Consider these two students.
Student 1 and Student 2 both take test 3 – a 55 question Multiple Choice test based on US History Standard 3 – The Gilded Age to the Progressive Era.
Student 1 scores an 80.5%.
Student 2 scores an 82.3%.
This is a grand total of about 1 question difference but the averaged score appears to indicate that Student 2 performed better on the test. If I were to put this in the gradebook as is the number would be calculated into the final average but it doesn’t really tell us much more than that.
Now consider that the test was a district mandated “Common Assessment” that generates a report that will provide data breaking down the students’ success according to the substandard indicators. Each US History Standard is divided into 6 indicators based on content and expected skill. Each question on the common assessment is linked to one or more of these 6 indicators. Now imagine if, instead of just giving back a raw score, the student were to get back a report showing how they did in each of the indicators! Much more useful! I documented how I use this to help grow mastery in this previous post.
Now Reconsider those two students with these data reports.
Student 1 did better with 3.CE and 3.P whereas Student 2 did better with 3.CO. and 3.CX. Similar raw scores but different strengths and weaknesses. I would like these differences to show up in the gradebook as a record for the student, myself, admins, and parents.
I also know that testing gives the illusion of accuracy. I want the gradebook to consider that both students probably missed some questions they knew (due to vocabulary, exhaustion, misreading) and answered correctly some questions they didn’t (good guessing, test taking strategies).
Using Powered Up XP (want to know more about the Powered Up XP grading?)
In my overall grade scheme students are attempting to get to 10,000XP. Since the district requires these common assessments to count as a “major” grade (don’t get me started) and that “major” grades must count as 50% of the final average (again… blech….) I start my semester knowing that roughly 5,000XP of the 10,000XP must come from the Common Assessments and projects. In Semester 2, I have 3 required common assessments and each has 6 tested indicators for a total of 18 indicators. If each indicator is worth 300XP then I get to 5,400XP which is roughly my goal. Yay!
Did you notice that the district doesn’t evenly distribute the number of questions in each indicator? Also, if you were to add the number of questions in each indicator the total would be greater than 55 because some of the questions are linked to multiple indicators and some of the “choose all that apply” questions are mistakenly worth more than the other questions. I don’t really care because I am going to grade these on tiers anyways.
XP Grading Tiers
|TIER LEVEL||Indicator Raw Score||Amount of XP|
So, now let’s consider Student 1 again.
On the tier system here is the XP Grading:
(last part of
|Raw Score |
For this test Student 1 has earned 1,550 out of 1800XP = about 86% of the possible XP.
Now look at student 2:
(last part of
|Raw Score |
Now this student gets 1500 out of 1800XP which is 83% of the possible XP. (yeah, I was a little generous in 3.CE)
The tiered system provides better data for the students than the raw score. It also allows me to account for the inaccuracy of the assessment and the odd unevenness of the question distribution on the test. I think of this as 6 assessments in one rather than 1 big assessment. It can then be used with the credit recovery system to promote growth towards mastery. Also, In almost every case that I have measured the percentage of XP earned is within a few percentage points and almost always favors the student.
These are the raw and XP scores in one class (half a class due to Hybrid schedule):
|Raw Score||XP Earned|
Finally, you may have noticed the built in 50% floor for students taking the test. This is intentional as I want all students to know they have learned something over the unit and that a single poor test will not destroy their chances at success.