Dropping Lowest Test Grade

[HigherEd7]

How do you feel about dropping the lowest test grade?

[Caracal]

How do you feel about dropping the lowest test grade?

There's nothing inherently wrong with it or right about it. It is mostly just a question of the overall structure and learning outcomes of a course and how those fit in with the grading scheme. One thing to consider is that in my classes it didn't lead to that many more As but did help a lot of people get Bs instead of Cs. I haven't done it for the last couple of semesters because I thought it was helping bad students out too much. I've been thinking about a modified version where students can choose to take a make up I give already to people who missed an exam, but they have to keep the grade they get on that instead of their lowest exam whether or not the grade is better.

[LibbyG]

Second to everything Caracal said.

An alternative I've used is to count the highest test grade twice in calculating that part of the course grade. It may not make a huge numerical difference (and mostly at the bottom of the range), but it helps dispel the idea that "One bad test sank my whole course grade!" It also helps students find the motivation to finish strong even in a rough semester.

[marshwiggle]

It also means not having to deal with as many excuses due to absence etc. "I'm dropping one quiz anyway" cuts off a lot of begging and explaining.

[Aster]

For clinical and practicum-based courses (e.g. many laboratory type courses) this is not advised, or even permitted.

There are also certain restrictions and recommendations for "final exams" where you really should not be letting students "opt out" or "drop" them. Cumulative or capstone-based final exams, for examples.

But if it's not a final exam, and not a practicum-based course, the option to drop major assessments from gradebooks can allow greater versatility in curriculum design. Many of my colleagues practice this and for (some) of those professors I respect it. I don't incorporate dropped exam policies myself, but I have alternative strategies to help students pick themselves up after the occasional assessment stumble.

One of the most well-known problems with a dropped exam policy is that there is a large subset of students that can harm themselves by misusing/abusing the policy. What can happen is that random student can decide that "I don't want to learn this stuff I'll just use this as my dropped exam." This mentality may encourage them to not even come to class, study, participate for a large part of the course. This will be a problem in courses using a scaffolded and/or integrated curriculum design. For example, if a student decides to not learn scientific notation for Exam 2 because he/she's "dropping that exam", he/she's going to get screwed when scientific notation knowledge is assumed and expected for Exam #3, Exam #4, etc...

Dropped exam policies can also foster negative learning behaviors that extend throughout the entire course. Students who choose to "take a break from classes" and opt-out of an early-term exam may find themselves trapped in a pit that they cannot (or will not) climb back out of. The rationalizations that they made for not studying for one exam now manifest into not studying for additional exams. It can turn into a positive feedback loop of failure.

So to me, dropped exams are a risky and restrictive move. You may have to simplify your curriculum plan to compartmentalize exam content. You might have to give a non-cumulative final exam to keep things "fair". You might incorporate more exams or reduce the value of exams in lieu of other assessments.

But yes, lots of professors use a dropped exam policy, and some of them even do it successfully.

[lightning]

Some faculty like dropping the lowest test grade because they then don't have to deal as much with make-up tests. This is especially useful at schools where students call in sick when they feel that they are under-prepared for a major test, and there is no testing center service that can take up the slack/slackers.

[marshwiggle]

So to me, dropped exams are a risky and restrictive move. You may have to simplify your curriculum plan to compartmentalize exam content. You might have to give a non-cumulative final exam to keep things "fair". You might incorporate more exams or reduce the value of exams in lieu of other assessments.

But yes, lots of professors use a dropped exam policy, and some of them even do it successfully.

In my case, I don't announce the drop policy in advance; i.e. I tell students about it when they come to ask, so it doesn't encourage the "preemptive drop" problems Aster mentioned. But it applies to everyone, unlike excused absences which some people use instead which only potentially benefit those who ask for them.

[HigherEd7]

Great responses and ideas by everyone.

