17 October 2005

More on grading/economics

Effort grades are inefficient. Instead of seeing how much time you can get your students to put into a homework assignment, state explicitly what is expected for each grade and then let them decide the level of effort they desire to exert to achieve a given grade. Although it may seems to a professor that Stochastic Decision Modeling is the most important subject ever, remember that the student is taking many courses as well as being involved in dozens of other activities.

As an example, let us say we have two student operating under two grading systems, one based on inputs and the other on outputs. Grading based on inputs cancels out some benefits for efficient students. Certain math classes where many similar, repetitive problems are required for homework is an example. The student that is good at math can probably learn the concepts necessary long before the assignment is completed; the rest of the time spent is wasted. On the other hand the weaker student may need extra practice to master the concepts; the effort required by the assignment is no more than an estimate of the effort required to master the concepts - it may not suit any one individual student.

On the other hand, grading based on output allows the more efficient student to achieve the same level of performance while saving time for other activities. In essence, she is rewarded in both a desirable grade and free time. At the same time, in terms of efficiently achieving performance, the weaker student is no worse off; he must still potentially complete the assignment and do more. In addition to all this is the fact that many students learn in different ways, and usually each individual knows which ways are best; again, the assignment is only an estimate of the best way to master the material - it may not suit any one individual student. So output grading is a pareto improvement over the input grading system.

No comments: