10 ms·

> For example, students who come into a course with little prior knowledge earn lower grades at the start, which means they get a lower final average, even if t

> For example, students who come into a course with little prior knowledge earn lower grades at the start, which means they get a lower final average, even if they ultimately master the material. I'm not sure this is the worst flaw with grading (the equity issues may be worse), but it's certainly the dumbest flaw. A scoring system for scholastic ability that disadvantages those that start off with little knowledge, even when they do well by the end of the course, is ridiculous on its face.

Double_a_92 5y ago
How does it disadvantage people that start from scratch though? Tests should be written in a way that everyone can pass them, with the knowledge thought in the course. If people that already have some related knowledge have an easier time studying, that's just life. How does it unfairly disadvantage you, if somebody else is already more skilled than you?
- matt321 5y ago
  >How does it disadvantage people that start from scratch though? In the same way that not being a genius is a disadvantage. Cant change it unless we handicap the smart people.
  - jdkee 5y ago
    “ In the year 2081, the 211th, 212th, and 213th amendments to the Constitution dictate that all Americans are fully equal and not allowed to be smarter, better-looking, or more physically able than anyone else. The Handicapper General's agents enforce the equality laws, forcing citizens to wear "handicaps": masks for those who are too beautiful, loud radios that disrupt thoughts inside the ears of intelligent people, and heavy weights for the strong or athletic.” See https://en.m.wikipedia.org/wiki/Harrison_Bergeron https://en.m.wikipedia.org/wiki/Harrison_Bergeron
- conradludgate 5y ago
  > earn lower grades at the start, which means they get a lower final average They get graded from the very beginning, not from a single final exam
  - Double_a_92 5y ago
    So? Each small test should be written in a way that you can ace with what you learned so far in the course. If you are not willing to study that, that's your personal problem.
- mckeed 5y ago
  They are talking about assignments, not tests. It applies more to skill-based courses than knowledge-based ones. If you're writing C papers at the beginning and A papers at the end, should you get an A or B?
  - Double_a_92 5y ago
    Why would you be writing C papers though? Either the tasks are badly designed, in a way that you can't properly solve them with what you have been thought so far... or you have all the possibility to write As in the beginning, even if you don't have prior knowledge.
- asoneth 5y ago
  The argument is that giving weight to the grades earned throughout the course may under-value students who made significant improvements. For example, take two students: Student A comes in barely meeting the prerequisites and struggles with low grades on initial tests and projects. They come to office hours, put in substantial effort, and eventually ace the final projects and tests. Student B comes in with substantial background experience and gets decent but not stellar grades on all tests and projects. Depending on how much weight is given to the final tests/projects student B may still end up with a higher grade despite the fact that they probably do not have as strong an understanding of the material as student A. I've had both kinds of students the handful of times I taught graduate-level seminars and it's one reason I weighted the final projects more heavily than initial projects. (Though I found grades on initial projects were still useful for other reasons.)
- enragedcacti 5y ago
  Imagine a course as teaching the content along a linear function from 0 to N understanding. With traditional grading we are effectively determining the area under the line for the given class, even though the goal of the class is to leave with a mastery of the content. Consider two students, one whose education path in the course is exponential and the other whose path is logarithmic where both paths intersect at the end of the course. According to traditional grading the logarithmic student would score far better in the course despite their knowledge being equal to the other student by the end. Which student would you rather be? who has more potential? Should the body of a course be treated like a series of 100m sprints or a marathon, and what is the net positive outcome through each approach?
- MiddleEndian 5y ago
  I think his point is that if Student A started with 50% knowledge and Student B started with 80% knowledge and by the end they both have 95% knowledge, Student A's grade for the course is penalized for the slow start even AFTER reaching mastery by the end. If Student A ends up only at 80% then so be it, but if they end up matched then they should end up with a similar grade.
- orwin 5y ago
  This very year, my sister started college after 5 years as a chef. Her math was shit, especially since it was already not her strongest subject in highschool, 5 years prior. She started very low and ended up average (she's going into biology). With help from me and Wolfram alpha (sry about that advertisement but it helped a lot). She was graded below average even though she showed more progress than at least half her class (since she started almost at the bottom and ended up in the middle). She doesn't care, but still think she is bad at math, when she just had poor teachers in HS and showed learning abilities at 23.
Aachen 5y ago
How so? If someone already knows the material or has a very solid framework to fit the extra knowledge into and needs zero catching up, of course they'll know the material better at the end of the course and/or will be able to apply it better. How should that not translate into a better grade? The only alternative I see is to let those less-prior-experienced students take a longer time and only then determine the final grade.
- etchalon 5y ago
  Or you could grade for mastery of material, instead of the individual steps taken to get there.
- rhn_mk1 5y ago
  That presumes people learn at equal rates. People don't learn at equal rates though.
- octokatt 5y ago
  Because the grading system _encourages_ students who already know the material, and _discourages_ students who don’t know the material. If someone is discouraged from a subject, they are less likely to put in the effort to succeed in the subject. This creates a self-perpetuating cycle, which ideally education would fix. Similarly, this sets students up for a false idea of their “good subjects”, when instead those are familiar subjects. A student who could be a polymath gets discouraged from pursuing multiple subjects because they’re already far ahead of the game in math, so they pigeon-hole themselves to the subject where they receive positive reinforcement. Essentially, grades as a feedback system are less than ideal, and we can do better.
