2007 January 17 / E-Mail Me

Reading Assignments

During the 2006-2007 school year at Duke University I incorporated formal reading assignments into my Laboratory Calculus I class. My colleague Jack Bookman ran parallel reading assignments in his sections. The idea for these assignments is not mine; I got it from Thomas Ratliff of Wheaton College, who spoke at a meeting of MAA's Project NExT. Thanks to that program, and to the Duke Math Department for sponsoring my participation.

The goal of such assignments is to get the students reading the book before class, so that they come into class with some idea of what's going to happen. This page gives concrete suggestions, some preliminary results, and even some software, to help you design and implement these assignments for your class.

How They Work

To make it concrete, suppose that Friday's class deals with Sections 3.3 and 3.4 of the book. Ahead of time, I read through those sections and come up with a couple of questions. I post these electronically; details on that are given below. The students respond electronically, by an established deadline before class. (My 2006-2007 classes met at 8:45 AM, so I set a deadline of 10:00 PM the night before.)

After the deadline, I read each student's responses. I grade the assignment on a Boolean scale: 0 or 1. Everyone who submits the assignment gets 1 point for it, except in rare cases when the response is trivial, such as "I didn't do it". (At the end of the semester, points for reading assignments constitute 5% of the grade, say.)

In class on Friday I work from the assumption that all students are familiar with the material. I begin by going over the reading questions, eliciting a variety of opinions from students, perhaps provoking a discussion, and certainly clearing up misconceptions. Then I launch into examples, problems, group work, etc.

Notice what I do not do in class: I do not spend a lot of time writing definitions and theorems, because the students have already seen them. They can reread them in the book, if they like. The book is a resource, not just a list of problems.

Designing Questions

Reading comprehension questions are not supposed to be as demanding as homework problems. They are just supposed to encourage the students to read. Designing them requires some care, and I don't consider myself proficient at it, yet. For example,

What does the second derivative tell you about the concavity of the graph of a function?

In Figure 3.5 in Section 3.3, for which values of x is the second derivative of the function positive?

Implementation Via Blackboard

Duke University provides commercial web-based software called Blackboard to its faculty, so that they can communicate with their students over the web. I began the 2006-2007 year running my reading questions through Blackboard, as follows.

1. In Blackboard I create an Assignment. If my reading questions demand pictures or complicated math, then I can attach a PDF file with that stuff; usually, however, plain text is adequate. I set a due date (including time of day) for the assignment, after which it will disappear from Blackboard.
2. Blackboard automatically creates an entry for that Assignment in the Gradebook.
3. Students log onto Blackboard, read the Assignment, and fill out a web form with their answers. Blackboard marks the Gradebook to indicate which students have submitted answers.
4. After the deadline, I log onto Blackboard and open the Gradebook. For each student I read her response and enter a score of 0 or 1.

Blackboard works, but slowly and awkwardly. As early as a few weeks into the term, the Gradebook is positively crammed with these little 1-point assignments. To read a single reading question response, I need to scroll down to find the student, scroll right to find the assignment, carefully click on a tiny "!" indicating that no grade has been assigned, wait for a web page to load, read the response, enter a 1 in the form, click Submit, wait for a web page to load confirming the submission, click OK, and wait for the Gradebook to load again. If I actually want to give feedback to the student, then that's another few steps.

Another problem is that Blackboard inexplicably strips carriage returns out of students' responses. In math this is a disaster. A student will type a sequence of equations, on separate lines, that elegantly lead to the desired answer or conclusion, but when I read them they are all crammed together, one after another, on the same line, resulting in a colossal mess of a paragraph.

Another problem is that assignments disappear after the due date has passed, so that students can no longer read them. To make them available after the due date, but not answerable for credit, I have to repost them elsewhere on Blackboard.

Implementation Via E-Mail

Running reading assignments through Blackboard was unpleasant enough that I switched to running them through ordinary e-mail.

1. I send an e-mail message to all of my students, with an easily recognizable Subject line such as "31LReading 3.3,3.4" As with Blackboard, I can attach a PDF if I like, but usually I stick with plain text.
2. The students Reply, typing their responses in the reply message. I require that they not mess around with the Subject line; it should be something like "Re: 31LReading 3.3,3.4".
3. After the deadline, I check my e-mail. I can easily pick out the reading questions by their Subject lines. I read each one. If I like, I give the student some feedback by sending a Reply message. I save the student's message into a designated folder in my e-mail program.
4. Whenever I want grades, I run a little program (detailed instructions included) that I wrote for this purpose. The program inspects the designated folder, and for each student records which assignments she submitted on time and tells me how many there were. I enter that number into my gradebook.

Please note: Using a program to grade the responses does not mean that I don't read the responses. I just don't mark them with a 1 anymore. If anything, I'm actually more involved with the responses when using e-mail, because I sometimes send the students personalized feedback, which I never did when using Blackboard.

Students generally preferred e-mail to Blackboard, as explained below.

Survey Results

I handed out informal surveys early in the term (the fourth week of the 15-week semester, with 39 responses) and late in the term (the final week, with 37 responses). The switch from Blackboard to e-mail occurred between these two surveys. Here are some modest results.

Late in the semester, the students reported spending an average of 17 minutes per assignment, with significant variation — a standard deviation of 12. (Early in the semester, they reported 29.6 minutes on average, but the question was asked differently and I suspect that comparisons are invalid.)

When asked to rate the value of these assignments in comparison to other class activities, the students said that reading assignments were less valuable than lecture/discussion and homework, comparable in value to labs, and more valuable than group work in class.

Students were asked whether they would read the book more, the same amount, less, or not at all, if there were no reading assignments. The early and late surveys both showed that reading assignments got the students reading:

Amount	Early Survey	Late Survey
More	0	1
Same	7	13
Less	23	19
None	9	5

Among the students who gave written reponses, many said that reading assignments should be kept, since they force the students to keep up with the material. A smaller number of students said that they still didn't read much, even with the assignments. I attribute this to the fact that most of the students had studied calculus before, and that the reading questions need to be improved, so that they actually encourage reading.

Finally, when asked "Which is more convenient for reading assignments, Blackboard or e-mail?" in the late survey, 22 students favored e-mail, 8 favored Blackboard, and 7 expressed no preference. A typical comment among e-mail supporters was "E-mail is harder to forget." The e-mail supporters also used many exclamation points.

Further results will be posted at the end of the school year.

Conclusion

I personally like the reading assignments. They don't demand too much of my time, and the time spent is almost pure feedback — reading the students' descriptions of math, in words. I like the fact that students are technologically limited to plain text responses; they are thereby required to move beyond opaque, uncontemplated symbol manipulation.

The assignments seem to improve students' preparation for class. They certainly change how I conduct class. I spend less time on definitions and theorems, and more time on examples and deeper problems. I intend to continue them.