Which polling method should I use for Peer Instruction?

posted February 10, 2016 and revised March 30, 2017
by Sam McKagan, PhysPort Director

Several research-based teaching methods, including Peer Instruction, CAE Think/Pair/Share, and Technology-Enhanced Formative Assessment, involve asking students to discuss and answer multiple-choice conceptual questions in class. There are at least three methods of collecting students’ answers to these questions: clickers, flashcards, and show of hands. Lasry 2008 found that when clickers and flashcards are used in a similar way, there is no difference in learning gains between the two systems. However, different polling methods make some instructional practices easier or harder, and there are advantages and disadvantages to each.

Method Advantages Disadvantages
Clickers / personal response systems: This is rapidly becoming the most popular polling method for Peer Instruction. In addition to commercial clicker systems, there are also methods of using students’ personal cell phones or laptops to submit responses.
  • They record how many students gave each response and can instantly create a histogram to give a visual summary of the responses to the instructor and students.
  • Data from clickers can be stored to be used for later formative assessment, grading, or research purposes.
  • Student responses are anonymous to other students – they cannot see what answers other students select so they are less likely to be swayed by peer pressure or feel anxiety about selecting a wrong answer.
  • Student responses appear anonymous to the instructor, which may lead to less anxiety about selecting a wrong answer.
  • Require time and effort to set up and may have technical problems.
  • More expensive than the other methods listed below. This expense may be born by the department or by the students.
  • If students forget their clickers or have technical problems, they cannot participate in voting.
Flashcards: Each student has a set of cards with colors and/or letters corresponding to a multiple choice answer. Students hold up their card in front of their chest so that the instructor can see it but other students can’t. Colors make it easy for an instructor to get a visual sense of the prevalence of each answer by a quick scan of the room. (download pdf card)
  • Cheaper than clickers
  • Faster and easier to set up than clickers and less prone to technical problems
  • If students are instructed to hold the cards to their chests and not look around the room when they answer, they are anonymous to other students, so that students are less likely to be swayed by peer pressure or feel anxiety about selecting a wrong answer.
  • Unlike clickers, they are NOT anonymous to the instructor, which may encourage students to take their answers more seriously.
  • If an instructor wants students to get a sense of how other students answered after all students have voted, they can ask students to hold up their cards and look around the room.
  • Results are not stored for future use for formative assessment, grading, or research.
  • No way to display results to all students without students who answered wrong standing out.
  • Requires time to distribute cards in large classes.
  • The lack of anonymity may lead some students to be reluctant to answer.
Show of hands: The instructor asks students who selected answer A to raise their hands, then those who selected answer B, etc.
  • Easiest method to implement, requiring no time to set up or distribute cards or clickers.
  • Free
  • Requires more time for each question because the instructor must ask for students to raise hands for each option sequentially, rather than collecting all responses at once.
  • Not anonymous to other students or to the instructor, so students may be embarrassed if they chose an unpopular answer and may be inclined to change their answer on the spot if they see that many students chose a different answer.
  • Results are not stored for future use for formative assessment, grading, or research.
  • No way to see a summary of all responses at once.

Here are couple blog posts with discussions of the pros and cons of different polling methods from Derek Bruff and Andy Rundquist.