Does anonymity matter? (#clicker series)

by Stephanie Chasteen on April 1, 2013

hand-on-clicker-100pxI have just finished reading an interesting article, “Anonymity and in class learning:  The case for electronic response systems,” (Freeman, Blayney and Ginns, Australasian Journal of Educational Technology, 22(4), 568-580, 2006).  It’s worth a read just for the very nice and thoughtful review of the literature on electronic response systems (clickers) and benefits of anonymity.

This research presents the results of a study of alternative response methods for in class formative questioning. Students’ anonymity from their peers and instructor was studied through a research design that maintained a constant interactive teaching strategy in a large lecture hall, in all respects except for the method used by students to respond to the in class questions. A handheld electronic response keypad was the only approach affording complete anonymity. Student perceptions of the benefits of anonymity were obtained from a survey conducted at the end of the course. The results suggest that anonymity is a critical factor affecting student willingness to participate with in class exercises. Furthermore, the results indicate that students’ propensity to engage with in class questions increases with the degree of anonymity provided to the student in revealing their response. The benefit of anonymity, combined with the increased availability and affordability of electronic response systems, will be of interest to academics keen to design engaging learning environments.

There is certainly a good argument to be made that anonymity should affect student learning — students are likely to be reluctant to share their ideas and to engage deeply in discussions if they are uncertain of their answers and fear public ridicule.  Some previous research here at CU suggests that this is probably even more true in non-majors classes, where students aren’t content experts.  The current research is a follow-up to a previous study by the same authors that compared raising hands to clickers, and found that students’ perceptions of their own understanding and interaction was higher when using clickers.   Were these results due to the anonymous nature of clickers?  That was the question underlying this new research.

The sample was a second year introductory management accounting course in Australia.  The instructors used Peer Instruction during questions, where students were encouraged to discuss the answer prior to the vote.   Students then voted with a clicker or by a show of hands.   It isn’t clear if students were given points for participation or for correctness.  However, it appears that, after the student vote, instructors typically displayed the histogram of student responses (or verbally indicated the correct answer), and then indicated why they favored the correct answer, rather than holding a student discussion about the reasoning behind the answers.  This, to me, frames the question as an “answer-finding” exercise rather than a “sense-making” one, and could sway the results.  Indeed, the researchers found that students were not in clear agreement as to whether or not peer discussion helped their learning; I would argue that this finding may be due to the fact that those peer discussions did not clearly relate to the post-vote discussion.  If instructors are not soliciting student ideas after the vote, then they have not shown that they value the reasoning that students generated during their discussions.

What is of particular interest is student responses regarding anonymity:

  1. I preferred answering questions when my answers were anonymous to the instructor:  68% agree
  2. I preferred answering questions when my answers were anonymous to other students:  62% agree
  3. Anonymity was more important with questions when I was uncertain about the answer:  63% agree

So, students do indicate that anonymity is important, but I notice that this is far from unanimous.  About 2/3 of students care that their answers are private, or that this is important when they are uncertain.

They also asked about student preferences for polling method, and electronic response systems were far and away the most popular method, followed by a show of hands. Individual responses (either by indicating that the student knew the answer, or by being selected by the instructor) were, not surprisingly, the least popular.  So, the most anonymous methods are also the most popular.

So, it’s good to know that students do prefer anonymity, but  — to the authors’ own admission — the focus on student self-report does limit the value of the findings.  I would be curious to know how such preferences correspond to instructors’ facilitation practices, for example.  I can imagine that an instructor who takes great pains to make it clear that the class is focused on discussing and debating ideas will have students who are more comfortable with non-anonymity, compared to an instructor who focuses on memorization and knowing the right answer.


Julie April 6, 2013 at 3:12 pm

#awesome post, thanks! In our lab, Lasry found that at least pertaining to physics and the FCI – there wasn’t a significant difference b/w flashcards and clickers on gain (see That surprised me! I think student preference for anonymity is a good enough reason to transition, but it is also interesting that at least in this one case “anonymity” did not correlate with gain on a standardized test. Have you found any competing evidence to the Lasry finding?

Stephanie Chasteen April 12, 2013 at 5:23 pm

Hi Julie, yes, I’m familiar with that paper. It also surprised me. However, there’s a piece of data that I often show from a study that compared correctness on questions answered through hand-raising, response cards, and clickers. They found that more students got the right answer when raising their hands, fewer when using response cards, and the least got the right answer using clickers. I use this to suggest that the instructor and students are getting a more accurate understanding of student understanding with clickers. They also found that response rates were higher with response cards and clickers. Here is a summary of that article:

Comments on this entry are closed.

Previous post:

Next post: