Shooting yourself in the foot for finals

As students brace themselves for final examinations, I would like to leak to the press the professoriate's most closely held trade secret: Multiple choice questions aside, grading papers and examinations can be subject to human error and, worse still, to personal bias.

As the further away an exam moves from answers expressed in mathematical symbols toward written prose, the room for error and bias increases. But even for examinations calling for answers rendered in concise mathematical symbols, errors and bias can creep in when graders grant partial credit for imperfect work that strikes them as on the right path.

Most universities take it on faith that professors and teaching assistants will do everything in their power not to let gender, race, politics or personal feelings about a student for any other extraneous reason enter their evaluations of that student's work. Absent that faith, a workable alternative would be to identify students on their examinations by a numerical code whose key is unknown to the graders. I have on occasion used that approach, and it worked well.

But there is one personal bias that even a numerical ID could not eliminate, namely, the ill will among graders triggered by poor workmanship on the examination itself, such as poorly labeled and carelessly rendered graphical displays and illegibly written mathematical expressions or prose. Such workmanship can trigger a grader's antipathy for at least two reasons.

First, graders at any good university are busy people, especially teaching assistants who take difficult graduate classes or are under severe pressure to finish their Ph.D.s and prepare for the job market. Their time has a high opportunity cost. Sloppily written or illegible exams exact an added time-price from graders, which may annoy them. Even if, in the end, graders have fully deciphered what students submitting poor workmanship sought to communicate, the by now annoyed graders would be likely to assign scores from the lower end of the range of defensible scores, even if only at the subconscious level. Thus, the answer to a question scored out of 20 points might receive a still defensible 16 points, rather than a more generous 18.

Second, busy graders are apt to interpret sloppily rendered homework or final examinations as signs of disrespect, which can further fuel the grader's antipathy. Poor workmanship suggests that students view graders as lowly servants whose duty it is to sort through whatever mess the student has left behind on paper. It is not a winning strategy when graders have discretion over one's grades.

What, then, can students do at this stage to bias graders in their favor, other things being equal (such as good will built up through good participation in precepts or seminars)? Let me make three suggestions.

First, students should think of themselves as high performance cars. No one in their right mind would put low-octane gasoline into a high-performance car. Why, then, do so many Princeton students stay up for last-minute cramming all night, only to perform sluggishly the next day on their exams, like high performance cars running on low-octane gas? After an all-nighter, possibly sustained by sundry pills, one's recall will be diminished along with one's reasoning power. In the ensuing frustration and possible desperation, one is then more likely to lapse into sloppy work or poorly rendered prose, which only compounds the problem. In the end, after receiving a poor grade, students who made this mistake may visit their professors to argue, correctly, that their final exam does not reflect what they actually have learned in the course. But what can professors do about it ex post without violating horizontal equity?

Second, even if students had been permitted throughout their high-school careers to submit barely legible handwriting and sloppy workmanship all around — and many high schools do seem to indulge students thus — it is not too late in college to train oneself in better workmanship on homework assignments and written examinations. To encourage students to do so, I explicitly reward good workmanship on homework assignments in my courses.

Finally, when answering a question, students should first tell graders briefly what they are trying to do, before presenting the details of their responses. Even on questions calling for essay-type answers, students should summarize up front the central thrust of their responses, as do newspaper columnists and writers of oped pieces. That courtesy alone might lift the score in the previous example from 16 to 18.

As is well known, the late Chairman Mao Zedong had a famous dictum on all facets of the human existence. In December of 1936, he penned this jewel: "What we need is an enthusiastic but calm state of mind and intense but orderly work." It is the perfect dictum for the season. Uwe E. Reinhardt is the James Madison Professor of Political Economy and a professor in the Wilson School. He can be reached at reinhard@princeton.edu.