https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym Made with whimsy en-us 2021-10-03 22:40:06 UTC 2021-10-20 16:56:45 UTC hello@padlet.com kaitlynzhou1 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794618917 Please write comments, questions, and concerns! Upvote posts you like! Please be kind and constructive!

We'll most likely share this whole padlet with some faculty and staff members. ]]> 2021-10-05 22:57:55 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794618917 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794626232 get to know the peeps around you :) 

]]> 2021-10-05 23:03:07 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794626232 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794628683
+8

(Still time!)]]> 2021-10-05 23:04:50 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794628683 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794630291 2021-10-05 23:05:57 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794630291 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794633270 2021-10-05 23:08:06 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794633270 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794634688 - It's actually a "non" enforced requirement. I know some > 2 years who haven't completed breadth

-but you need to complete it eventually right?
-- yes to graduate

- I don't mind eventually, but why by year 2? How about sprinkling out research-irrelevant courses over time?

- If it isn't actually required, why call it a requirement? Why not call it a recommendation?

- They shouldn't say this is a requirement. It is not...]]> 2021-10-05 23:08:59 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794634688 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794639624 The university level website had some scary text about declaring candidacy by the end of second year and it being a big deal if you don't but my advisor and upper years told me it doesn't matter.

In general it seems like the scary language about program requirements and what milestones should be completed by when is really out of step with what actually happens

+1]]> 2021-10-05 23:12:26 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794639624 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794649085 Unless someone literally wrote in "third gender" on the form

- I agree the wording is awkward but would not recommend "other" as the solution. See https://interactions.acm.org/archive/view/july-august-2019/how-to-do-better-with-gender-on-surveys]]> 2021-10-05 23:18:42 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794649085 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794661072 2021-10-05 23:26:15 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794661072 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794671485 I mean, good to know, not great that it is the case

It could also be because not everyone was looking to align with people they rotated with (eg. If you know your main advisor but want to rotate outside of the department for the experience)]]> 2021-10-05 23:32:57 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794671485 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794679824 Also I don't think "really easy" is the right takeaway for the data? Manageable might be a better takeaway?

Also what is the benefit of breaking up these results by area? Seems less relevant for this question than for things that have to do with specific geoups and advisors like rotation difficulty.
^because it. Unevenly impacts people from different sub areas? ]]> 2021-10-05 23:38:29 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794679824 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794685486 This would be a strengthening of what we did last year (someone had to "vouch" for them).

You could also do this for the whole department, but it doesn't seem needed elsewhere.

- cap the number of AI/ML admits to match the spots available]]> 2021-10-05 23:42:17 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794685486 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794687000 "How important" question does not imply that the department needs to improve.

+1]]> 2021-10-05 23:43:17 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794687000 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794691282 2021-10-05 23:45:48 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794691282 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794692426 I agree that there should be more accountability from faculty about which students they have advocated for/committed to working with and how many students they can accept. It feels like during the admissions process, faculty are incentivized to convince students to come to Stanford even if they have no solid commitment to working with that student and thus little stake in their long term success. ]]> 2021-10-05 23:46:28 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794692426 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794692737 I would say X % did not get a alignment offer after 3rd rotation instead of 8 students. Otherwise, faculty might think only 8 that is great because they don't see in that moment that only ~50? people answered.

+1]]> 2021-10-05 23:46:40 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794692737 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794698903 Might be a good idea to include high level summary items of each section, and suggested action items at the end of the presentation?]]> 2021-10-05 23:50:20 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794698903 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794701184 +2]]> 2021-10-05 23:51:31 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794701184 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794710899 - Unevenly affect students from different backgrounds / undergrad institutions
- response times for waivers are excessive

+6

----
Adding on:

Even though I come from a background that let me waive a lot of courses and was very math + cs heavy, I found the requirements quite stressful. In particular trying to get an A-   , rather than having to just pass or even meet a lower bar like a B- which was the requirement in my previous grad program. I also know not all stanford PhD programs have this A- requirement 

The timing of the offerings of breadth requirement courses also makes it very difficult to finish them by year 2

+5
---
Thanks for the work you all have done on this survey and on sharing the results, but I was a bit unhappy at the way that breadth requirements were cast as "pretty easy" for most students; it seems that differs quite a bit among students depending on their background and research area. Were there results on the extent to which the breadth requirements impacted students' ability to work on research (in similar ways as rotations were investigated)? What's worrying to me is that since the breadth requirements more strongly impact students with non-traditional backgrounds, these students are bearing a greater burden that may further get in the way of research productivity. 

