Discussion about this post

User's avatar
Michael B. Horn's avatar

Enjoy what you're creating here as always! I'm intrigued with Nitkin's idea.

Here was my take on this, which I see his idea as a step toward ultimately:

https://michaelbhorn.substack.com/p/high-grades-are-presumably-the-goal?utm_source=publication-search

Daniel Paulson's avatar

First, we have to determine what a grade is. What does it mean? How do we determine grades? If a grade is a sorting of learning accomplishment, there are covariables that must be addressed. IQ is one. Cognitive development and the ability to deal with complexity are others. Absenteeism and socioeconomic pressures, such as the need to work after school, also could affect grades. The other factor is that we are in a time of teaching to objectives and goals. In that paradigm, we want everyone to master reading fluently, basic math facts, and the three branches of government.

If we wish to see grades in a normal statistical distribution, i.e. the curve, then we have to construct tests that sort. The best items on such a test will have a 50% pass/fail ratio. Some questions will be passed by everyone, and some will be failed by everyone.

What does that tell you about what a student knows or does not know? Maybe 50% of students cannot name the three branches of our government. Maybe the average student can read only 50% of the words in a list. What does the A-F grading system indicate in a system that wants everyone to accomplish the expected goals? To determine mastery of an objective, the teacher must identify a point on the distribution curve that indicates mastery. Is that the mean, or one standard deviation above the mean? In that system, we have to be comfortable with some students not being able to meet our expectations.

When you ask the grading system to determine mastery of objectives, you do not want a normal distribution. You want a distribution of scores that are bunched together at the high end of the scale. A-F grades are difficult when you want to signify mastery of an objective. Maybe the distribution curve could represent the number of objectives mastered, but you still do not want a normal distribution, with a few demonstrating mastery at high rates and most students mastering maybe 50% of the objectives.

Grade inflation is only a concern if you expect some students to master all the goals, most to master some of the expectations, and some to be wasting time in school.

1 more comment...

No posts

Ready for more?