Grading Quizzes

Standards-Based Grading

Video Transcript:

In the last video, we learned how to create concept quizzes, and we saw that they are based on three levels of questions. Now it’s time to dive into how to grade these quizzes. Is it similar to grading a traditional test? Are we allowed to give partial credit? Does it take a long time?

Let’s find out.

To begin, let’s get another look at the structure of a typical quiz. There are always 6 questions, and they are split into 3 difficulty levels.

There are 3 80 level questions which represent proficient understanding, 2 90 level questions which represent advanced understanding, and 1 100 level question which represents mastery understanding.

I like to specifically label these sections so there aren’t any surprises in how they are graded. Students know what they need to do to earn each grade, and the sections make grading fast and easy for teachers.

So, how do students earn their grades?

In general, students need to get all 3 80 level questions correct to make an 80. Then, they need to get both 90 level questions correct as well to receive a 90. Finally, they need to get the 100 level question correct in addition to the 80 and 90 level questions to receive a 100.

That’s the simple formula for grading quizzes, but it leaves us with quite a few questions. What happens if students only get some but not all of the 80 level questions correct? What if a student gets the 100 level correct but misses other questions?

And do students always need to attempt all 6 questions?

Let’s start with the latter question first. Do students need to attempt all 6 questions? Or can they just work on certain levels?

The short answer is yes, students always need to attempt all 6 questions. This is true for several reasons. First, if a student truly has mastery, 100 level understanding, then they should be able to answer all 6 questions correctly, not just 1 challenging question.

On the flip side, I never want a student to simply settle for proficient, 80 level understanding. So if a student were to try and correctly answer 3 80 level questions and stop, they wouldn’t have the opportunity to challenge and push themselves to think through more advanced questions. Proficiency isn’t the goal. Deep understanding is.

Therefore, on either end of the spectrum, students must attempt all 6 questions. If they don’t, I won’t accept the quiz.

Okay, now to the fun part. How do we score quizzes that don’t fit neatly into our simple formula of getting all questions correct in certain categories?

I first want to note that you are free to use your professional judgment. My thinking may be different from yours, and that’s okay. The key is making sure you are consistent with all your students. Once you set certain benchmarks, stick with them.

With that said, here’s where I landed.

Let’s look at several cases, and for each case, I’ll tell you the score I’d give and my reasoning for it.

Case 1. Students get all 3 80 level questions correct. In this scenario, we know the student will earn at least an 80 because they have shown proficient understanding. Now it depends on how they do on the 90 and 100 level questions.

Here’s the breakdown. If students get 1 90 level question correct and the rest incorrect, they receive an 85. Again, students need to get both 90 levels correct to reach the 90 level range, but if they get one correct, they’ve shown that they are more than proficient. Therefore, I give them the middle score between 80 and 90.

Next, if students get both 90 level questions correct, but the 100 wrong, then they receive a 90. And of course, if all questions are correct, they receive a 100.

In addition, it’s not uncommon for students to get the 100 level question correct but one or both of the 90 level questions incorrect. In this case, I basically use the 100 level question as a 10 point bonus. So, if they miss both 90 level questions, they receive a 90. If they get 1 90 level correct along with the 100 level, they receive a 95.

What’s my reasoning for these scores? Let’s go back to our rating scale again. 90-99 represents advanced understanding, and our rationale for a student who is in the 90-99 range is that the learner has demonstrated outstanding scholarship and fluent understanding of the specific knowledge and skills.

Therefore, if a student can successfully reason through the 100 level question, they have shown at minimum advanced understanding. Then, it’s just a matter of finishing off those 90 level questions to advance higher up the rating scale.

Overall, I always give grades in increments of 5. I don’t add a point here and a point there for specific work shown.

Case 2. Students get all the 90 and 100 level questions wrong.

Here’s the scores I give. If students get all 3 80 level questions correct, they receive an 80. If students get 2 of the 3 80 level questions correct, they receive a 75, and if students get 1 of the 3 80 level questions correct, they receive a 70.