[Chemystery]

I replace the lowest exam score with that of the cumulative final exam, if it is higher.  I usually do not tell my class about this until after the last midterm exam.  That said, one of the courses I teach is a two-semester course, so they already know about the policy going into second semester.  So far, I haven't seen any evidence that those students are blowing off an exam because they know it can be replaced. 

[Antiphon1]

I allow corrections on exams which results in raising the grade by 1/2 the points missed.  There is no moral or ethical reason to choose corrections over dropping a grade or vice versa.  My personal preference is requiring effort for awarding points.  Further, students don't learn anything from failing a test; they will learn something finding the answers to questions they didn't know.  On the same tack, I also don't curve grades at the end of the semester.  Students are given adequate opportunities to earn extra points if they are willing to do the work.  It depends on how you view awarding grades.  I believe grades should mostly reflect the student's effort not my willingness to dispense mercy.

[downer]

There is no function to drop the lowest grade on Blackboard gradebook, so it is not something I would consider doing.

[polly_mer]

There is no function to drop the lowest grade on Blackboard gradebook, so it is not something I would consider doing.

Psst: https://kb.uc.edu/KBArticles/Blackboard-DropLowestGrade.aspx was the first listing on a web search.

Back on topic:

As Aster wrote, the course must be carefully planned to make dropping the lowest exam be a good idea.  The question, as always, is what pedagogical purpose are you trying to achieve with the course design? 

I've taught classes in which pretty much everyone bombs the first exam every term.  A colleague suggested using a group rewrite of the test so that people learn the material as they earn back some points and that worked pretty well for a course that had a large level of group work anyway.  Using that method every exam helped students who were borderline C/D, but were diligently attending, finish the term on the C side.

A different course worked pretty well to have the first exam really early as the wake-up call and just have more exams during the term; mathematically, having one bad exam out of six total is less bad than having one bad exam out of three total.  This method ensured that students who had a slow start weren't doomed.  This also ensured that students could earn an A in the class instead of being mathematically excluded after the first exam.  As Caracal wrote, this tended to help the C students earn Bs more than the D students earn Cs.

A third course worked pretty well to have the final exam be very heavily weighted as cumulative and having the exams during the term be less heavily weighted.  That first bombed exam was then less than 10% of the overall course points so that no one was excluded from an A because of one bad day.  Almost no one ever earned an A in this class, but by the end, most people had done enough to demonstrate being ready to go on to the next course in the sequence.

So, what's the purpose of dropping the exam and is that the only way to achieve that purpose while still helping students learn what they are supposed to learn in the class?

[marshwiggle]

As Aster wrote, the course must be carefully planned to make dropping the lowest exam be a good idea.  The question, as always, is what pedagogical purpose are you trying to achieve with the course design? 

...
  A colleague suggested using a group rewrite of the test so that people learn the material as they earn back some points

...

A different course worked pretty well to have the first exam really early as the wake-up call

...

A third course worked pretty well to have the final exam be very heavily weighted as cumulative and having the exams during the term be less heavily weighted.


....

All* of these (and various other approaches) implicitly assume that students need some room to adapt as the term goes on, and that a good ending is better than uniform performance during the course.

Does anyone feel that isn't a good idea, (i.e. students should start as strong as they end), and if so, why?


(Even things like having the first test early as a wake up call still uses the timing to give students more of a chance to adapt.)

[downer]

There is no function to drop the lowest grade on Blackboard gradebook, so it is not something I would consider doing.

Psst: https://kb.uc.edu/KBArticles/Blackboard-DropLowestGrade.aspx was the first listing on a web search.

Hmm. I will have to check that out in detail. I guess it makes a difference whether test scores are recorded as points or percentages.

Not all my tests are worth the same number of points, nor are they all worth the same proportion of the final total. so it might get complicated.

[Golazo]

In some classes I let the student substitute the final exam grade for the weakest unit grade, if the final is higher. This seems to work better for encouraging students not to simply write a unit off. But I've also been moving away from as many unit exams in favor of other requirements that are easier for me to grade given my teaching load.