Hacker News Reader

Entry level courses are suppose to provide the perquisite background knowledge. You cannot take Data Structures until you've taken Computer Science 1 & 2. If students do not learn programming basics in Computer Science 1 & 2 they should be failed, where they can either retake the class or move to something that better aligns with their abilities and interest. The introductory classes serve to weed out the students that will not perform well in the department. Every department has these.

At my university (15 years ago), less than 1/4 of the people that started Computer Science 1 ended up starting Data Structures, but over 90% of the students that started Data Structures graduated from the department.

My thoughts? Utilize these introductory courses to set the standards expected from students, and expect a lot of freshmen to drop out. Additionally, I do put some blame on high schools for not teaching students fundamental skills like how to take good notes, how to read books, how to write sentences, and how to sit still for an entire lecture. If the standard that college students are educated to falls, then that blame belongs to the colleges.

You're kinda ignoring my point though - as a teacher (if I was), there is no way for me to unilaterally control what students have learned in those prerequisite classes. And clearly that's not working. Students entering my hypothetical classroom are coming in with insufficient knowledge to be successful. What can I alone, right now, do about it in my class?

Sure, I agree it's a systemic problem that needs to be addressed holistically, but I also need a plan for what to do in my classroom right now with these particular students. And I don't feel good about either passing those who haven't learned the material or failing 80% of the class.

If I was ignoring your point, it was unintentional because I was thinking about it through a systems level perspective and not at an individual lecturer's perspective. I didn't realize that was a concern of yours.

One thing I would try is to give students a scary speech on day 1. "This isn't going to be an easy class, I expect you to read the material, I will not be sharing my lecture notes, I'm a tough grader" etc etc.

I think an entrance exam is acceptable.

Having re-read my prior comment, I agree that I was unclear. Thanks for your comment. It sounds like we're probably 90-100% in agreement.

I'm thinking for intro just tell the students they are guaranteed a b- if they skip everything. Just acknowledge the transactionalism.

But the entry to the next class level requires a passing grade on the final exam, and DON'T water that down.

And of course, the exorbitant cost of college was unmentioned. These people are buying two new cars a year to go there. That uod the ante on the transactionalism substantially, and colleges everywhere would rather mint b+ degrees for 250k than impose standards

Quals are a standard part of graduate school. Many MS-only programs don’t require them, but most PhDs have done them. Quals work great as gatekeepers. But I think they can only work for small cohorts who have already self-selected into a challenging program and fewer students will give into the temptation to cheat (some will still cheat but it’s a lower proportion). Part of the secret sauce toward graduate quals is that most of the time the faculty know who they are going to accept before you even take the test- the test results probably only rarely flip their opinion of their students.

Quals will never be implemented at large in undergraduate mass-market coursework. The need for a placement exam on day one is supposed to be satisfied by prerequisite coursework. The fact that you have to pass a pre-calculus course before starting calculus is enforced by the school. Transfer credit from another school is supposed to be vetted by the registrar. And for the most part it is enforced, but the students still suck. Partly because if you struggled to get a C in pre-calculus then you’re not actually ready for calculus, especially after a couple week break in the summer or winter. Plus, a decent portion of students cheated to pass their pre-calc class. We could easily raise the pre-requisite requirements to a B+ or better, but that won’t actually work. There would be increased pressure to cheat, plus the alumni would stop their donations when their kids are forced to drop out.

The same thing would happen with a qualifying exam on day one. Many kids would cheat, so you’ll still get a bunch of unqualified students in your course. If you somehow managed to keep students from cheating you’d have so many students dropped on lesson two that you would break the school’s course scheduling system every semester. Would those failing students need to take their prerequisite courses again, or should they get to try again on your qualifying exam next semester without any extra courses? Either way is a disaster. The school would absolutely not let this go on for long in any decently sized course.

I don’t know what the answer is. It’s easy to say “just enforce the standards” but if nobody else is doing so then your efforts are wasted and you’ll probably get fired anyway.

> Quals will never be implemented at large in undergraduate mass-market coursework.

I believe proctored math placement tests are still common upon matriculation at less-selective colleges (e.g., directional public schools). Usually Accuplacer or done in ALEKS. That said, the outcome these days may be corequisite section placement rather than remedial course placement. Colleges have to balance readiness against the graduation delays that remediation adds (which often lead students to drop out entirely).

There are schools that use placement exams to place new students into their first course level, partly because there is inconsistency between grades from different high-schools, and even standardized test scores aren't amazing predictors of college course performance (they are correlated, just not as much as we'd like). The school I teach at does this, and we often place students who got as high as a 4 on their AP Calc AB exam back into introductory calculus to take again.

GP seemed to be talking more about a qualifying exam at the beginning of every course. I don't think that will ever happen mostly because of what you stated as, "the graduation delays that remediation adds." Can you imagine how long it would take average students to graduate, or the dropout rate, if we made each student pass a placement exam before every course? We couldn't implement this without a radical change to how universities work.

> trying to make up for years of poor education

Poor prior knowledge is one thing. But the larger issue discussed in the article isn’t so much lack of knowledge, but lack of skills (like being able to read and understand longer “adult” texts) and the average general attitude. You can’t fix any of those quickly.

I had a professor for my systems programming class give a qualifier exam day 1, worth 1% of the final but you get an F if you fail. It was just fizzbuzz, a student failed then complained to the dean, so they werent allowed to do it.

As a teacher I would be okay with that - I tried my best and did what I could.

I think the core problem is that administration would not allow you to do this.

> What am I supposed to do? Keep standards high and fail them all? That’s not an option for untenured faculty who would like to keep their jobs.

At the State university where I teach, literally part of our mission statement is to graduate every student who we admit. It has become a big part of the messaging from upper administration in the last few years.

Why is that a priority? I would think maintaining the reputation of a StateU diplomas would be more important for the school and graduates.

Love it.