July 31st, 2013

Better Group Work Experiences Begin with How the Groups Are Formed


Like many matters regarding teaching and learning, there isn’t one best way to put students into groups. The best way is related to what you want students to learn from their group experience. Here’s a brief discussion of how that works for three common ways of forming groups.

Randomly formed groups – Students join with others sitting nearby or the teacher creates groups using some random method like birthdays, house numbers, last digit in a cell phone number, etc. The advantage of this approach is that it’s quick. The method is also a good one to use if you want students to meet more of their classmates. However, it’s not a good method if you want students working in groups where there is a diversity of backgrounds, experiences, and knowledge. Indeed there’s research indicating that you shouldn’t use randomly formed groups when students are tasked with completing a project because there’s no guarantee that the group will have the necessary range of skills to do so.

Student-formed groups– Students form their own groups, selecting members from among their classmates. This is often the methods students prefer. If they know others in the class, they tend to select their “friends.” If they’ve taken classes together before, they select others they’ve worked with successfully and avoid those not as motivated as they perceive themselves to be. Students report higher levels of satisfaction with group experiences when they have selected their members. However, this may not be the best way to form groups when the goal is high-quality products. To work effectively with “friends,” students must transition from social relationships to task-oriented ones and that transition isn’t always easy. Furthermore, this method of forming groups doesn’t help students learn how to work with people they don’t know.

Teacher-formed groups – Teachers assign students to groups using any number of different criteria. The one usually mentioned first is ability and there are differing views as to the merits of putting the best students together, those performing less well together, or creating groups with a range of abilities. When the least able students work together, they learn the least about the content and group processes. They stand to lose the most. When the brightest students work together, you often end up with a group with lots of leaders and few followers.

There are other criteria teachers can use to form groups and they may be more important than ability. For example, what skills, previous experiences, and background knowledge are needed to successfully complete the project assignment? Some teachers use that question to generate a list of what would help the group do its work. For example, if the assignment is to construct a website, the group needs one or more members with web development experience and knowledge, somebody with good graphic design skills, and at least one person who writes well, for starters.

In addition to task criteria, groups also need members who can contribute those intangible things that help groups function effectively—groups needs leaders, members who can encourage others to participate and contribute, and members who aren’t afraid to challenge ideas and question the group’s decision-making processes. Some teachers transform this collection of experiences, skills, knowledge, and abilities into a survey which students complete. Teachers then use the data to form groups. If it’s an upper-division course and students have had previous experiences doing group projects, the teacher might involve them in creating this list of relevant background knowledge and experiences. When teachers are transparent about the criteria being used to form groups, groups are able to begin their work together by discussing who has these resources and how the group will use them.

When teachers use knowledge and skill criteria to form groups, students have a greater chance of experiencing a group that accomplishes more than they could as individuals. It’s also a good method if the goal is for students to learn how to work with others—others they don’t know well, possibly some they don’t even like. This mirrors what happens in professional situations. We don’t get to chair search committees and populate them with our favorite colleagues, nor should we.

Designing good group learning experiences involves thoughtful planning and that starts with how the groups should be formed.

Please share your favorite method for putting students into groups as well as other strategies that help ensure a positive group work experience for your students.

  • Carlos Diaz

    Prior to teaching, I conducted Managerial Grid Seminars, and formed teams randomly with success. Since then, I have formed groups in my college classes in that fashion, and with few exceptions, they worked. Students always complain about not being with their friends, neighbors and whatever, but at the end of the semester, on their evaluation of the team experience, they appreciate the effort and learning experience. I believe that acquiring the ability to work with others is a valuable tool for what comes next: REAL WORK.

    • Joni

      What is a Mangerial Grid Seminar?

  • Glena Andrews

    I value group work because I believe it is good practice for the work world. We usually don't get to pick our friends for committee work, task forces, project work and other types of working groups. I have yet to find a student who looks forward to group work. I often hear "I hate groups" comments followed by "I always have to do all the work" comments. I have used several methods to assign groups including major (mixing them in the group membership), classification (having upperdivision students with lower division students), and background (those having foundation classes with those without). Sometimes I have the group elect a leader, other times I allow them to let the leader emerge as they begin. When they are informed that they will evaluate the group process including their own contribution as well as their group members, I seem to have more successful groups.

