# Exam philosophy

I'm curious if anyone knows of any books, studies, or other resources on the philosophy of creating and grading mathematics exams. After working as a graduate TA for 4 years and dealing with a wide variety of issues creating and grading precalculus and calculus exams, I feel like I've developed a lot of personal opinions about how things "should" be done. I've even had quite a few "discussions" (borderline arguments) about exam fairness and so forth, so now I'm wondering if there is any well-repsected consensus opinions out there. I'm being deliberately non-specific, so perhaps this question might work best as a community wiki?

• I doubt you'll find much studies in this area, specifically for undergraduate education. Opinions, however, there are aplenty here... Apr 22 '14 at 18:23
• @icurays1 why not share some of your opinions? Apr 18 '16 at 17:12
• @DavidSteinberg I was trying to avoid giving opinions, instead looking for more rigorous studies. Also, it's been a while since I taught - I switched to an RA a while back so I haven't thought about these things in a while. Apr 18 '16 at 17:55

Since you remark that your question is "deliberately non-specific," here is a (necessarily) incomplete response: First are two links to documents about assessment that might be of interest, and then two grading schemes that I have encountered in mathematics courses.

Documents:

As far as the philosophy of creating examinations, early work on this was done by former AERA president Michael Scriven. In particular, he was the first to distinguish between formative assessment and summative assessment; this was done in "The Methodology of Evaluation" (1966) which can be found in full (as a pdf) here. The citation is:

Scriven, M. (1966). The methodology of evaluation (Vol. 110). Boulder, CO: Social Science Education Consortium.

For a more recent discussion of formative vs. summative assessment, see, for example, the following:

Taras, M. (2005). Assessment–summative and formative–some theoretical reflections. British Journal of Educational Studies, 53(4), 466-478. Jstor link.

Here is a relevant excerpt from the latter link:

Insofar as creating (mathematics) examinations is concerned, it will be important for you first to decide whether you intend the assessments strictly as summative instruments, or whether you wish them to serve a formative purpose as well.

1. Indicate to students roughly what will be covered; give three examinations; multiply the better two scores by 3, then take an average of the tests (i.e., if scores are $a \geq b \geq c$, then use $(3a+3b+c)/7$ as the student's average). Subsequently, use the course final as at most half of the student's grade; in particular, it can be worth more than half of the grade if the score is higher than the average going into the test.
2. Set the total number of points for a course at $N$ and allow for some fixed number of points that can be earned during the semester (through examinations, quizzes, etc). Make the final examination worth $N-($student's earned points$)$. The consequence here is that everyone enters the final with the possibility of getting full marks for the course, but also wanting to have accumulated points during the term.