  - Aachen 5y ago
    > Similarly, this sets students up for a false idea of their “good subjects”, when instead those are familiar subjects. A student who could be a polymath gets discouraged from pursuing multiple subjects I had never considered it that way, interesting perspective. Not yet sure how this will work in the grander scheme, I'm thinking of the impact on the value of a diploma when it's based on learning ability or broader knowledge as compared to being specialized in one thing. Something I'll be thinking about in the future.
blodkorv 5y ago
How far should we strech this? there are students who come into courses with superior memory or iq and get a head start from that.
- jrgoff 5y ago
  Not sure what your getting at here? The person you are commenting on wasn't talking about punishing people for having a head start they were saying that people's final grade should be based on what they are capable of at the end of the class, not at the start of it.
  - blodkorv 5y ago
    Why not, there will be for sure students who have the ability to grasp a course fast and effectily to compete with those with prior knowledge, why not grade and reward that ability.
    - mech987 5y ago
      If you grade improvement rather than mastery, you incentivize feigning incompetence at the start of the course to game the system. The closest thing I ever saw to this was my high school weightlifting class. Your bench press single rep max was a high percentage of your overall grade, and was tested several times over the course of the semester. You were graded on improvement. The teacher made sure not to mention this before the first test, to keep people from cheating, but a few kids were in the know and faked a low max. From a philosophical/values point of view- what is more important for education- incentivize all students to try to achieve the mastery level expected, or incentivize all students that as long as it looks like they're making progress, that is enough? I'd go with mastery. The appearance of progress is too easy to game, too subjective, and not easily applicable to varying courseloads.
    - lostcolony 5y ago
      It is rewarded, simply by the fact that they then have to put less effort into the class. Why should it be graded? The point of the class is not to learn quickly and effortlessly, but to learn the material completely. If by the time the class is over the student has learned it well, they should get a grade that reflects that learning. I mean, by your logic, should we give extra credit to the students who never show up to office hours in college, since clearly they're learning it 'faster and (more) effectily(sic)'?
micromacrofoot 5y ago
Shouldn't the course be designed in a way where everything on the exam has been taught in the class? People with prior knowledge will still have an upper-hand in everything (even beyond school), so I'm not sure that it's even a solvable problem. I guess perhaps one could give a baseline exam on day 1 to qualify prior knowledge and give a handicap to people if they're coming in blind? Though if I knew that purpose I'd game the system, so it probably still isn't fair.
notreallyserio 5y ago
> A scoring system for scholastic ability that disadvantages those that start off with little knowledge, even when they do well by the end of the course, is ridiculous on its face. This is one reason I did poorly in school and eventually dropped out. Teachers chose (either independently or by arrangement) to weigh all grades equally, whether they were at the start of the class or the end. So if you did poorly at the start, you might as well give up, you'll never dig yourself out of that hole.
specialp 5y ago
I had such a situation when I first entered college. I came from a disadvantaged background, and was not a good writer. My freshman composition class was taught by a woman who had us write synopses of what I thought of at the time as boring, long-winded New Yorker articles. She ruthlessly hammered me on grammatical errors. I ended up withdrawing from the course. Then next semester I had a the same class but a different professor. He was much different. He had us write about our life experiences and things that interested us. You would get the grammatical corrections as feedback but you would not get blasted for them. Over that one semester I used the literary handbook I was given, and got a lot better at it. It really helped me. In my future liberal arts classes I had to take, I got good grades on my papers and they were well received.
- Auracle 5y ago
  I was required to take an art course in college, and I chose an intro level graphic design class. I’m terrible at all things art other than photography, but I thought graphic design made sense being a CS major an all. My experience was terrible. There was no real artistic instruction. For instance, during our first or second week we were told how to use the pen tool in Illustrator and or homework was to make a logo for a made up company. I, having absolutely no art background, did a bad job. She actually projected my work in front of the class and had me stand up in front to be ridiculed. I decided to drop the class. When I brought her the form to sign all she said was “thank you,” which I very much took to mean “thank you for dropping out of my class.” Heaven forbid you actually need to do your job, professor.
  - sitkack 5y ago
    I am sorry for your experience. Most college level art classes have a pretty strict entrance requirement where portfolios are judged pretty harshly from incoming high school students. Incoming students are basically already required to be good. This instructor was not prepared for you and treated you poorly. Another instructor might have handled it much better.
    - navbaker 5y ago
Clubber 5y ago
> For example, students who come into a course with little prior knowledge earn lower grades at the start, which means they get a lower final average, even if they ultimately master the material. One of my strategies in college was to really study on the first exam. The logic being is it's just introductory material and therefore the easiest exam to make an A on. Subsequent exams tended to get into the nitty gritty and require a lot more information to be understood / retained. This held true on any course. In my experience, this negates the quoted point that people who had little prior knowledge would do poorly early on. Granted I went to college a long time ago. Things may have changed.