    • Carlos Diaz

      Ditto on the learning value. Compliments on trying different methods. On group leadership, the group leader will emerge naturally, sometimes change during the semester due to bad results.
      For better results during the semester, I have found a critique methodology that works: first, the team has to report on each task "who did what" telling the professor and the rest of the class which individual did what part of the work assigned. Then each individual "grades" each other member (usually between four to seven) on a scale 1/10 their contributions to the team effort. I encourage them to be honest on both efforts, and IT WORKS after one or two assignments. This technique I carried over from my Grid Seminars, thank you Dr. Blake and Mouton.

  • Giles Brennand

    Does anybody have a postgraduate level case where students have to assign various individuals to, say three, teams with quite different responsibilities?

    I think the discussion about their results, the process they followed, the considerations they made would be very helpful for the students selecting members of teams in future

  • Glenda Brown

    I teach an on-line program evaluation course. In real life, faculty do not get to chose which faculty are on their team; but, when doing program evaluation (especially when preparing for accreditation), the end results impacts the entire team. In the course, the end results impacts each member's grades. It is traumatic for students when the group results are less than the ability of an individual member of the team. My courses are six weeks long, which is not always sufficient time to develop synergy among group members. I frequently wonder if there is some way to handle groups so that the more capable students are not penalized for working with less talented or less motivated students. I like the idea of taking into consideration the peer reviews when determining final grades.

  • Debbie Johnston

    I teach a first semester course in which students must work together in franchise teams running a virtual business. I let them select their teammates, but only by following a specific process. At the beginning of the semester (before they set up teams), I ask the students to identify (on 3×5 cards) their target course grade as well as the amount of time they realistically think they will put into their studies each week. (I record this information for my records, since it serves as a great discussion starting point when students miss their target grade in tests…i.e. "You said your grade goal was 80% but you only achieved 60%. What do you think you will need to do differently in order to achieve your goal?") (Continued in next post)

  • Debbie Johnston

    (Continued from previous posting)
    When we set up teams in Week 3 of the semester, I return the cards to the students so they can find teammates with similar grade goals and who are willing to make similar time commitments. The next step is to identify a two-hour block of time that everyone in the proto-team can commit to each week for their mandatory meeting. (If members cannot find a common time, they split up to work with different students who have both similar goals and availability.) The third step in the process is to discuss who will play which roles in the team (these roles are specified by me). Once again, if problems arise, they can split up and find different teammates. This whole process takes no more than an hour for a class of 40 people. Their first deliverable as a team is to build a contract, and all of this "background information" must be included within it so I know they have followed the process appropriately. (Continued in next post)

  • Debbie Johnston

    (Continued from previous posting)
    I've used this method for the past several years and experienced remarkable success with it. While the process does not eliminate all team problems, I would estimate that it has reduced them by about 75% compared to either letting the students choose their teammates without a process, or to having me randomly assign them to groups. People with similar grade expectations work together, so it is rare that one person ends up doing the bulk of the team's work. Because they are willing to dedicate similar amounts of time to the work each week and have already agreed upon meeting days/times, these issues are also eliminated. In addition, there are no arguments over roles since those were agreed-upon at the outset. When someone doesn't pull their weight as the semester progresses, I've observed that the teams are much faster at proactively identifying and addressing the issue themselves because they had that initial "expectations" discussion at the outset. I think this happens because the students had control over the selection of their teammates, so they take stronger ownership of the group's success.

  • Tom Smith

    In our Senior Capstone Seminar, we put groups of five or six together by program (we attempt to have each student from a different major) and then balance the groups by gender. We have had very good results.

  • LLappin

    I started using Myers-Briggs scores for creating the first groups at the beginning of the semester. I have 5 groups of 5 at the end. I ignore the first two letters of the score and group them by the last two. This is inexact, but it breaks up clicks, gives them a topic to discuss (their scores and what they mean) early in the semester. After that, for the second project, they form groups based on interests (for a writing assignment). Students need help designing their own group contracts and I monitor their group time in class–to be sure they are on task (I ask for a list of the tasks they will be working on for each project).
    Group work divides the session so that I am not always lecturing. I usually start with questions about the topic, discussion, then a short lecture, some individual writing, then group work. an hour and a half goes by very quickly.

  • Pingback: Advice on strategically creating groups | teachingpublicspeaking()

  • Pingback: To Group or Not to Group | Expanding the Learning Journey()