Developing effective, accountable teams in sports journalism ethics class
- 7 hours ago
- 5 min read
Ethics in Sports Journalism is a three-credit, in-person class for sports journalism undergraduate students at Arizona State University. I have taught it and its graduate equivalent 13 times since I joined the faculty in 2015.
The course’s summative assessment is to research and present a case study of something a sports journalist or sports media organization did that students consider unethical. For example, Sports Illustrated using artificial intelligence-generated bios in its content. NFL reporter Dianna Russini and New England Patriot’s head coach Mike Vrabel’s relationship. Or Pat McAfee amplifying an internet rumor about Ole Miss student Mary Kate Cornett on his show. In groups of two or three people, students research several elements of the case, then presents their findings to the entire class.
I have seen multiple student groups present high-quality, insightful case studies. I genuinely enjoy watching and grading their presentations. But several years ago, I began to notice a pattern: Students, whose groups presented A+ case studies, telling me after-the-fact that they had actually done most of the work themselves. And to make matters worse, all group members received the same stellar grade.
I wanted students to work in groups. To become teams. But I certainly didn’t want to reward free riders, either. I needed a way to structure effective groups, to catch underperforming students early, and to allow students a way to constructively hold each other accountable.
I dug into the cooperative learning research. My findings went on to win the 2025 AEJMC Mass Communication & Society Teaching Ideas Award, and was published last week in Teaching Journalism & Mass Communication.
Since the 1980s, education psychology scholars such as brothers David W. and Roger T. Johnson at the University of Minnesota have been examining and re-examining cooperative, competitive, and individualistic goal structures’ effects on student achievement in American education. Multiple studies have since-then confirmed that cooperation is more effective on individual achievement and productivity than interpersonal competition. Richard M. Felder and Rebecca Brent found that struggling students are less likely to give up, and successful students find gaps in their own knowledge by explaining concepts to other students. Small-group learning can result in longer information retention, deeper levels of understanding, and enhanced communication skills among student participants.
Anyone who has worked in a small group knows that these benefits are not automatic. Many students strongly – and I do mean, strongly – dislike working in groups. For cooperative learning to be effective, individuals’ outcomes must depend on both their own actions, and those of their group members. This positive interdependency, Morton Deutsch penned in 1949, influences group members through three processes: substitutability, cathexis, and inducibility. Substitutability is the extent to which one person’s actions can be substituted for another. The second process is cathexis. According to the Chicago Psychoanalytic Institute, cathexis (kuh-THEK-suhs) is “the mental energy a person invests in themselves, others, or objects. … it explains how we become attached to people, ideas, or goals, and how that investment impacts our emotional well-being.” The third variable is inducibility, or a person’s willingness to accept input from others and to adjust their behavior accordingly.
A component of these processes is the ability to hold each other accountable. R. W. Brown proposed an adjustment factor that allows students to assess themselves and each other on a class project. Let’s say, for example, there are four people in a group. They earn an 80% on a class project. They “grade” each other based on their perception of that person’s contributions. If one student awarded themselves 87.5%, and their teammates gave them “grades” of 100%, 87.5%, and 87.5%, the instructor would take the average of these four votes (90.62%) and divide it by the team score (80%) to calculate what an adjustment factor. For this hypothetical student, 90.62/80 = 1.13. Then, this adjustment factor is subtracted from the average to calculate the student’s final grade. In this example, 90.62 - 1.13 = 89.23%. So, though the team received 80% for its project, this individual student would instead receive 89.23%.
I combined and modified Barbara Oakley and colleagues’ Team Member Evaluation measures, Richard Felder and Rebecca Brent’s self-assessment of team function guidelines, and Brown’s autorating system into one framework, piloting these measures across three sections between 2022 and 2025. For the Spring 2022 undergraduate and fall 2025 masters classes, I put students in three-person groups. For the Fall 2023 undergraduate class, students self-selected into pairs.
After teams formed, groups met in class to have their first team meeting. During this meeting, students completed an open-ended questionnaire. Two weeks later, teams met again during class, completing another questionnaire. After their presentation, students completed the third and final questionnaire that measured substitutability, inducibility, and cathexis.
Overall, 104 students participated. All were undergraduate students, except the 38-student Fall 2025 masters class.
I ran a multiple linear regression to test how substitutability, cathexis, and inducibility predicted students’ experiences working in groups. The model was statistically significant, explaining about 80% of the variance in students’ experience working in their groups.
These measures, as well as students’ ability to assess their peers, opened my eyes to group dynamics I otherwise would not see. For example, Students R, K, and E were in a group together and earned an 83% on their case study. In their exit questionnaire, Student R said Student E “hardly did anything for the project. He kind of just rode our coattails. In all honesty, [Student K] did the majority of the work.” Student R went on to describe the above-and-beyond things Student K did to bring the project to the finish line. With Brown’s (1995) adjustment factor, Student E’s grade would have dropped from an 83% to a 73%. Group hero Student K, on the other hand, would have seen his grade rise from an 83% to a 97%.
One thing I did with the fall 2025 masters students that I didn’t do with the others – and that I will do going forward – is that I selected their groups based on students’ availability outside of class. At the beginning of the semester, I had them give me their schedules, and I organized them in three-person groups based on overlapping availability. Oakley and colleagues recommended doing this, as well as making sure team members who hail from relatively at-risk groups not be isolated. “Relatively at-risk” varies by discipline. For example, women in engineering studies have a relatively higher dropout risk, so Oakley et al. recommended putting first- and second-year women in groups of either “all women, two of each, or two or three women and one man, but not two or three men and one woman.”
I also saw a relationship between degrees of substitutability, cathexis, and inducibility, and whether students self-selected or were placed in groups. Students who chose their own group had far more negative experiences than those whose groups I selected. This echoes what scholars such as Oakley and colleagues already concluded: Self-selected groups are more likely to be comprised of friends, and therefore, more likely to cover for each other than to tell the instructor when there is a problem.
When using the Brown (1995) adjustment factor, I recommend having at least three people in a group and following up with students who score a peer abnormally higher or lower than the other group member did. For example, one of my students rated a team member all 1s, which stood for “strongly disagree” on the 5-point Likert scale, when answering statements about that team member’s contribution. The third group member, however, evaluated that same person with all 5s (“strongly agree"). It turns out, the student who gave the teammate all 1s thought 1 stood for “strongly agree” instead of “strongly disagree.” In such instances, an instructor needs to contact students for clarification before posting that grade.
The full study is available at Teaching Journalism & Mass Communication.