Jeffrey Evans Stake
The faculty of Indiana University School of Law—Bloomington has adopted guidelines for grading standard courses. Those guidelines include a maximum standard deviation and a maximum and a minimum for the mean of the grades awarded JD students. There are many legitimate ways to award grades to students, including those based on established criteria and those that are purely comparative. For comparative grading, I have suggested some principles in my article, "Making the Grade: Some Principles of Comparative Grading," 52 J. Leg. Ed. 583 (2002).
To make it easier to adhere to those principles, I have created spreadsheets for grading that are not too difficult to use. Those spreadsheets, which may be downloaded below, may make it easier to follow the guidelines.
If you have any comments or questions, please send them to me at email@example.com.
This spreadsheet converts exam scores into grades. You can enter from 1 to 10 subscores, on any numerical scale, and the spreadsheet will calculate a total for each student using the weight you have given to each part. You can raise or lower grades by changing the target mean, and change the spread of the grades by changing the target standard deviation. The program uses the Z-score method of scaling grades, but that feature can be toggled off. The spreadsheet is set up for the Indiana University Maurer School of Law grading scale, but it can be modified to accommodate other grading scales with up to 10 grading intervals. The spreadsheet will accommodate a couple hundred students. If the mean and standard deviation should be calculated without some of the students, they can be excluded from those statistics even though the spreadsheet still calculates grades for them.
This spreadsheet calculates the average grade and the standard deviation of the grades. You enter the number of students receiving each grade and the program tells you the mean and standard deviation for that set.
The following table provides one example of the kind of distribution that can result from processing student scores into grades using EZ-Score. This particular distribution was created from 139 actual student scores on four parts of an examination. Applying a target mean of 3.246 and a target standard deviation of 0.356 to this particular set of scores resulted in an actual mean of 3.250 and an actual standard deviation of 0.370.
|Mean = 3.250|
|Standard deviation = 0.370|
You may view the sample spreadsheet (in Corel Quattro Pro or in Microsoft Excel) that created this distribution of grades. But do not try to modify this spreadsheet for grading a class of students. Instead use one of the downloadable templates available in the first two links on this page.