Another thing that students have discussed is that the current apportionment of classes to breadth categories is very out-of-date; a large proportion of classes (AI, CompBio, Security, DB, Graphics, HCI) is lumped into the "Applications" bucket, so students who are not focused in the Theory or Systems areas often have to complete more breadth requirements even if they came from a CS background. 

+5]]> 2021-10-05 23:56:49 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794710899 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794712082 When you break down results by subfield, some of the results may be more variable since they have fewer respondents. Maybe you can annotate each column with the number of students in that subfield, or combine related subfields into larger chunks.

-EDIT: merged ]]> 2021-10-05 23:57:28 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794712082 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794719184 On some graphs it was unclear what 1 and 5 meant—would be great to have explicit labels (I.e. 1=low, 5=high) 

-merged]]> 2021-10-06 00:01:39 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794719184 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794721152 The AI + applications category feels like too big of a bucket compared to the other two (theory and systems). Not sure this is because the number of “application” subfields has expanded with time, but it seems like it’s grouping together a lot of very different areas and therefore kind of downplaying their relative value.
+4]]> 2021-10-06 00:02:47 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794721152 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794731169 2021-10-06 00:08:44 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794731169 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794797219 "If you take cranberries and stew them like applesauce, they taste much more like prunes than rhubarb does." - Groucho Marx

]]> 2021-10-06 00:39:11 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794797219 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794824493 It would be beneficial to have more (non-sensitive) information about the population of students:

Prior background before coming to stanford (e.g.) did they do a masters, or are they coming straight from undergrad. 

There may be more insight into the rotation experience conditioned upon experience]]> 2021-10-06 00:50:16 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794824493 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794825649 2021-10-06 00:50:48 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794825649 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794852664 I strongly agree with the things that people mentioned during the meeting and in the pad, but adding on to make sure this gets surfaced.

Breadth requirements are not useful.
In addition to everything that's been said, it feels like breadth req heavily favors those in certain areas (e.g. sys, theory) both in terms of being able to waiver, AND in terms of whether the content learned is actually useful. For many of the courses I took for my breadth req, I still cannot fathom a way that they would be useful and this belief seems to last even among the senior students or PhD graduates. I feel like there is a heavy opportunity cost here because there are courses I would have taken if it weren't for the breadth req. that could significantly actually help my research and career.

Importantly, given that the field of CS has grown so much over the years both in terms of its size and topics, it seems unreasonable to assume that learning about some of the topics asked in the breadth req. will be universally helpful to all students.

Breadth req should not be the "threshold"
One argument against removing the breadth req. appears to be that it's the minimum "threshold" or "bar one has to pass" to get a CS PhD from Stanford. I think this only makes sense if the req puts everyone on a level playing field, which it very clearly does not; it very strongly (and explicitly) favors those from large/well-known CS departments and fields such as sys/theory, and those from better-represented communities who's been exposed to CS and CS education at an early age. I've seen cases of admitted students from under-represented communities (both in terms of demographically and academically) turn away from Stanford partially due to this and the general "elitist" aura that such req asks for.

What do other schools do?
From what I understand, other peers schools (e.g. CMU, Washington, UIUC, Berkeley (?), haven't checked MIT) do not have these sorts of breadth requirement. Often they give much more flexibility to the students; let the students pick the courses that will "actually help" the their career, and ask for a lower grade cutoff (e.g. B or B-).

This puts Stanford at a significant disadvantage 1) when we are recruiting new students (i.e. students from under-represented communities turning down Stanford). 2) This also puts the current students at a significant disadvantage -- when other PhD students from our peer institutions are focusing on their research and courses that actually help their pursuit, students at Stanford spend two if not more years of their time focused on getting a good grade in a class that has nothing to do with their work.

---
Comment: I strongly disagree with the comparison to other schools. Stanford's flexible breadth requirements structure relative to its peers is actually one of the reasons I chose to come to Stanford. Breadth courses do exist at other places -- see, e.g., https://www.cs.washington.edu/academics/phd/courses/quals -- and often cannot be waived. The biggest difference is the grade cutoff mentioned above, which works fine for some students in some courses and not fine for other students in other courses. Some schools also arguably do better than others at counting special topics courses toward breadth requirements, but my info on that is very anecdotal.

—-
No! The biggest difference is the courses included in the categories^ in each of the categories on the link posted, everyone is likely to find something at least tangentially related to any kind of CS research.]]> 2021-10-06 01:01:51 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1794852664 https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1799947230 Maybe the future working group can be a more representative group, with people from more CS focus areas and backgrounds in general. This way the issues faced by different people from more diverse backgrounds or minorities can be probed in the survey.

+1]]> 2021-10-07 14:15:45 UTC https://padlet.com/kaitlynzhou1/5iz27era1jwdbaym/wish/1799947230