generally handle their life without your supervision, you can trust that all of them are able to communicate and to understand most of what you communicate to them.
So the subject matter expertise in kidnergartens is how to teach stuff to kids. Its not about holding a fork, or to not pull someones hair. Just as the subject matter expertise in an university can be maths. You rarely have both, and I don't understand how you suggest people get both a phd in maths, do enough research to get to be a professor and at the same time get a degree in education?
I was an instructor for a college credit eligible certification course. While I think that education degree is more than you need, providing effective and engaging instruction is a skill and is part of actual teaching at any level. Concepts like asking a few related open ended, no right answer questions at the beginning of a new topic to prime students’ thinking about that topic. Asking specific students “knowledge check” or “summarize/restate this topic” questions throughout keeps students from checking out. Alternating instruction with application type exercises help solidify concepts. Lesson plans/exercises/projects that build on each other and reincorporate previous topics. Consideration of how to assess students between testing and projects, for example a final vs a capstone project.
If you are just providing materials and testing, you aren’t actually teaching. Of course there are a ton of additional skills that go into childhood development, but just saying adults should figure it out and regurgitating material counts as “teaching” is BS.
Just watch out for who is certifying how things should be taught. It’s honestly one reason education is so bad and so slow to change.
Edit: and why perfectly capable professionals can’t be teachers without years of certification
> I've always found it weird that you need teaching certification to teach basic concepts to kindergartners but not to teach calculus to adults.
I think this is partially due to the age of the students, by the time you hit college the expectation is you can do a lot of the learning yourself outside of the classroom and will seek out additional assistance through office hours, self study, or tutors/classmates if you aren't able to understand from the lecture alone.
It's also down to cost cutting, instead of having entirely distinct teaching and research faculty universities require all professors to teach at least one class a semester. Usually though the large freshman and sophomore classes do get taught by quasi dedicated 'teaching' professors instead of a researcher ticking a box.
>don't know that they necessarily know how any better than the administrators.
If someone is doing something day in and day out, they do gain knowledge on what works and doesn't work. So just by doing that the professors typically know much more about how people should be taught than the administrators. Further, the administrators' incentives are not aligned towards insuring proper instruction. They are aligned with increasing student enrollment and then cashing out whenever they personally can.
This is very different in France. Studying to be a teacher at university level is a big deal.
Since the reform on University administration circa 2011, a big push was done towards 'evaluation continue' (basically regular tests), which now last until your third year in some Uni, to make public universities more like private schools, and against 'partiels' (two big batteries of standardized tests in person, with thousands in the same area, with only pen and papers, one early January, second in may, every year, over a week).
That push was accelerated because of COVID, but with the 'AI homework', it gave teachers a possibility to argue against that move and the trend seemed stopped last year (I don't now yet if it has reverted). In any case, I hope this AI trend will give more freedom to teachers, and maybe new ways of teaching.
And I'm not a big Llm fan in general, but in my country, in superior education, it seems good overall.
1 reply →
Once you're an adult some of the best lessons come from having bad teachers.
Adult here! No.
Adult here; there are lessons everywhere, but people learn very differently.
Nobody knows "how" things should be taught. Pedagogy is utter disaster.
I am pretty sure that early childhood education (until fifth grade) is a very active area of research in all highly developed nations. Almost, by definition, it you want to (a) become or (b) stay a highly developed nation, you need to have a high quality public education system.
My mother was a first grade teacher for 30+ years. In her school system, first grade is the year that students learn to read. Each year, she was also required to take professional training classes for a certain number of days. She told me that, in her career, there were many changes and improvements and new techniques developed to help children learn how to read. One thing that changed a lot: The techniques are way more inclusive, so non-normie kids can learn to read better at an earlier age.
1 reply →
A PhD was historically a teaching degree: that’s what the D stands for.
No?
PhD - Philosophy Doctor
5 replies →
The instructors may not know the absolute best way to teach, but I think they do know more than the administrators. All my interaction with teacher training suggests to me that a large proportion of it is basically vacuous. On dimensions like the ones under discussion here (e.g., "should we use AI", "can we do this class online"), there is not really anything to "know": it's not like anyone is somehow a super expert on AI teaching. Teacher training in such cases is mostly just fads with little substantive basis.
Moreover, the same issues arise even outside a classroom setting. A person learning on their own from a book vs. a chatbot faces many of the same problems. People have to deal with the problem of AI slop in office emails and restaurant menus. The problem isn't really about teaching, it's about the difficulty of using AI to do anything involving substantive knowledge and the ease of using AI to do things involving superficial tasks.
