Monday, November 25, 2013

The Stats Survey Write-Up

Each year in early November my statistics classes conduct a survey of our school's students.  We spend days brainstorming questions, winnowing down the list, rewording, etc, until we get 20 questions we can ask.  Then we go into classes all over school, randomly select students to survey, gather the data, and input it into a spreadsheet (it's just easiest that way).  Then there's a write-up.

I don't give a lot of written instruction on the write-up for a couple reasons.  The first reason is best explained by an analogy:
When you're taking your driving test, the instructor doesn't say, "Put on your blinker, check your blind spot, merge into the right lane, and then turn right up at the next corner."  He says, "Turn right up at the next corner," and expects you to know that you're supposed to do all those other things.
If I tell them what to do I'm not evaluating if they know that they're supposed to do it.  I am also skimpy on the instructions because the students are all, with the exception of one junior, college-bound seniors.  They should be able to synthesize and create, contrast and hypothesize.  If I tell them everything I want them to do I may as well give them a fill-in-the-blank sheet instead of assigning a report.

I do, though, give a lot of verbal instructions and reminders of what I expect.  If they take notes in class, or have exceptionally good memories, great.  If they don't, well, they're probably not going to do well.  The overarching instruction is to demonstrate what you've learned in class so far.  I do list some specifics I want them to address; how they address those topics, and the detail they choose to employ, are up to them.

The write-ups were due last Thursday; I'm not so mean as to require them on the Friday before a break, and I thought it cruel to have such an assignment hanging over students' heads all break by having it due after break.  The reports are written by teams of 4 or 5 so there shouldn't have been that much work for any one person to do.  After all the data was entered there was a week and a half to do the write-ups.

The survey asked 20 questions on a variety of topics, and each team had to choose a topic of interest and focus their write-up on that topic of interest.  Some teams chose specific questions, others chose groups of questions that all related a topic; for example, some groups chose to focus on the three questions that had slightly different wording on the two different versions of the survey, and to write about the differences in responses to those questions.

I usually don't bring work home with me but chose to bring these surveys home over break and to grade only 4 per day; that should get me through all of them by the end of break.  The grades on the first four?  15/30, 17/30, 27/30, 29/30.

Update:  I've graded the next four.  The scores were 17, 21, 24, and 27.

Update #2, 11/26/13:  Tonight's scores were 24, 28, 29, and 30.

Update #3, 11/27/13:  Today's results were 24, 26, 27, and 29.

Update #4, 11/30/13:  I graded the final four today:  18, 28, 28, and 28.


maxutils said...

While I agree with most of your post ... and maybe the rest, if you were more specific?
i think it's fair for a high school student to get a general, bullet point, set of standards ... in the case of statistics, I would say: a)introduction of basis of the survey b) basis of model c) results d) conclusions. I know that seems basic, and I may well have missed something you would have wanted ... but those first four scores weren't good.

maxutils said...

not a bad average at all...