- lostcolony 5y ago
  > One of my strategies in college was to really study on the first exam Why did you really have to study if it was so easy? Right, because you didn't know it. For others who came in knowing it, they didn't have to 'really study'. And this still holds true for every successive test. I feel like that kinda negates your negation. To speak of specifics, I took calc 1 twice. First time was due to inconsistencies in the teacher's grading, but whatever; the second time I took it, I was very familiar with the concepts, and made an easy A. Others struggled. It's fair to say I should have gotten the A at the end because I knew the material, and it's fair to say the people who got an A on every test (those 'easy' first ones you mention being included) should receive an A, but was it fair to say others, who by the end had the same understanding I and the other A students did, should have gotten a B even if they aced a cumulative final, because they struggled with the concepts initially and did less well on the earlier tests?
  - Clubber 5y ago
    > > For example, students who come into a course with little prior knowledge earn lower grades at the start, which means they get a lower final average, even if they ultimately master the material. >For others who came in knowing it, they didn't have to 'really study'. And this still holds true for every successive test. Ya I see your point. The first test was the easiest and the most likely to get an A assuming I didn't know it. Of course if you study for the first test and get an A because it's the easiest and someone who already knows the material doesn't study and gets an A, then one isn't lower than the other. In other words, this gap only seems to apply if both groups (familiar and unfamiliar) don't study much or at all on the first test.
NWoodsman 5y ago
I failed Physics 101 in my first semester of college because in the first week the professor was covering everything to know about vectors, velocity, acceleration, etc. I never heard the word "vector" before, much less understood what a velocity vector is. Was I to blame for not knowing that? I certainly suffered the consequences, but it seemed like a failure on behalf of the teaching staff and curriculum.
- gruez 5y ago
  Presumably they thought it was already covered in high school physics/math?
  - throwawayboise 5y ago
    Yes. All college courses, even "100" (intro/freshman) level, have some expectations about what the students already know. Students who don't have that background should take remedial courses or go to junior college or just be prepared to do a lot more outside work on their own. It should not be a university's job to "fix" unprepared students, whether due to the student's prior indifference to his education, or because that eduction was provided poorly.
- larkost 5y ago
  Did you raise your hand after a couple of mentions and ask what a velocity vector was? Or go see the professor either right after class, or at their next office hours? Or talk to a TA about this? Those all seem like appropriate solutions to the problem, whereas your response here, blaming the teaching staff and curriculum, seems to indicate you did not do what you should have.
  - seba_dos1 5y ago
    Kinda off-topic, but as someone with social anxiety and (strongly suspected, but not confirmed yet) ADHD I'm a bit jealous of people for whom raising a hand to ask a question during a lecture or using professor's office hours is just a simple decision away (provided they care enough to actually do that). For me it's mostly like these options didn't exist at all even if I actually wanted to use them, for several reasons. I wonder what could be done there to better accommodate neurodiversive students.
yboris 5y ago
In my math department many professors said that as long as we can do well on the final, that will be the heaviest-weighted of our grades. It seems sensible, at least at the college level, to have 90% of the grade depend on the (cumulative!) final.
- Auracle 5y ago
  I guess the one big gotcha is that if you’re having a bad day - you’re sick or couldn’t sleep the night before - that could really affect you negatively.
  - lostcolony 5y ago
    While we're in the land of hypotheticals, it also seems sensible to plan for a makeup test for people who can't make the main one, and not make it require a signed affidavit from a senator, a doctor, and a notable charity to take that one.
  - MikeTheGreat 5y ago
    I think this is one of the reasons why a lot of students prefer assessment models that reward 'effort over time' - taking 5 tests over 15 weeks helps reduce the downside of doing badly on a single exam/homework/etc. (Which is not to say that all students prefer this approach, but for those that do prefer it this is a common reason) Of course, this then leads us back to the problem of assessing people early on in the term before they've had a chance to build up some momentum in their knowledge & learning. But yeah - there's this sea-saw balancing act between a small number of higher-stakes assessment vs. spreading the assessment out & assessing earlier as well as later when designing assessments in formal education.
  - bglazer 5y ago
    It's also extremely anxiety inducing (for some people)
    - yboris 5y ago
      One of my professors (a jolly fellow) joked about us having "street cred" based on our past work. So if we've been reliably good at proofs, when another proof had gaps, he said he'd assume we knew but just went too quickly, but someone with lower "street cred" would just get marks off ;)
vondur 5y ago
Some of my Chem professors would give you the grade of your final exam if it were higher than your grade at that point in time. They figured if you pulled off a better final exam score, then you deserve that as your final grade as it's a cumulative exam. They also mentioned it rarely happens.
- dehrmann 5y ago
  I'm pretty sure I had a history professor unofficially do something like this. I got B's on the midterms and needed a 100 on the final to get an A. I knew I did better on the final and couldn't think of a question I missed, but 100 still seemed unlikely. I suspect I got a ~98, and he ran with the logic that if you're demonstrating a very solid understanding of the material on the final and close to a cutoff, you earned the higher grade. Also, history. No one was getting 98% on my physics or math exams.
