Long story short:

How to modify an old test and score it with an SBG mindest:

  • Chunk test by standards/objectives instead of type
  • Each chunk is worth 4 points and graded for proficiency
  • Find the average rubric score (total points/# of objectives)
  • Use Rubric → %  scale to determine score for gradebook

    Teacher takeaways:

  • Less time to grade
  • Score means more to me (true proficiency vs. point mongering)
  • Students lacking proficiency don’t get confidence completely deflated (maintains hope)
    • Kids still failed, but less did

    Busted Myths (Schimmer, 2016):

  • Myth 1: SBG makes it easier for the students
  • Myth 2: SBG is more work for the teacher

Full Story:

I took my first stab a standards based grading (SBG) today. My super excited side wants to just dive in, rewrite all my curriculum/assessments, and start over….but reality check says very few people will want to join me on that path and it’s the END of 3rd quarter, so implementing a new way of grading and looking at assessment would be hard on everyone involved.

So I wanted to look at ways to start small. These changes would be too small for students/parents to get scared by, and would be a manageable way for any teacher during any semester to ease into. I decided to look at a summative assessment I already had written and had actually already given to my students and start there. I didn’t change anything about the test only how I scored it.

Traditionally I would have scored the test by counting how many steps or parts to the problem there was. This process of just preparing a test for scoring was cumbersome…and not fool proof. It would take time to decide points for each question and then would spend a lot of time while grading going back and forth making sure I was being consistent and fair…so grading took me hours to do (I HATE GRADING).

When looking at the test this time I divided it by standards/objectives, plan to give one grade for each objective. This meant that sometimes I would look at a  whole problem, sometimes a couple problems, and sometimes I divided one problem into multiple groups. Overall I found there were 17 objectives the students were being tested on. I gave each of the 17 objective groups a score of 4 regardless of how many steps completing that task require. This gives even weight to each objective instead of weighting problems by effort/time to complete.

It took me about ten minutes to decide scores and grading the test only took a fraction of the time it normally does!

The ease of grading came from having a consistent way of scoring every problem. There were several options to consider when it came to scoring, but I decided to use a basic 5 point rubric for scoring each problem (Schimmer, 2016):

4- Advanced: Fully met standards even in new scenarios, GOT IT

3- Proficient: Fully meeting expectations, GOT IT with MINOR MISTAKES

2- Developing: Mostly met expectations, ON TRACK FORGOT KEY POINTS

1- Novice: Not meeting expectation, COULD IDENTIFY SOME RELEVANT MATERIAL

0- No Evidence: Insufficient evidence to determine understanding, BLANK OR IRRELEVANT.

To communicate grades to students I needed to consider how the 4-0 scale translated to percentages. The 4-0 should correlate to the A-F scale, but if you put it in your grade book as 4, 3, 2, 1, 0 you would get 100%, 75%, 50%, 25%, and 0%. Which doesn’t really match the rubric above and overly emphasizes the lower half of the % scale. Schimmer (2016) suggests a modified scale: 4-95%, 3-85%, 2-75%, 1-65%, 0-50%. The other thing to consider is how many grades. I could enter 17 individual grades, one for each objective or one score for the whole unit. When I shift completely to SBG I think that individual grades would be great, but that might freak my kids and parents out in the 3rd quarter…so I opted for one grade. Again, I can’t just tally the scores and give a % that would again ruin the scaling and wouldn’t tell me where someone sits overall in the unit. Instead I totaled the score and divided by 17 (the number of objectives) this gave me a score between 4 and 0…I then assigned a percentage with a more detailed scale than the one above, adapted from Schimmer (2016).

Scoring in this way showed me who gets it and who still needs help and how much. It also didn’t dig my struggling students into a points hole that they cannot get out of or that will deflate their confidence.

Schimmer, T. (2016). Grading from the inside out: bringing accuracy to student assessment through a standards-based mindset. Bloomington, IN: Solution Tree Press.

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 )

Google+ photo

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

Connecting to %s