Going into Math 320, Elementary Probability and Statistics, I expected to be taught and tested on probability and statistics by Professor Edward Spitznagel. For the first test of the class – as well as for my first test of the semester – I studied for the mathematical material presented in the textbook and in the lectures: correlation, r2, least-squares regression lines, categorical variables, ordinal variables and a number of other terms that most readers right now couldn’t care less about. On the exam, there were, sure enough, questions on this mathematical material, and coming into the lecture hall on Monday evening to take the test there were, sure enough, questions on correlation, r2, least-squares regression lines and the calculations and interpretations thereof.

There were, in addition, the following questions in the exam:

“In one of the exercises on displaying numeric data, a histogram of singers’ heights showed what feature?”

This was the entirety of the fifth question of 26 10-answer multiple choice questions in the test. I was not being asked to interpret, calculate, or analyze data from the exam. I was not being asked to apply a concept from the textbook to another problem. I was being asked to recall the statistical particulars of a single exercise from the textbook.

“In one of the exercises involving the plotting of numeric data, a histogram had a gap in the middle corresponding to a very noticeable drop in the time series plot. What numeric variable was being graphed in both cases?”

This was also the entirety of the question. Hopefully, I thought, it would be the last sort of question on the exam. I had studied for the test by understanding the concepts of the textbook exercises. I had not studied for the test by memorizing the answers to the textbook exercises.

“In the Western Hemisphere study of female life expectancy and number of children a woman has, one country yielded an extremely influential point in the scatter plot. Which country was it, and why was its point so influential?”

This was, of course, also the entirety of the question, taken from a textbook exercise. I considered myself lucky to have remembered that the country’s name was Costa Rica, although I had not taken notes about Costa Rica when the professor discussed the exercise in class.

“Which federal agency had the highest rate for both assaults per 1,000 and killed-or-injured per 1,000?”

The ten answers to this question had nothing to with statistical concepts. The correct answer was “G) National Park Service”. I guessed “B) DEA (Drug Enforcement Agency),” and later received four points off my exam.

“John Tukey, inventor of the box plot, the stem-and-leaf plot, and scads of other simple and effective ways to look at data, also had the knack of coining words and phrases that have become part of standard English. Which one of the following words did Tukey invent?”

I considered myself very lucky to have remembered that Tukey had something to do with the word “software,” especially when in class I had not taken notes or studied about Turkey and his influence on the English language.

“Which of the following people was Francis Galton’s cousin?”

I had not studied for anything regarding Francis Galton’s family tree, although I did remember that he had something to do with the invention of the regression line. I had spent one or two hours studying how to compute and calculate least-squares regression lines.

In an e-mail, Professor Spitznagel wrote this:

“I gather that some of you are in a state of shock about being asked some content-and-context questions, rather than all plug-and-chug. I do all sorts of statistical studies, and the easy part is the computation. The much more challenging part is explaining exactly what the results mean, to my clients or my colleagues.”

Explaining data and results was, apparently, part of the first two questions above, and those two questions also involved recollection of textbook exercises – not mere analysis or explanation, but actual memory of the contextual particulars of the exercises. To answer the other three problems correctly, I had to remember certain facts that somehow seemed more appropriate to a game of “Jeopardy!” than to a university exam.

Does college actually test us on what we learn? Does a precious 4.0 in a class represent what we have learned, or does it just measure our capacity to stay awake at three in the morning cramming facts for the upcoming chemistry exam, or (even more inaccurately) our luck that we had spent more time on electron orbitals than on wave-particle duality? As an English major who has (or at least will have) more essays than exams to deal with, I’d like to say it doesn’t apply to me, but it sometimes feels like spitting facts and sentences back onto a paper is all that’s needed for a decent grade. Of course Washington University will remain a research university, and there’s nothing wrong with that as long as undergraduates are given fair consideration.

Professor Spitznagel also wrote in his email:

“Don’t worry; it won’t be on the next test.”

Hopefully, “it” refers to the kind of questions I have described.

David is a sophomore in Arts & Sciences. He can be reached via email at dssong@artsci.wustl.edu.

 Print This Post

This entry was posted on Wednesday, October 18th, 2006 at 12:00 pm and is filed under Forum. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.