Monthly Archives: December 2013

[SBG] How I’m Doing Exams

This is my first year using SBG and, as a result, I’m looking at grading totally differently.  I got to the final and realized that it doesn’t make sense to be using a 1-4 mastery system working on specific standards all year and then at the end of the year just grade the final out of 50 or 100 points or whatever.  So instead, I’m simply making the final a continuation of the year but with a twist.  Student get to choose what they want on their final.

What are you, nuts?

Yes, but let me explain the setup first.  First of all, students earn a 1, 2, 3, or 4 on every assessment throughout the year, depending on how much mastery of the topic they’ve shown me.  Only 3’s and 4’s are passing. Sometimes I give multiple grades on an assessment because multiple standards were tested (example: labs).  Students can sign up to show me a previous standard and they essentially re-take a quiz[1].  I take the last two grades in any standard, average them, and that’s their grade for that standard.

At the end of the year, for most students, there are plenty of standards for students to improve.  So I let them sign up for which standards they want on their exam.  However, it’s a bad idea for them to only sign up for topics they have 4’s on already.  I explain it to the students like this:

Because I average your grades, then if you just sign up for things you already have 4’s on, then you can’t improve your overall grade at all, you can only do worse!  But if you only sign up for things that you only have 1’s, even if you totally bomb the exam, you can’t do worse!  And the great thing is I test for understanding, so you’re bound to only improve your grade!  So please sign up for the topics that you are doing the worst in and study for them!

Now, I did tell the students that they only get to sign up for 3 standards.  In Precalculus, I’m giving them 3 other standards, so they do have to be ready for everything throughout the year, but practically I’m trying to help students out here so I’m mostly giving standards that a large part of the class has room to improve.

Dude, but that means that you’ve got to make X number of different exams!  That would take days!!

Not quite.  In practice, the exam becomes just a bunch of quizzes put together.  So creating the exam just means I need to make one quiz for each standard.  This year I got through 19 standards and students only signed up for about 14 of those.  That means once I created those 14 quizzes, I just had to stand next to the copier putting together each student’s individualized exam by copying out the right quizzes.  I actually did that today and for my 1st period class of only 15 students, it took me 30 minutes (after I had all the quizzes created, which took a few hours because this is the first year I’ve done this).  So if the average teacher has 100 to 120 students, it would only take about 2 hours to put together all the exams: not too bad.  True, it’s not your multiple-choice exam “press print and collect the exams”, but I think this time is well invested.

Grading them will be another beast to tackle, but I’m not too worried.  The way I grade is by giving one grade measuring a student’s understanding across an entire quiz, so it’s actually faster than tallying up points.  Often times I know what a student will get before I grade every problem, and during the year I only continue grading every problem so that students have a complete quiz to reference and work on to improve.  Because they won’t get their exam back, I might not have to grade all of every quiz[2].  Again, it won’t be the super-fast “run-them-through-the-scantron-multiple-choice”, but I believe that open-ended problems where students show their work is a much better assessment of a student’s abilities than multiple choice questions.

But how do you know students have retained all the information if you’re not testing them on everything at the end of the year?

Ideally I’ve already re-tested students multiple times on every standard, so students that have 3’s and 4’s legitimately retained the information and the need for a traditional (summative) exam diminishes.  Unfortunately, I didn’t assess as much as I could and should have this year[3].  So I’m putting the most essential parts of the class on the exam.  The more experience I have with this grading system, the more comfortable I’ll be that I really tested whether students had retained the information before the exam rolls around.

Other Little Details

I used Google Forms to record students’ choices for exams.  And no, not every student signed up for standards on their exam even though we did it as a class on a day that every student was present.  But if they didn’t sign up, that only hurts them because then I just chose the standards I thought would be most important for them to know.[4]

I haven’t given the exams so I’m really curious to see whether students’ grades significantly change.  Fortunately I have an awesome principal who allowed me to try this out because he understands my system and agrees that this would be a good way to test students at the end.  We’re given a lot of freedom because our principal trusts us.  Which is awesome and what really should happen everywhere.

I should probably add that this idea is probably influenced by conversations with Jim Pai (@paimath) who teaches in Canada and those guys have much better ideas about the way remedial teaching should go.  First of all, the entire system uses SBG.  When a student fails a class because they didn’t get one part of the class, then the student takes summer school, but only on the topic(s) that they failed in.  So they don’t have to retake Algebra I if they failed to learn how to isolate variables, they just go to summer school and learn that topic.  Why can’t American schools do that?


[1] By “retake”, I mean take the same standard, but there are different questions.  Ideally there are different types of questions, but realistically that hasn’t happened enough.  But this is just my first year doing this so I’m hopeful for the future.

[2]But I still probably will because of my slightly OCD nature.

[3] Even though I broke my Google spreadsheet because I recorded over 100 different grades for students, but more on that in a to-be-written post.

[4] If I was feeling nice, then I looked at their grades as I made their decision.  Actually, if I was feeling nice AND I had time, then I did this.  If I didn’t have time, I just copied another student’s exam.


Filed under Teaching

Quiz Corrections

Before I started Standards Based Grading, I used to use Test Corrections where students could make up half of the points.  Now that I don’t use points (or give “tests”), that system didn’t make sense, so I didn’t use it.  However, I really liked the idea of students reflecting on their mistakes, as well as the chance to fix their mistakes, so I decided to bring it back with “Quiz Corrections”.

If students do the following three steps:

  1. Get the right answer.
  2. Show your work (or explain why it’s the right answer).
  3. Write two (2) sentences why you got it wrong.

then they get a 4 (mastery) on the standard.  This only works because I take the average of their last two grades in a standard, so getting a 4 after getting a 1 gives them a 2.5 (still not passing), which means to actually earn a 4, they have to come in and re-take the standard and get a 4 on the quiz (which I’m cool with).  Here’s a list of reasons why I think this is a good system:

  • Students must do every single problem in order to earn the 4–no partial credit.  This rewards students who were “pretty close” on the quiz because they have less to do.
  • Before this, students had to retake a quiz twice in order to have a chance at getting a 4 if they didn’t the first time around. Now they have the option to spend some time reflecting on what they got wrong and then taking a quiz to demonstrate understanding later.  They’re looking at the material at least 2 more times on their own outside of class, so I win.
  • This is good for students who prefer to write and who can improve through writing.  The more options students have to demonstrate mastery, the better the students are for it.
  • I do put a time limit on it of “within a week of me handing it back” so I don’t get a bunch of really old quizzes that I have to remember how to grade.
  • Students have (another) reason to pay attention in sessions when their peers are teaching them material that we’ve already learned.
  • Some students were frustrated by the fact that they had to take 2 quizzes.  They would rather spend much more time doing something to make sure they get a 4 than studying and perhaps getting another low grade.  (They don’t realize that effective studying would essentially guarantee them a 4 on the quiz.)

Below is the paper that I give to the students.  I think it’s a bit wordy, and I need to get better at striking the fine line between “informative enough” (or covering possible questions) and information overload so they throw the sheet out the window first chance they get.  Next year I’ll definitely shorten it, but this is good for now.


Filed under Teaching