What’s my reasoning for these scores? Let’s go back to our rating scale again. Our rationale for a student who is in the 70-79 range is that the learner has demonstrated an “emerging” understanding of specific knowledge and skills.

Based on that, if a student gets any of the proficient level questions correct, I believe that they are at a minimum emerging in their understanding. Therefore, I give them a 70 if they get one correct. If they get 2 correct, it shows they’re really close to proficient, but not quite there, so I give them a 75.

Case 3: Students get all 90 and 100 levels correct, but one or more 80 levels wrong.

In this case, I usually just take 5 points off for each 80 level question that is missed. So 95 for 1 miss, 90 for 2 misses, and 85 for 3 misses. 2 to 3 missed 80 level questions are rare in this case, so you’ll mainly see a simple calculation error in one of the 80 level questions if students miss one.

Case 4: Students get all 6 questions wrong.

What do we do if students get no questions correct? Do they receive a zero? Not at all.

Let’s look again at our rating scale. For students in the 60-69 range, our rationale is the learner has provided evidence for assessment but does not yet demonstrate understanding of specific knowledge and skills.

Therefore, I really lean on the quality and quantity of work shown when determining if a student should be in the 60-69 range, or if they should drop down to the 50-59 level.

So, students receive a 65 if there is quality work shown, and it seems they were close to getting a question or two correct. Maybe their steps are almost there, but not quite yet. This student receives a 65.

For students that show effort, but they aren’t as close with their work shown, they receive a 60. Again, use your professional judgment here. It’s often clear when seeing quiz after quiz what type of work samples are 65’s vs. 60’s.

Finally, how do we know when a student should be in the 50-59 range? Let’s look again at our rating scale rationale. For the 50-59 range, the rationale is that the learner has not provided enough evidence to assess proficiency.

This sentence is the key. Students in this range simply have not provided enough work to assess. Either they didn’t try at all, or they made a minimal attempt. 50 or 55 just depends on whether minimal effort is apparent (that’d be a 55), or no effort is apparent (that’d be a 50).

Those are the most common cases I’ve encountered, and I found that it was pretty easy to determine levels of understanding based on the quiz set up. However, if you have interesting combinations of right and wrong questions, the overall key is to think back to the rating scale.

Take a step back and look at the overall body of work of the quiz you’re grading. After giving it a good look, which category of understanding does it seem to fit? You’ll have a gut feeling for which one it is, and my recommendation is to trust your intuition.

As we wrap up this video, I want to address two common questions.

First, do students need to write explanations for each problem on the quiz.

I’ve actually gone back and forth about this over the years. Some years I worked really hard to make sure students wrote explanations. Other years, I didn’t require them and only required full work to be shown.

And that’s actually where I land now. I told students that explanations can only help their cause because it allows me to see their thinking better, especially when they get questions wrong. So, it’s certainly not a bad thing to include them.

However, I do not require explanations because the great majority of the time, I can see student thinking if they show their work fully. Therefore, I require all work to be shown, but explanations are optional.

A second common question is about the low score for quizzes being a 50 instead of a zero. What’s up with that?

Although it seems strange to give a student who didn’t provide much work a 50, I’ve found that this is crucial to keep students engaged and to create an environment where students can bounce back from mistakes.

As I mentioned earlier in the workshop, at the beginning of my career, I had a lot of students give up early in the year after making low grades. Students would make a 20 or 30 on a test, and then they’d quit for the rest of the year. This was because they were in such a big hole, they realized they’d never be able to get out of it, and that’s actually solid reasoning.

However, now that students cannot make lower than a 50 in my class, every student is always able to bounce back. A 50 can be brought up. It’s not too big of a hole. And I’ve had many students make turnarounds because of this.

Overall, I found that it was extremely rare for students to quit on me after implementing this system. On the flip side, all the students that would have quit before instead stayed engaged, and many started to have success when they normally wouldn’t have.

Well there you have it. We’ve seen how to grade concept quizzes, and I’m a big fan of this process. I found that it cut grading time significantly, and the scores truly match student understanding. I hope you find the same when you give it a try yourself.

