I’m *still* grading that 10-point calculus homework assignment. One exercise gave the students a table of data and asked them to (a) create a scatter plot, (b) fit it with a linear trendline, and (c) use the trendline to predict a data value not on the table. Students had the option of doing this exercise using Excel and sending it to me as an email attachment.

What could possibly go wrong with that, you ask?

– A student saved the file with the extension .xlr, as opposed to the usual Excel file extension .xls. I don’t have any tool on my computer that will let me look inside this file to see if there’s legitimate work in it. So I have to have the student resubmit the spreadsheet instead, and I have no way of knowing if the original work was just BS, other than the fact that the .xlr file is 12K large. [*Student hack of the day*: If you have to turn in work in the form of a computer file, and you are pressed for time, "accidentally" change the file extension by one letter. By the time the prof gets to it and contacts you about the misspelled file extension, you’ll have had time to finish the work, and send it to the prof in the right form*.]

– A student sent one email with a spreadsheet that had the scatterplot but no trendline and no predictions. On his main writeup, the student said the rest of the work was in a second email. Which I never got. Which raises the old issue: If a student claims s/he sent you an email with graded work in it, and you didn’t get it, what do you do? Did the email fail to arrive because of a mail server glitch, or because the address was wrong and it didn’t bounce, or because it never really got sent in the first place, etc.? Who bears the burden of proof here?

– One student literally cut and pasted the scatterplot + trendline into the main writeup (which is a little messy but OK for now), and in part (c) just gave the answer for the predicted data value… with no clear justification. That earns the student a 1 out of 10 on the assignment. Another student typed in the value of the predicted data point on top of the trend line about where the data point should appear, which to me constitutes a sort of graphical justification of the answer (i.e. student explicitly, if wordlessly, connected the answer with the graph). I’ll have a lot of fun explaining to the 1/10 student that the difference between his/her graph and the other student’s was the physical location of his/her answer on the page.

And that’s just from one section I’m teaching. The other is coming right up.

* I am not suggesting this student did this.

Tags: Calculus, grading, Excel, teaching with technology

Pingback: Casting Out Nines»Blog Archive » A case study in grading

Why couldn’t you just rename the .xlr file back to .xls and open it in Excel?

You can avoid these problems by providing students with the Excel file, along with textboxes for them to enter answers, and worksheets for the scatter plot and trendline. Linear regression in calculus?

RWP – I actually really want students to get good at creating a spreadsheet themselves with the intent of accomplishing a particular task. So I want to leave the data entry and table formatting up to them.

We’re using Excel in calculus to (a) generate tables of data from formulas, and (b) take tables of data and fit them with different kinds of trendlines, and then use the formulas for the trendlines for taking derivatives and so forth. It’s our way of turning tabular data into formulas which we then do calculus on. We’re not getting into the nitty-gritty of regression analysis.

We not only give out the templates in files, we have VBA code in the project that grades every assignment. There is not even one point out of the total thousand that is up to the instructor. This is also why students cannot use Macs — because if they do, they’ll get a zero.

Creating templates, setting up problems, we do all that in class, or I should say, they do it in class. But for exams and assignments, everything is VBA graded so 24 hours later, all the grading is done, turned in, and available for them to see online.