PollEverywhere: Summing up

[This post wraps up my reflections on my pilot of PollEverywhere this past spring. If you missed my last two posts, I discussed how I used PollEverywhere in my data analysis course, and student reaction to it.]

So here are my general thoughts on PollEverywhere, particularly relative to standard clickers:

Advantages

• Convenience for students. They all have cell phones so they never 'forget' their device the way they do with clickers.
• Easy to ask open-ended questions. Even on clicker systems that have this feature, it is generally easier with cell phones/laptops.
• Relatively low-cost. If you have fewer than 30 students, the service would be totally free; if you need to track more responses, there is a cost for a PE account that someone (you, your institution, or the students) will have to bear. For students who do not have unlimited texting, there may be costs related to sending/receiving messages; the total cost will depend on how many questions you ask (in a previous post, I calculated that for my class, assuming 10 cents per text, it would still be cheaper for students than buying a clicker).
• Relatively low-commitment. Even if clickers are used a lot on your campus (as is the case at SDSU), PE can be a great complement for faculty who only want to use the technology occasionally. For example, in my spring writing course, there was one class meeting where I wanted to survey the class about some plagiarism issues, but since it was only that one day, it really didn't seem worth having them buy and register clickers. In the past, I probably would have either skipped the surveying or I would have asked students to answer the questions before class; instead, I used PE (note that in this situation, I did not need to track responses so students did not need to have accounts or register their phones).
• Few problems with software. This is a relative thing - many faculty on my campus have had major issues with the clicker software we've been using. In comparison, PE was really easy and almost always worked well. Not perfect but glitches were pretty rare.
  

Disadvantages [with the caveat that some of these things seem like a pain to me just because I'm used to clickers; if you've never used clickers, you may not find most of these issues particularly problematic. Also, according to a comment on my first post, the PollEverywhere folks are working on several of these concerns!]

• Lack of integration with University systems. Students have to register on a separate site and getting grades into the course management system requires more work than most clicker systems.
• Multiple submissions from same person recorded as separate responses. If you allow multiple responses (which allows students to change their answer after submitting something), you'll have to deal with sorting those out afterwards.
• Integration with PowerPoint is kind of clunky. I do everything in PowerPoint and prefer not to switch to a separate site to ask questions so I embedded all my PE questions into PowerPoint. To do this, you create all the questions on the PE site and then export a PowerPoint file. The questions themselves are shockwave flash objects embedded in the slides that I would then cut and paste into my class PowerPoint presentation (where I also created a timer, which I'm used to having for clicker questions). It's not difficult; it's just more time-consuming and not nearly as seamless as with clicker systems.
  
• Difficult to ask questions on the fly. Many clicker systems have a way to get responses to questions that have not been prepared ahead of time (e.g., asking a question verbally or creating a new question on the fly). It isn't hard to create questions in PollEverywhere but in order to generate the codes for responses, you do have to go into the system and create the question with answer responses. I think trying to ask something that you didn't prepare ahead of time would be too time-consuming to do spontaneously in class. It occurs to me as I write this that one possible workaround (if you anticipate needing this option) is to create a generic question so the codes are already created...
• Need consistent cell service. Most of my students do not have laptops so if cell service doesn't work for someone, there is no alternative way for them to submit responses to the system. I really am not sure what I would do if I had more than two or three of these students. In my opinion, this is the biggest reason not to have PollEverywhere completely replace clickers (e.g., if a campus is trying to decide on one standard, as SDSU does).

Other issues [Not necessarily good or bad but stuff you should think about if you're considering using PE]

• Students have cell phones out during class. Duh. This is not a problem for me but I know it is a huge issue for other people. Aside from issues related to students cheating (which I didn't worry about because all my questions were low-stakes) and students texting their friends (which I don't worry about in general), the one thing I was a bit concerned about was whether students would remember to turn off their ringers. I put a reminder on the first slide to silence phones (there's an image in my previous post) and I have to say, I don't think I had any more ringers go off than in a typical semester; if anything, I think there may have been fewer.
  
• Monitoring answer distribution is kind of clunky. This is true with the eInstruction clicker system as well - if you want to see what the answer distribution looks like but don't want students to see it (e.g., if you think the distribution is likely to be mixed and you plan to re-ask the question), the system itself doesn't really let you do this since whatever you see on the computer is what is shown on the projector screen. My workaround is that I usually 'freeze' the projector (so the image on the screen doesn't change even if what I see on the computer monitor changes). This is only possible in some smart classrooms on my campus, not in others.
  
• No technology is 100% perfect. You still need to have a plan for how you will handle things like the system being slow, or students who insist they submitted responses but the system didn't register them. As with any technology, it's best if you go in with a flexible attitude. One of my students commented, "I loved polleverywhere... my only concern would be if other professors would be able to utilize the program as efficiently... Sometimes when professors who are not tech savvy attempt to use such programs, it sometimes eats up more class time just for them to figure out how to use it." PollEverywhere is pretty user-friendly but if you're the type of person who gets flustered easily when things don't go entirely according to plan, you may want to do a lot of practice runs before integrating into a live class.
  

Overall, I think PollEverywhere is a great service, particularly for faculty who want to incorporate clicker-type activities but don't want to make their students buy/register a clicker, or if you really want to ask open-ended questions. I do think it is best for low-stakes activities, particularly given the issues I encountered with cell reception, but it would be great as a back-channel or in classes where the technology is only needed once in a while.