  - Cpoll 5y ago
    That might have just been a standard bell-curve. If you got the best score, you might have been bumped up to 100%.
- milesdyson_phd 5y ago
  This was standard in the Chemistry department at my university. It also saved my ass twice.
choeger 5y ago
But how do you measure progress, then? If you are already on the backfoot, don't you want some kind of concrete measurement of how you improve? Admitted, this would require to use grades slightly differently (not as aggregation towards an end result) and it would require the teachers to repeat tests multiple times. But shouldn't that be how learning is organized?
- etataetaet 5y ago
  I think most people can measure progress without grades! For me at least, getting a 90% on a test is usually the same as a 100%. That 10% difference is almost always just silly math errors or simple misunderstanding and not a fundamental issue.
  - choeger 5y ago
    I don't think that most people can do that. In fact, I think only the ones that already understand a topic can realistically give an estimate of the percentage of their understanding. Someone who "got" 25% or 45% of some material will usually not see their progress.
    - etataetaet 5y ago
      Fair enough! If you don't understand the material it makes sense you wouldn't be able to properly judge how well you're doing. Thanks for the thoughtful response
  - MikeTheGreat 5y ago
    I'm going to do the obligatory thing & cite the Dunning-Kruger Effect here :) Also, from personal experience as someone who's taught computer programming (US college 1st & 2nd year courses - listed as "computer science" but really mostly computer programming): I've seen a _lot_ of people tell me some variation on "My program won't compile, I'm totally stuck, but those are just (implied minor) details - I really do understand what's going on!". Bonus points for when their code indicates that they've missed huge concepts at both the language level ("What do you mean I can't do arrayName = arrayName + 1 in Java?") and/or algorithmic level ("The variable defined by the outer loop of the bubble sort was never used so clearly that loop was useless so I removed it"). Ok, I'll stop ranting :) My point being that you may be able to measure your own progress but not everyone is. Plus, certification / credentializing is one of the services that education provides to the the rest of society.
jkubicek 5y ago
Freshmen year of college, I was part of an experimental program that solved this exact problem. At the beginning of the semester we were given a large binder filled with competencies we were expected to master. Every question on every test was labeled with a competency that the question assessed. Once you demonstrated knowledge of the competency you no longer needed to demonstrate it again. You marked it as complete in your binder, attached the graded test as evidence, and turned it in to your advisor. Advanced students could go into the final, answer only the questions that tested competencies they hadn't previously mastered, then turn in their final after just a few minutes of work. Students who struggled early in the semester could rally through the last few tests, ace all the questions, and demonstrate enough of the required competencies to end the semester with a decent grade. It was an intensely _weird_ way to be graded and required a massive amount of overhead and bookkeeping from both students and faculty. I'm not sure it was a good idea, but I did do much better that year than any of the later years. My freshmen 4.0 sustained my total GPA through some much more difficult times.
- noasaservice 5y ago
  That actually sounds like the Diablo 2 skill tree. And that sounds awesome.
  - slowmovintarget 5y ago
    If the university did it for all subjects, it would look like Path of Exile's skill graph [1]. In the game, when you graduate to high-level game play, you get a "whole-nuther" graph like the first to start populating. [1] https://poeskilltree.com/?v=3.17.0#AAAABAMAAA== https://poeskilltree.com/?v=3.17.0#AAAABAMAAA==
mrosett 5y ago
The unwritten policy in many math departments is that if you get an A on the final you get an A in the class.
- Spivak 5y ago
  It's explicitly written in basically all of my math classes. Your grade was always the better of your final or your coursework. I get the perverse incentive of fucking off and taking the final but nobody actually does that, shits hard yo. The game is actually trying to get your grade high enough that if you bomb the final you're still good.
eyelidlessness 5y ago
That is an equity issue. Those students who are disadvantaged at the start are often disadvantaged in several ways. And it perpetuates itself: those lower scores will be used to hold the same students back or otherwise decelerate learning, leaving them once again starting off disadvantaged the next term. It also shapes their perception of their ability to advance, and importantly it shapes that perception for their educators and their parents/guardians.
MikeTheGreat 5y ago
It's worth pointing out that the 'vision' for formal education is that you show up, the teacher teaches you what you need to know, and then the grades are supposed to assess how well you learned the material that the teacher taught you. I'm not claiming that it always works this way; a long-term transformative process isn't going to be as simple as "show up -> get taught -> git gud". There are substantial and structural issues with how formal education works - for example, it's well known amongst people who study education in the US that the single biggest predictor of your grades is the socio-economic level of your parents. But please let's not forget the "teaching" part of education :)
Aeolun 5y ago
Recently someone on here suggested something like taking either the grade on the final test, or the weighted average grade of the coursework, whichever is higher. That seems like it would work to me.
zajio1am 5y ago
I do not know about US university customs, but on a university (in Europe) which i attend the only purpose of intermediate grades for a course (if they are used) was to assess whether a student is allowed to go for the final exam of the course, and then only the the grade from the final exam counted.
- pkaye 5y ago
  In my experience at US universities, our course grades were determined by 2-3 midterm exams and a final exam. A common weighting scheme was 3 x 20% for the midterms and 40% for the final exams. Sometimes they had 10% weight for homework, labs, projects. There were some courses where there was a final exam but the rest as projects, papers and presentations.

