ACM Logo  An ACM Publication  |  CONTRIBUTE  |  FOLLOW    

Getting Students to Talk in the Online Course

By Michelle Everson / July 2011

Print Email
Comments (1) Instapaper

©2011 Michelle Everson

Often when I meet other online instructors, I immediately ask them if they use discussion assignments in their courses. My own courses involve quite a bit of asynchronous discussion among students, and I'm always looking for new ideas and new ways of structuring my discussion assignments. Even though keeping up with student discussions in an online course can be a lot of work for the instructor, I personally could not imagine a course in which my students were not activity engaged in dialogue of some sort—I want to provide them as many opportunities as possible to learn from each other.

It's thus surprising when I hear time and again that some of my peers who also teach online statistics courses do not attempt to involve their students in discussion. Sometimes, there are logistical reasons for this, such as teaching very large courses (e.g., 100 or more students) and not having the time or the resources (e.g., help from teaching assistants) to keep up with the volume of discussion that would likely take place; or not being sure about how to grade discussion assignments. Other times, however, I get the impression that some instructors just don't know what they should ask their students to talk about. They worry about creating assignments that would lead to little discussion on the part of students, especially if students all arrive at the same answer right away leaving little to reflect on and react to. Although there are certainly questions we might ask in the statistics classroom that do just have one right answer, there are many other questions we can ask that can be approached in multiple ways, or that might yield multiple correct solutions.

In my own courses —which typically have up to 40 students —I break students into smaller groups of five to six, and they work together to complete a series of small group discussion assignments. Typically, the groups have up to a week to talk through the assignment before one member summarizes the group response to the discussion assignment and submits that summary to me. Over the past several years, I have been on a quest to find appropriate discussion questions and to pose them to my online students. If I were to offer some advice to the new online instructor ready to create discussion assignments that will actually lead to some good "discussion," I would encourage them to think about asking a variety of different kinds of questions that will likely lead to a variety of responses.

One of my favorite assignments is one in which I present each group with five problems that they must solve. The problems require an understanding of sampling distributions, and I ask the group to divide the problems so that each member of the group is responsible for trying to teach the other members how to arrive at the answer for one particular problem. Because I want to make sure that students focus not only on the problem they are required to teach but on all the problems, I tell the students they each will receive a half point of extra credit for each problem that the group correctly solves. This motivates the students not only to get the correct answer to their problem but to make sure the other problems are answered correctly. One member of the group is responsible for summarizing the discussion and sharing the groups' final answers with me. This assignment always generates a lot of discussion primarily because the problems tend to be challenging for the students, and students are sometimes tempted to approach the problems in ways that will not yield correct answers. There are also sometimes multiple ways to solve each problem, and students quickly see this as they discover their own way of working toward the problem differs slightly from how a peer approached the problem, even though the answers they arrived at were the same. I particularly enjoy this discussion because it provides me with many opportunities to question students along the way about their understanding of certain concepts and ideas, and my own involvement in the discussion often spurs on even more discussion among the group.

Another approach that can easily be done is asking students to brainstorm. I do this often in my statistics courses when I present students with a news article in which the results of a study are reported. I ask students to critique the study, identifying its strengths and weaknesses. I also ask them to think about how they might design an even better study. Typically, an assignment like this is one of the first discussion assignments I will incorporate into my introductory statistics course, just as students are beginning to learn about issues related to sampling, surveys, experiments, data collection, and data description. It's rare that students will all identify the same strengths and weaknesses, and they often have a variety of ideas for how to design a new study, so there is typically a lot to talk about.

Discussions can also arise by asking students to explain, in their own words, what a particular concept means to them, and to come up with some of their own unique examples. For instance, I have a discussion assignment related to variability that I sometimes use and part of the assignment involves asking students to attempt to explain what variability is to someone who knows little about statistics. Another tactic I employ is to ask my students to complete an anonymous survey at the beginning of each semester so that we can later use the survey data in several activities throughout the course. For this particular assignment, I might ask the students to predict which variables in the data set would have the most and the least amount of variability (and why) and to then test their predictions by using statistical software to explore the data. Students often pick different examples to focus on and this leads to discussion in which students compare and contrast their chosen examples.

I often ask students to attempt to relate what they are learning about to their own lives or their own fields of study, and this often generates a lot of discussion since students are not all studying the same thing and do not generally all have the exact same interests. When we begin to cover hypothesis tests, I may ask the students to come up with a research question from their own fields of study that is of interest to them and that would lend itself to a particular type of analysis technique (like a two-sample t-test or a paired t-test). Students will then share their examples and talk about issues such as using appropriate notation to write their hypotheses, testing important assumptions, making errors, and trying to determine if their results are both statistically and practically significant.

Those of us who teach know that there are several misconceptions or misunderstandings students might have, and sometimes, an understanding of these misconceptions or misunderstandings can form the basis of a good discussion assignment. For example, I presented students with several statements about confidence intervals where each statement involved an interpretation of a confidence interval, and only one interpretation was correct. Students were asked to identify the correct interpretation and explain not only why that interpretation was correct, but why the other interpretations were not correct. I noticed when using this assignment that some students knew what the correct statement was but couldn't easily explain why it was correct, or why it was better than the other statements. Other students struggled with the assignment and immediately chose an incorrect interpretation as being correct, and this again provided me with some opportunities to question them and to try to get them to think about things in a different way.

I often tell my students that no matter what the discussion topic is, there is always something that can be discussed, and whenever it seems like discussion is at a standstill or that the students are arriving at similar initial responses to my discussion questions, I will attempt to steer things in a different direction by introducing more questions for the group to think about. I hope this will not only lead to more discussion but model the kinds of questions students can be asking each other.

So, with this in mind, do you think there are things that you can ask your students to talk about? If you come up with some great new ideas, remember that I'd love to hear about them!

About the Author

Dr. Michelle Everson is a lecturer in the Quantitative Methods of Education track within the Department of Educational Psychology at the University of Minnesota. She has been teaching online for six years and teaches introductory and intermediate statistics courses, in addition to a course called "Becoming a Teacher of Statistics." In 2009, she received a Distinguished Teaching award from the College of Education and Human Development at the University of Minnesota. She is the editor of resources for the Consortium for the Advancement of Undergraduate Statistics Education, and the statistics editor for MERLOT.


  • Fri, 09 Sep 2011
    Post by Ryan Tracey

    This is excellent practical advice, thanks Michelle.

    I particularly like how you refer to real news articles. To me, authenticity is the driver of discussion because it prompts the learner to *apply* their skills.