Next up in our workshop, we’re going to see what the actual gradebook looks like. How does it all come together? We’ll find out soon.

6 thoughts on “Grading Quizzes

    1. Hey Tony,

      Teacher discretion comes in when students don’t get all the 80 level questions right, or all the 90s, etc. Ideally, a student may get all the 80 levels right, for example, but miss the 90s and 100 level resulting in the student receiving an 80 as their grade. However, in many situations, the student will get some of the 80 levels right, or some other combination, and it’s up to the teacher to determine what level the student is at.

      It all comes back to the rating scale. The teacher is trying to determine which category the student falls into. So, based on the evidence a student has shown (work shown and written explanations), a teacher determines what rating a student is. If a student is close to getting all the 80s right, but has a small error or two, then he or she may receive a 70 or 75 (emerging category). If the student isn’t as close, then he or she will probably be a 60 or 65 rating (not yet category).

      When grading quizzes, it’s actually easier to determine grades than it may sound. Levels tend to form as a teacher grades one paper to the next. The teacher starts to see patterns in mistakes and gives consistent grades based on the patterns he or she notices. Really, the first class of the day is probably the most important, because that class is setting the standard for what ratings are given for the rest of the classes.

  1. I’m sorry if you’ve mentioned this already. I know that you’ve discussed grading discretion, but I wanted to get your input on this one: I have had two concept quizzes come back to me that are nearly perfect. The explanations are great. The work is done in the prescribed manner. But there is a small error in the proficient level problems. Something like dropping the negative at an inopportune moment. The explanation for the same problem is errorless.

    What would you do?

    1. Hey Michael thanks for asking! I would definitely give these students a full credit 10. This is another example of why I really like SBG. We can see that these students have great understanding of the concept. No need to punish based on small errors.

      Let me know if that answers your question!

      1. Yes, that was my instinct. It helps to hear it from a steady hand. I’m still unsure about when and where that makes sense, but it is inherently subjective. For instance, if the concept covered is “Logarithm Properties” and that includes equation solving skills, then what do you do if a student correctly applies all logarithm-related properties/steps but messes up an ordinary equation-soliving step (like dividing both sides where they should have added)? The concept involves logarithms, but it also involves solving equations. The “Algebra 1” skills are still important. What would you do?

      2. That’s a great question. My depth of content knowledge doesn’t go past Geometry at the moment, so I can’t give specific opinions about Logarithms. However, maybe some general principles can help.

        I’ve found in the past that sometimes I need to either get more specific with what concept I’m assessing or just decide to teach a concept but not formally assess it with a quiz.

        For example, in Algebra 1, I used to assess solving one-variable inequalities. This led to a similar situation to your Logarithm example. Whenever it came time to grade the quizzes, a lot of my grading ended up focusing on solving equations errors instead of things like graphing the answer on a number line or general inequality errors. It was hard to specifically assess the inequality aspect, and it felt like I was just re-assessing solving equations.

        After thinking about ways around it (and not really finding any), I decided to remove the concept from the list of concepts I formally quizzed. I still spent several days teaching the concept, but I didn’t quiz because the concept didn’t align well with the grading process.

        Another possibility is to break the broader concept of “Logarithm Properties” into smaller chunks. Again, I don’t have the content knowledge to help here, but I’m reminded of when I used to assess systems of equations. I used to assess the whole concept including writing out equations from word problems and the different methods of solving a system. Later on, I decided to break the concept into 2 parts, solving systems and writing systems. This allowed me to focus on 2 concepts that I felt were really important while assessing both at a deep level.

        Another example could be linear equations. If I just have a concept on the checklist called linear equations, then it’s hard to focus my assessment. Should I focus on the different forms of linear equations? Graphing? Writing equations? It was more helpful to break it into smaller concepts like graphing slope-intercept form, writing slope-intercept form, and calculating slope.

        It’s definitely a fine line at times figuring out how to get specific with concepts without getting too specific and having too many concepts to assess.

        Hope that helps!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.