Double_a_92 5y ago

How does it disadvantage people that start from scratch though? Tests should be written in a way that everyone can pass them, with the knowledge thought in the course. If people that already have some related knowledge have an easier time studying, that's just life. How does it unfairly disadvantage you, if somebody else is already more skilled than you?

matt321 5y ago
>How does it disadvantage people that start from scratch though? In the same way that not being a genius is a disadvantage. Cant change it unless we handicap the smart people.
- jdkee 5y ago
  “ In the year 2081, the 211th, 212th, and 213th amendments to the Constitution dictate that all Americans are fully equal and not allowed to be smarter, better-looking, or more physically able than anyone else. The Handicapper General's agents enforce the equality laws, forcing citizens to wear "handicaps": masks for those who are too beautiful, loud radios that disrupt thoughts inside the ears of intelligent people, and heavy weights for the strong or athletic.” See https://en.m.wikipedia.org/wiki/Harrison_Bergeron https://en.m.wikipedia.org/wiki/Harrison_Bergeron
conradludgate 5y ago
> earn lower grades at the start, which means they get a lower final average They get graded from the very beginning, not from a single final exam
- Double_a_92 5y ago
  So? Each small test should be written in a way that you can ace with what you learned so far in the course. If you are not willing to study that, that's your personal problem.
mckeed 5y ago
They are talking about assignments, not tests. It applies more to skill-based courses than knowledge-based ones. If you're writing C papers at the beginning and A papers at the end, should you get an A or B?
- Double_a_92 5y ago
  Why would you be writing C papers though? Either the tasks are badly designed, in a way that you can't properly solve them with what you have been thought so far... or you have all the possibility to write As in the beginning, even if you don't have prior knowledge.
asoneth 5y ago
The argument is that giving weight to the grades earned throughout the course may under-value students who made significant improvements. For example, take two students: Student A comes in barely meeting the prerequisites and struggles with low grades on initial tests and projects. They come to office hours, put in substantial effort, and eventually ace the final projects and tests. Student B comes in with substantial background experience and gets decent but not stellar grades on all tests and projects. Depending on how much weight is given to the final tests/projects student B may still end up with a higher grade despite the fact that they probably do not have as strong an understanding of the material as student A. I've had both kinds of students the handful of times I taught graduate-level seminars and it's one reason I weighted the final projects more heavily than initial projects. (Though I found grades on initial projects were still useful for other reasons.)
enragedcacti 5y ago
Imagine a course as teaching the content along a linear function from 0 to N understanding. With traditional grading we are effectively determining the area under the line for the given class, even though the goal of the class is to leave with a mastery of the content. Consider two students, one whose education path in the course is exponential and the other whose path is logarithmic where both paths intersect at the end of the course. According to traditional grading the logarithmic student would score far better in the course despite their knowledge being equal to the other student by the end. Which student would you rather be? who has more potential? Should the body of a course be treated like a series of 100m sprints or a marathon, and what is the net positive outcome through each approach?
MiddleEndian 5y ago
I think his point is that if Student A started with 50% knowledge and Student B started with 80% knowledge and by the end they both have 95% knowledge, Student A's grade for the course is penalized for the slow start even AFTER reaching mastery by the end. If Student A ends up only at 80% then so be it, but if they end up matched then they should end up with a similar grade.
orwin 5y ago
This very year, my sister started college after 5 years as a chef. Her math was shit, especially since it was already not her strongest subject in highschool, 5 years prior. She started very low and ended up average (she's going into biology). With help from me and Wolfram alpha (sry about that advertisement but it helped a lot). She was graded below average even though she showed more progress than at least half her class (since she started almost at the bottom and ended up in the middle). She doesn't care, but still think she is bad at math, when she just had poor teachers in HS and showed learning abilities at 23.

Aachen 5y ago

How so? If someone already knows the material or has a very solid framework to fit the extra knowledge into and needs zero catching up, of course they'll know the material better at the end of the course and/or will be able to apply it better. How should that not translate into a better grade? The only alternative I see is to let those less-prior-experienced students take a longer time and only then determine the final grade.

etchalon 5y ago
Or you could grade for mastery of material, instead of the individual steps taken to get there.
rhn_mk1 5y ago
That presumes people learn at equal rates. People don't learn at equal rates though.
octokatt 5y ago
Because the grading system _encourages_ students who already know the material, and _discourages_ students who don’t know the material. If someone is discouraged from a subject, they are less likely to put in the effort to succeed in the subject. This creates a self-perpetuating cycle, which ideally education would fix. Similarly, this sets students up for a false idea of their “good subjects”, when instead those are familiar subjects. A student who could be a polymath gets discouraged from pursuing multiple subjects because they’re already far ahead of the game in math, so they pigeon-hole themselves to the subject where they receive positive reinforcement. Essentially, grades as a feedback system are less than ideal, and we can do better.
- Aachen 5y ago
  > Similarly, this sets students up for a false idea of their “good subjects”, when instead those are familiar subjects. A student who could be a polymath gets discouraged from pursuing multiple subjects I had never considered it that way, interesting perspective. Not yet sure how this will work in the grander scheme, I'm thinking of the impact on the value of a diploma when it's based on learning ability or broader knowledge as compared to being specialized in one thing. Something I'll be thinking about in the future.

