Saturday, January 9, 2010

Clickers vs cards

I use peer instruction in most of my introductory courses.  When I first started I used colored cards for the students to use when answering multiple choice concept questions but when I had a chance to switch to clickers (personal response systems) I jumped at it.  I was the first person on campus to use them so I was sort of on my own with the hardware and software.  I lugged around the receivers and passed out the clickers on the first day of class, claiming that I'd charge the students $30 if they didn't turn them in.

I did that for around 4 or 5 years until one summer I realized how unexcited I was for a new class that was starting.  I tried to pin down my own emotions and realized that it was the lugging of the receivers and also having to always hook up my laptop at the beginning of the class.  So, just because I was lazy, I went back to colored cards.  I haven't switched back since and I thought I'd write a little about why.

First, why they compare so well for me.
I only ever used the clickers to collect and display answers for multiple choice questions.  I never tracked individual students (though I could have) and I never used the clickers to collect graded assignments like quizzes.  Since that's all I ever did, it's clear that colored cards don't have an immediate disadvantage.

Where you'd think clickers would win:
There are a few arguments in the pedagogy literature that talk about why student learning improves with clickers that wouldn't seem to work as well with cards.  The first is anonymity.  The students are unable to tell what their neighbors have voted on and this provides students cover to really think about what they're doing and not be intimidated by who they think the smart students are.  Scholars have also written about how students enjoy using clickers in class.  It keeps them engaged and it's fun to play with toys.  One more argument for clickers is that the students can see the histogram of the class vote projected.  This allows them to see that not all have said the same thing but that there are one or two answers where a lot of them have fallen.  This then tends to aid in the "convince your neighbors" portion of the peer instruction pedagogy.  There are more arguments for clickers but I'll just stick with those three for this post.

How do the cards stack up?
 First anonymity.  With cards students can watch what others do and be influenced.  I use cards that are only colored on one side so that at least the classmates behind you can't see what you're voting for.  In my experience I haven't seen a lot of influence happening in the first vote though I would say they use the cards to visually communicate with each other across the room during the "convince your neighbors" portion.  This is probably a win for clickers but not a huge win.

Next, the fun factor.
My students tend to have fun with the cards and appreciate the fact that they always work.  If they forget them they find interesting ways to vote like pointing at articles of clothing of the appropriate color.  Of course, sometimes I embarrass them in an effort to get them to bring them in the future by asking the students to stand and shout their vote instead.  The students have fun with cards primarily through my confidence level approach (see below).

Finally the histrogram: 
 When my student vote my eyes very quickly discern the color that is winning and I communicate that to them orally.  I feel, at least, that I'm able to provide them nearly as much detail about the distribution by simply describing it as they'd get by seeing it.  I admit I don't have much data to back that up, of course, but I will say I haven't heard complaints about it.

Where cards win:
It's very easy, of course, to administer the card approach.  I cut the cards, laminate them, and hand them out on the first day of class.  The cost is pretty low so if I don't get them back it's not a big deal.  There's no receiver to adjust, no software to play with, and no batteries to change or make available.  Laminating them makes them last for quite a while (3 years so far with no losses yet!).

Confidence level
The biggest unexpected benefit I've found with the cards is how the students can communicate their confidence level on a vote.  With clickers (at least the PRS ones I have) there are modifier buttons for the students to choose low, medium, or high confidence when voting and that is color coded in the histogram. With the cards I simply tell the students that their confidence level is the height they hold their cards above their head.  My students have a lot of fun with this.  Some have been known to stand on their desks and reach the ceiling with their cards while others will slouch and nearly drop their card to the ground.  I find this analog scale of confidence to be very useful to me as the instructor.  I often pay more attention to the confidence displayed than to the votes themselves.

Loss of anonymity
What's most interesting to me is that the confidence level is clearly not anonymous.  This creates a very interesting classroom dynamic as students can see that there is some confidence in the class in some cases and in others they can feel consoled by the lack of confidence anywhere.  I feel that this mix of anonymity (present, at least to some degree, with the vote itself while absent for the confidence level) is really useful in the classroom.  I especially like when there are more than one super confident students who don't agree with each other.  The class seems to get excited about the battle that shapes up with those "captains".

Final thoughts
In head to head comparisons looking at learning outcomes, cards and clickers are neck and neck. There are advocates on both sides (you can see where I stand) and I would encourage people to really think about which features they're looking for.  The confidence level aspect that I discuss here is something that I've only recently really put some thought into, especially the split anonymity, but I'd love to hear some differing opinions.