Indicators Met
The indicators met metric measures performance by grade level and subject. Long division. Adjectives and adverbs. Kingdom, phylum, class, order, family, genus, species. It's all here, my friends. I like to call this metric No Subject Left Behind. The motivation is to ensure students are successful in each subject, since the performance index only examines overall test scores.
The Ohio Department of Education metric sets a goal for the percentage of students that must pass a subject. If you meet the goal, you get a point. If you're one student short, too bad. It's like filing your taxes on time. Miss that April 15 deadline by one second and maximum penalties for you! Letter grades are then computed from the point totals.
​
If we take this approach, first we need to set the goal, say something like 80%. Looking at the ODE calculations, the goals are
different for each subject and grade level, whaaaat? I can only surmise that this was done in response to complaints that the scores were too low. Heads up, readers! That's a sign of a problem, and we need to fix it. We're looking for a way to show subject deficiencies while maintaining an accurate overall performance portrait. Let's start with a...
Thought Exercise
In fact, the "met/not met" classification is a bit harsh, because we're losing too much information.
​
Imagine you're grading the following tests. Each test covers 2 subjects with 5 questions per subject. The pictures show which questions were answered correctly. Intuitively, what letter grade would you give each?
Got them graded? Good. Now, let's look at what an indicators met score would be.
​
There are two subjects. Let's say the pass rate for each subject is 80%. For these tests, that's 4 questions correct out of 5. Since there are only two subjects, we'll have three possible letter grades: A for passing both, C for passing one, and F for passing neither. Our indicators met grades are: A, C, C and F.
1. Show the subject score distribution with a uniform pass goal for all subjects.
2. Replace "met/not met" with a point bonus.
3. Consider renaming this metric to Subject Achievement.
Test 3 has four questions correct and receives a C. Test 4 has six questions correct but receives an F. Huh?
​
Hmm. We like having information on subject performance, but the "Met/Not Met" dichotomy doesn't gel with our expectations of rewarding achievement. What's a better way to illustrate subject performance? How about our friend, the distribution?
Pass Rate Distribution
Alright. We'd like to see the pass rate in each subject, for each grade, with emphasis on places where the goal was not met. How about this?
Recommendations
Here's a bar chart for a sample district with a mean pass rate equal to 80 (the goal). We keep a uniform goal for all subjects and grade levels. Missed goals are indicated by dark bars. They pop out, don't they?
​
We can use the distribution to identify patterns that lead to improvement suggestions. For example, this next district has low math scores across the board. It should focus on improving its math curriculum.
Whoa! Not a single math goal met. A "met/not met" classifier would give a big fat 0. Fortunately, we can see from the distribution that the pass rate was over 70% in most cases. With a little work, this district will be meeting goals soon.
​
This next district has a problem in grades 4, 5, and 6. It should investigate the root causes and focus on improvements in those grade levels.
​
​
Overall, the distribution gives us much more insight than a single indicators met letter grade. Hooray!
Bonus Met!
We've seen that "met/not met" doesn't adequately reflect the underlying subject achievement. Is there a way we can reward districts that meet the goal while giving partial credit to others?
​
Let's try a bonus. We'll convert out left-hand axis to points. First, we award points for the actual pass rate (say, up to 60 points). Then, we award bonus points for goals met (say, another 40 points). We'll get a distribution that looks like this. We don't necessarily have to show this distribution on the report card. It's so you get a feel for how the bonus affects the point total.
Then, we add up all the points for each subject and convert to a letter grade for that subject. We can calculate an overall letter grade too, if we like.
​
The bonus strategy is flexible - the bonus amount can be carefully calibrated to reward high achieving districts while not discouraging struggling districts.
The important takeaway is to show the pass rate distribution. Then, districts will know where to improve.
​
Congratulations! We've completed this section. Here are the recommendations.