blodkorv 5y ago

How far should we strech this? there are students who come into courses with superior memory or iq and get a head start from that.

jrgoff 5y ago
Not sure what your getting at here? The person you are commenting on wasn't talking about punishing people for having a head start they were saying that people's final grade should be based on what they are capable of at the end of the class, not at the start of it.
- blodkorv 5y ago
  Why not, there will be for sure students who have the ability to grasp a course fast and effectily to compete with those with prior knowledge, why not grade and reward that ability.
  - mech987 5y ago
    If you grade improvement rather than mastery, you incentivize feigning incompetence at the start of the course to game the system. The closest thing I ever saw to this was my high school weightlifting class. Your bench press single rep max was a high percentage of your overall grade, and was tested several times over the course of the semester. You were graded on improvement. The teacher made sure not to mention this before the first test, to keep people from cheating, but a few kids were in the know and faked a low max. From a philosophical/values point of view- what is more important for education- incentivize all students to try to achieve the mastery level expected, or incentivize all students that as long as it looks like they're making progress, that is enough? I'd go with mastery. The appearance of progress is too easy to game, too subjective, and not easily applicable to varying courseloads.
  - lostcolony 5y ago
    It is rewarded, simply by the fact that they then have to put less effort into the class. Why should it be graded? The point of the class is not to learn quickly and effortlessly, but to learn the material completely. If by the time the class is over the student has learned it well, they should get a grade that reflects that learning. I mean, by your logic, should we give extra credit to the students who never show up to office hours in college, since clearly they're learning it 'faster and (more) effectily(sic)'?

micromacrofoot 5y ago

Shouldn't the course be designed in a way where everything on the exam has been taught in the class? People with prior knowledge will still have an upper-hand in everything (even beyond school), so I'm not sure that it's even a solvable problem. I guess perhaps one could give a baseline exam on day 1 to qualify prior knowledge and give a handicap to people if they're coming in blind? Though if I knew that purpose I'd game the system, so it probably still isn't fair.

notreallyserio 5y ago

> A scoring system for scholastic ability that disadvantages those that start off with little knowledge, even when they do well by the end of the course, is ridiculous on its face. This is one reason I did poorly in school and eventually dropped out. Teachers chose (either independently or by arrangement) to weigh all grades equally, whether they were at the start of the class or the end. So if you did poorly at the start, you might as well give up, you'll never dig yourself out of that hole.

specialp 5y ago

I had such a situation when I first entered college. I came from a disadvantaged background, and was not a good writer. My freshman composition class was taught by a woman who had us write synopses of what I thought of at the time as boring, long-winded New Yorker articles. She ruthlessly hammered me on grammatical errors. I ended up withdrawing from the course. Then next semester I had a the same class but a different professor. He was much different. He had us write about our life experiences and things that interested us. You would get the grammatical corrections as feedback but you would not get blasted for them. Over that one semester I used the literary handbook I was given, and got a lot better at it. It really helped me. In my future liberal arts classes I had to take, I got good grades on my papers and they were well received.

Auracle 5y ago
I was required to take an art course in college, and I chose an intro level graphic design class. I’m terrible at all things art other than photography, but I thought graphic design made sense being a CS major an all. My experience was terrible. There was no real artistic instruction. For instance, during our first or second week we were told how to use the pen tool in Illustrator and or homework was to make a logo for a made up company. I, having absolutely no art background, did a bad job. She actually projected my work in front of the class and had me stand up in front to be ridiculed. I decided to drop the class. When I brought her the form to sign all she said was “thank you,” which I very much took to mean “thank you for dropping out of my class.” Heaven forbid you actually need to do your job, professor.
- sitkack 5y ago
  I am sorry for your experience. Most college level art classes have a pretty strict entrance requirement where portfolios are judged pretty harshly from incoming high school students. Incoming students are basically already required to be good. This instructor was not prepared for you and treated you poorly. Another instructor might have handled it much better.
  - navbaker 5y ago

Clubber 5y ago

> For example, students who come into a course with little prior knowledge earn lower grades at the start, which means they get a lower final average, even if they ultimately master the material. One of my strategies in college was to really study on the first exam. The logic being is it's just introductory material and therefore the easiest exam to make an A on. Subsequent exams tended to get into the nitty gritty and require a lot more information to be understood / retained. This held true on any course. In my experience, this negates the quoted point that people who had little prior knowledge would do poorly early on. Granted I went to college a long time ago. Things may have changed.