I totally agree. I think the neo-liberal university model is the real culprit. Where I live, Universities get money for each student who graduates. This is up to 100k euros for a new doctorate. This means that the University and its admin want as many students to graduate as possible. The (BA&MA) students also want to graduate in target time: if they do, they get a huge part of their student loans forgiven.
What has AI done? I teach a BA thesis seminar. Last year, when AI wasn't used as much, around 30% of the students failed to turn in their BA thesises. 30% drop-out rate was normal. This year: only 5% dropped out, while the amount of ChatGPT generated text has skyrocketed. I think there is a correlation: ChatGPT helps students write their thesises, so they're not as likely to drop out.
The University and the admins are probably very happy that so many students are graduating. But also, some colleagues are seeing an upside to this: if more graduate, the University gets more money, which means less cuts to teaching budgets, which means that the teachers can actually do their job and improve their courses, for those students who are actually there to learn. But personally, as a teacher, I'm at loss of what to do. Some thesises had hallucinated sources, some had AI slop blogs as sources, the texts are robotic and boring. But should I fail them, out of principle on what the ideal University should be? Nobody else seems to care. Or should I pass them, let them graduate, and reserve my energy to teach those who are motivated and are willing to engage?
I think one of the outcomes might be a devaluation of the certifications offered in the public job marketplace.
I can say from some working experience in the United States that way too many jobs require a university degree. I remember being an intern or my first job after uni (which I struggled a great deal to complete), looking around and thinking: "There is no way that all of these people need a uni degree to do their jobs." I couldn't believe how easy work was compared to my uni studies (it was hell). I felt like I was playing at life with a cheat code (infinite lives, or whatever). I don't write that to brag; I am sure many people here feel the same. So many jobs at mega corps require little more than common sense: Come to work on time, dress well, say your pleases and thank yous, be compliant, do what is asked, etc. Repeat and you will have a reasonable middle class life.
7 replies →
> Some thesises had hallucinated sources, some had AI slop blogs as sources, the texts are robotic and boring. But should I fail them, out of principle on what the ideal University should be?
No, you should fail them for turning in bad theses, just like you would before AI.
That's probably what should happen, but it's not what happens in reality. In grading I have to follow a very detailed grading matrix (made by some higher-ups) and the requirements for passing and getting the lowest grade are so incredibly low that it's almost impossible to fail, if the text even somewhat resembles a thesis. The only way I could fail a student, is if they cheated, plagiarised or fabricated stuff.
The person who used the AI slop blog for sources, we asked them to just remove them and resubmit. The person who hallucinated sources is however getting investigated for fabrication. But this is an incredibly long process to go through, which takes away time and energy from actual teaching / research / course prep. Most of the faculty is already overworked and on the verge of burnout (or are recovering post-burnout), so everybody tries to avoid it if they can. Besides, playing a cop is not what anybody wants to do, and its not what teaching should be about, as the original blog post mentioned. IF the University as an institution had some standards and actually valued education, it could be different. But it's not. The University only cares about some imaginary metrics, like international rankings and money. A few years ago they built a multi-million datacenter just for gathering data from everything that happens in the University, so they could make more convincing presentations for the ministry of education — to get more money and to "prove" that the money had a measurable impact. The University is a student-factory (this is a direct quote by a previous principal).
11 replies →
You should fail them.
The larger work that the intellectual and academic forces of a liberal democracy is that of “verification”.
Part of the core part of the output, is showing that the output is actually what it claims to be.
The reproducibility crisis is a problem Precisely because a standard was missed.
In a larger perspective, we have mispriced facts and verification processes.
They are treated as public goods, when they are hard to produce and uphold.
Yet they compete with entertainment and “good enough” output, that is cheaper to produce.
The choice to fail or pass someone doesn’t address the mispricing of the output. We need new ways to address that issue.
Yet a major part of the job you do. is to hold up the result to a standard.
You and the institutions we depend on will continue to be crushed by these forces. Dealing with that is a separate discussion from the pass or fail discussion.
Fail them. Only let the ai generated text that has been verified and edited to be true to pass.
If they want to use AI make them use it right.
> Some thesises had hallucinated sources, some had AI slop blogs as sources, the texts are robotic and boring. But should I fail them, out of principle on what the ideal University should be?
I don't think you should fail them - instead, give them feedback on how to improve their thesis themselves, and how to make better use of tools like ChatGPT.
If instead of flat out failing to turn in their thesis, instead they are submitting work that needs more iteration due to bad use of AI, that sounds like a net win to me. The latter can be turned into something useful.