Article: Freeman, Blayney, & Ginns (2006)
Reference: Freeman, M., Blayney, P., & Ginns, P. (2006). Anonymity and in class learning: The case for electronic response systems. Australasian Journal of Educational Technology, 22(4), 568-580.
Summary: In an effort to investigate the importance to students of the anonymity afforded by classroom response systems, Freeman, Blayney, and Ginns surveyed the 139 students in an introductory management accounting course at the end of the term. Notably, the majority of students in this course were female (73%), 20-22 years old (93%), with non-English speaking backgrounds (82%).
Each three-hour class session in this course featured a number of “formative, mainly rules based, multiple choice questions.” Students responded to these questions via clickers or hand-raising in alternating class sessions and were encouraged, but not required, to discuss their answers with their peers before responding. The instructor would announce the correct answer to each question immediately after the distribution of responses was shared with the class and then practice “agile teaching” by using the results of the question to guide subsequent lecture and discussion. This alternating system of response methods ensured that students were given opportunities to experience a response method that allowed students to remain anonymous (clickers) and one that did not (hand-raising).
Sixty-eight percent of students agreed or strongly agreed with the statement “I preferred answering in-class questions when my answers were anonymous to the instructor.” Almost as many students (62%) preferred answering questions when their answers were anonymous to their peers. About the same number of students (63%) agreed or strongly agreed that “anonymity was more important with in-class questions when [they] were uncertain about the answer.”
Additionally, students were asked to rank in order of preference four potential response methods-clickers, hand-raising, volunteering to respond when they knew the answer, and “cold-calling.” Average student rankings of these four methods indicated a preference for anonymity, as the methods were ranked (on average) in the same order as I just listed them.
The authors conclude from these survey data that students value the anonymity (peer-to-peer anonymity and student-to-instructor anonymity) enabled by a classroom response system. Since other response methods, such as hand-raising and response cards, are less anonymous, this argues for the use of clickers.
The authors make a couple of additional points about the potential interplay of culture and use of clickers. One is that peer instruction was not as highly valued by these students as other aspects of learning with clickers, according to survey results. It’s possible that the cultural diversity of the students (as measured by proxy by the percentage of students with non-English-speaking backgrounds) made peer instruction less useful for these students than for more homogeneous students surveyed in other studies who found peer instruction more useful.
The other is that demographic variables, including native language, had no significant impact on the student responses to survey questions, particularly the questions about the importance of anonymity. “Using first language as a rough proxy for culture, and in particular openness to criticism, these results might be seen to contradict Banks (2003) who suggested that cultural background could impact preference for ERS usage.” The authors suggest further study of the role of culture in students’ valuing of anonymity.
Comments: For a relatively simple survey-based study, this article raises some interesting questions. For instance, given that the instructor of this course practiced agile teaching, altering his instruction based on the distribution of student responses to in-class questions, I would think that response methods that generated more honest responses from students would lead to more useful agile teaching. That is to say, if students were less likely to indicate confusion about a question when the hand-raising method was used (because their answers would not be anonymous to their peers), then the instructor might overestimate his students’ comprehension of the topic at hand when relying on the response distribution and subsequently spend less time on the topic than might actually be warranted. If true, then students who prefer anonymous response methods and who have instructors who practice agile teaching might benefit more from an anonymous response method than students whose instructors do no practice agile teaching. Since I have yet to read a study that compared use of clicker questions with and without agile teaching, it’s unclear at this point how important this issue might be.
The questions raised about the interface of culture, participation, criticism, and anonymity are also very interesting ones. They are ones that haven’t been on my radar until recently, when I heard a presentation by Parvanak “Pary” Fassihi, who described her use of clickers to engage students in English-as-a-Second-Language (ESL) classes at Boston University and the Boston campus of the Showa Women’s University (of Japan). She shared that some of her students come from cultures where it is seen as impolite to disagree with others publicly. These students are often very hesitant to engage in the small-group and classwide discussion of clicker questions that Pary tries to generate in her courses. (I’ll note here that I interviewed Pary for my book. She had a lot of interesting things to say about the role of clickers in her language instruction courses, but this observation about cultural views on public disagreement didn’t come up in our interview.)
Pary’s observation is consistent with the finding by Freeman, Blayney, and Ginns that their students, many of whom came from diverse cultures given that few of them spoke English as a first language, did not value peer instruction as highly as might be expected. I imagine it is also consistent with Banks’ suggestion, mentioned in the article, that “cultural background could impact preference” for clicker usage. I haven’t read the Banks article, but I hope to do so soon and review it here on the blog.
One criticism I would have of the article at hand is that comparing the clicker and hand-raising response methods doesn’t quite isolate the effect of anonymity since clickers offer a number of advantages over hand-raising that might be at play here. Given the nature of the survey questions used, this isn’t a significant issue, since student perception of anonymity was the focus, not a control-group-study comparison of two difference response methods. However, it is possible that the students’ preference for clickers over hand-raising over hearing from student volunteers could be an indication not of their preference for anonymity but of their preference for response methods that encourage more complete and/or more independent participation and/or greater accountability for participation. Thus, I would argue that those data in particular don’t necessarily imply that students value anonymity when responding to in-class questions.
(Cold-calling also leads to more complete and independent participation and greater accountability for participation since students have to stay on their toes in case they are called upon. However, I’m comfortable arguing that the stress generated by this method would outweigh any perceived benefits regarding participation or accountability.)
A better comparison might be between having students respond via clickers with and without a display of individual participant responses. (Many classroom response systems have the ability to display on screen not only a histogram showing the distribution of responses but also a list of individual students and their particular responses.) That would better isolate the anonymity factor from other advantages of clickers over alternate response methods. It’s not a perfect solution, in part because students in a large class aren’t likely to know each other’s names and so might not be bothered by having their names displayed next to their responses and in part because this is a feature of classroom response systems that is infrequently used by instructors so its use in this context might seem too artificial. Also, this plan would not isolate peer-to-peer anonymity from student-to-instructor anonymity. However, I think it would yield more relevant data than comparing clicker use with hand-raising.