lostcolony 5y ago
> One of my strategies in college was to really study on the first exam Why did you really have to study if it was so easy? Right, because you didn't know it. For others who came in knowing it, they didn't have to 'really study'. And this still holds true for every successive test. I feel like that kinda negates your negation. To speak of specifics, I took calc 1 twice. First time was due to inconsistencies in the teacher's grading, but whatever; the second time I took it, I was very familiar with the concepts, and made an easy A. Others struggled. It's fair to say I should have gotten the A at the end because I knew the material, and it's fair to say the people who got an A on every test (those 'easy' first ones you mention being included) should receive an A, but was it fair to say others, who by the end had the same understanding I and the other A students did, should have gotten a B even if they aced a cumulative final, because they struggled with the concepts initially and did less well on the earlier tests?
- Clubber 5y ago
  > > For example, students who come into a course with little prior knowledge earn lower grades at the start, which means they get a lower final average, even if they ultimately master the material. >For others who came in knowing it, they didn't have to 'really study'. And this still holds true for every successive test. Ya I see your point. The first test was the easiest and the most likely to get an A assuming I didn't know it. Of course if you study for the first test and get an A because it's the easiest and someone who already knows the material doesn't study and gets an A, then one isn't lower than the other. In other words, this gap only seems to apply if both groups (familiar and unfamiliar) don't study much or at all on the first test.

NWoodsman 5y ago

I failed Physics 101 in my first semester of college because in the first week the professor was covering everything to know about vectors, velocity, acceleration, etc. I never heard the word "vector" before, much less understood what a velocity vector is. Was I to blame for not knowing that? I certainly suffered the consequences, but it seemed like a failure on behalf of the teaching staff and curriculum.

gruez 5y ago
Presumably they thought it was already covered in high school physics/math?
- throwawayboise 5y ago
  Yes. All college courses, even "100" (intro/freshman) level, have some expectations about what the students already know. Students who don't have that background should take remedial courses or go to junior college or just be prepared to do a lot more outside work on their own. It should not be a university's job to "fix" unprepared students, whether due to the student's prior indifference to his education, or because that eduction was provided poorly.
larkost 5y ago
Did you raise your hand after a couple of mentions and ask what a velocity vector was? Or go see the professor either right after class, or at their next office hours? Or talk to a TA about this? Those all seem like appropriate solutions to the problem, whereas your response here, blaming the teaching staff and curriculum, seems to indicate you did not do what you should have.
- seba_dos1 5y ago
  Kinda off-topic, but as someone with social anxiety and (strongly suspected, but not confirmed yet) ADHD I'm a bit jealous of people for whom raising a hand to ask a question during a lecture or using professor's office hours is just a simple decision away (provided they care enough to actually do that). For me it's mostly like these options didn't exist at all even if I actually wanted to use them, for several reasons. I wonder what could be done there to better accommodate neurodiversive students.

yboris 5y ago

In my math department many professors said that as long as we can do well on the final, that will be the heaviest-weighted of our grades. It seems sensible, at least at the college level, to have 90% of the grade depend on the (cumulative!) final.

Auracle 5y ago
I guess the one big gotcha is that if you’re having a bad day - you’re sick or couldn’t sleep the night before - that could really affect you negatively.
- lostcolony 5y ago
  While we're in the land of hypotheticals, it also seems sensible to plan for a makeup test for people who can't make the main one, and not make it require a signed affidavit from a senator, a doctor, and a notable charity to take that one.
- MikeTheGreat 5y ago
  I think this is one of the reasons why a lot of students prefer assessment models that reward 'effort over time' - taking 5 tests over 15 weeks helps reduce the downside of doing badly on a single exam/homework/etc. (Which is not to say that all students prefer this approach, but for those that do prefer it this is a common reason) Of course, this then leads us back to the problem of assessing people early on in the term before they've had a chance to build up some momentum in their knowledge & learning. But yeah - there's this sea-saw balancing act between a small number of higher-stakes assessment vs. spreading the assessment out & assessing earlier as well as later when designing assessments in formal education.
- bglazer 5y ago
  It's also extremely anxiety inducing (for some people)
  - yboris 5y ago
    One of my professors (a jolly fellow) joked about us having "street cred" based on our past work. So if we've been reliably good at proofs, when another proof had gaps, he said he'd assume we knew but just went too quickly, but someone with lower "street cred" would just get marks off ;)

vondur 5y ago

Some of my Chem professors would give you the grade of your final exam if it were higher than your grade at that point in time. They figured if you pulled off a better final exam score, then you deserve that as your final grade as it's a cumulative exam. They also mentioned it rarely happens.

dehrmann 5y ago
I'm pretty sure I had a history professor unofficially do something like this. I got B's on the midterms and needed a 100 on the final to get an A. I knew I did better on the final and couldn't think of a question I missed, but 100 still seemed unlikely. I suspect I got a ~98, and he ran with the logic that if you're demonstrating a very solid understanding of the material on the final and close to a cutoff, you earned the higher grade. Also, history. No one was getting 98% on my physics or math exams.
- Cpoll 5y ago
  That might have just been a standard bell-curve. If you got the best score, you might have been bumped up to 100%.
milesdyson_phd 5y ago
This was standard in the Chemistry department at my university. It also saved my ass twice.

choeger 5y ago

But how do you measure progress, then? If you are already on the backfoot, don't you want some kind of concrete measurement of how you improve? Admitted, this would require to use grades slightly differently (not as aggregation towards an end result) and it would require the teachers to repeat tests multiple times. But shouldn't that be how learning is organized?

etataetaet 5y ago
I think most people can measure progress without grades! For me at least, getting a 90% on a test is usually the same as a 100%. That 10% difference is almost always just silly math errors or simple misunderstanding and not a fundamental issue.
- choeger 5y ago
  I don't think that most people can do that. In fact, I think only the ones that already understand a topic can realistically give an estimate of the percentage of their understanding. Someone who "got" 25% or 45% of some material will usually not see their progress.
  - etataetaet 5y ago
    Fair enough! If you don't understand the material it makes sense you wouldn't be able to properly judge how well you're doing. Thanks for the thoughtful response
- MikeTheGreat 5y ago
  I'm going to do the obligatory thing & cite the Dunning-Kruger Effect here :) Also, from personal experience as someone who's taught computer programming (US college 1st & 2nd year courses - listed as "computer science" but really mostly computer programming): I've seen a _lot_ of people tell me some variation on "My program won't compile, I'm totally stuck, but those are just (implied minor) details - I really do understand what's going on!". Bonus points for when their code indicates that they've missed huge concepts at both the language level ("What do you mean I can't do arrayName = arrayName + 1 in Java?") and/or algorithmic level ("The variable defined by the outer loop of the bubble sort was never used so clearly that loop was useless so I removed it"). Ok, I'll stop ranting :) My point being that you may be able to measure your own progress but not everyone is. Plus, certification / credentializing is one of the services that education provides to the the rest of society.

jkubicek 5y ago

Freshmen year of college, I was part of an experimental program that solved this exact problem. At the beginning of the semester we were given a large binder filled with competencies we were expected to master. Every question on every test was labeled with a competency that the question assessed. Once you demonstrated knowledge of the competency you no longer needed to demonstrate it again. You marked it as complete in your binder, attached the graded test as evidence, and turned it in to your advisor. Advanced students could go into the final, answer only the questions that tested competencies they hadn't previously mastered, then turn in their final after just a few minutes of work. Students who struggled early in the semester could rally through the last few tests, ace all the questions, and demonstrate enough of the required competencies to end the semester with a decent grade. It was an intensely _weird_ way to be graded and required a massive amount of overhead and bookkeeping from both students and faculty. I'm not sure it was a good idea, but I did do much better that year than any of the later years. My freshmen 4.0 sustained my total GPA through some much more difficult times.

noasaservice 5y ago
That actually sounds like the Diablo 2 skill tree. And that sounds awesome.
- slowmovintarget 5y ago
  If the university did it for all subjects, it would look like Path of Exile's skill graph [1]. In the game, when you graduate to high-level game play, you get a "whole-nuther" graph like the first to start populating. [1] https://poeskilltree.com/?v=3.17.0#AAAABAMAAA== https://poeskilltree.com/?v=3.17.0#AAAABAMAAA==

mrosett 5y ago

The unwritten policy in many math departments is that if you get an A on the final you get an A in the class.

Spivak 5y ago
It's explicitly written in basically all of my math classes. Your grade was always the better of your final or your coursework. I get the perverse incentive of fucking off and taking the final but nobody actually does that, shits hard yo. The game is actually trying to get your grade high enough that if you bomb the final you're still good.

eyelidlessness 5y ago

That is an equity issue. Those students who are disadvantaged at the start are often disadvantaged in several ways. And it perpetuates itself: those lower scores will be used to hold the same students back or otherwise decelerate learning, leaving them once again starting off disadvantaged the next term. It also shapes their perception of their ability to advance, and importantly it shapes that perception for their educators and their parents/guardians.

MikeTheGreat 5y ago

It's worth pointing out that the 'vision' for formal education is that you show up, the teacher teaches you what you need to know, and then the grades are supposed to assess how well you learned the material that the teacher taught you. I'm not claiming that it always works this way; a long-term transformative process isn't going to be as simple as "show up -> get taught -> git gud". There are substantial and structural issues with how formal education works - for example, it's well known amongst people who study education in the US that the single biggest predictor of your grades is the socio-economic level of your parents. But please let's not forget the "teaching" part of education :)

Aeolun 5y ago

Recently someone on here suggested something like taking either the grade on the final test, or the weighted average grade of the coursework, whichever is higher. That seems like it would work to me.

zajio1am 5y ago

I do not know about US university customs, but on a university (in Europe) which i attend the only purpose of intermediate grades for a course (if they are used) was to assess whether a student is allowed to go for the final exam of the course, and then only the the grade from the final exam counted.

pkaye 5y ago
In my experience at US universities, our course grades were determined by 2-3 midterm exams and a final exam. A common weighting scheme was 3 x 20% for the midterms and 40% for the final exams. Sometimes they had 10% weight for homework, labs, projects. There were some courses where there was a final exam but the rest as projects, papers and presentations.