Learning vs. Getting Things Done

The paradox that connects parenting, peer mentoring, and group work

A team of economists investigated how people working at a Fortune 500 company interacted before and after the Covid shutdown. They did this to find the effects of in-person interaction on work.

They found that “sitting together” reduces productivity in the short term but boosts it in the long term. Or as they put it, “We find being near coworkers has tradeoffs: proximity increases long-run human capital development at the expense of short-term output.” When people talk at work they get less done, but they build long-term learning connections. Tell that to your boss when he tells you to stop talking and get back to work. “You don’t understand. I’m being productive in the long term.”

The short-term productivity losses were greatest for senior workers, who were also more likely to provide mentorship.

This got me thinking about group work in schools. Take this paradox, for example: group work projects in school are great for learning, but high-achieving students hate group work.

This is despite evidence that group work, well, works. One meta-analysis found that group work outperforms individual work by 0.3 standard deviations on average – quite a strong result. Exactly which group members benefited the most varied from study to study, but results were often evenly distributed between students, and occasionally high-GPA students learned the most. These are “good students,” right? Shouldn’t they appreciate the extra learning the group work brings?

They do not. This study of biology students in South Korea gives voice to countless GPA-chasing scholars immiserated by perfectly successful group projects. High-GPA students feel pulled in different directions. They have to both teach their peers and finish the project. They must please the teacher by delegating work but would rather just do it themselves. And they feel that their social and academic duties are at odds. Individual work gives them more control.¹

Here’s where I think the disconnect is: learning isn’t the same as getting things done. High-achieving students want to succeed at the assignment. They want to do a good job, to be productive. But their teachers have a different, ulterior, sinister motive: they want their students to learn. Group work is good for learning. But how many students would be willing to take more responsibility, work harder, or even take a lower grade if it meant they (or their peers) learned more? Some, but that would take uncommon maturity.

Students haggle over prices at Mini-Economy Market Day, a group project.

This is an especially pressing issue in the working world. People learn on the job. But in the working world, the goal is productivity. Learning is a side benefit that brings future productivity. Where does learning on the job come from?

Until now, most economists have treated learning on the job as coming from two sources: a) better work environments, and b) some workers are better learners. That is, learning on the job is either about where you are or who you are.

It turns out that those are not the main factors. Economists Will Jungerman, Jacob Adenbaum, and Fil Babalievsky² used French data to find that more than 50% of “Learning on the Job” (the title of their paper) comes from peers. Specifically, more productive workers teach less productive workers. Dr. Jungerman and his colleagues found that the larger the gap in work tenure between work teams, the more the less-skilled workers learned.³

The thing is, teaching the Young Guns slows the Old Hands down. A lot. Dr. Jungerman et al. were able to measure productivity at work (the French government is serious about tracking what people are doing on the job!). They found that more-skilled workers are much less productive when teaching the less-skilled ones. But companies were willing to pay them even more despite the lost work. As Dr. Jungerman told me: “The punchline [is that] more-skilled workers get compensated [paid] for training their less-skilled coworkers. Losses in production from being around worse coworkers are outweighed by the gains in training.” The larger the gap in skill between workers, the greater the gains in learning. The best workers are the best teachers, but teaching slows their work.

The experience + learning gap is the largest and most important at home. It’s between parents and small children. If you’re a parent who is watching a toddler, you can’t get anything done. The thing is, you try. The toddler will be momentarily distracted by something and you furtively try to do a tiny bit of a chore. That’s when the child notices that they no longer have your undivided attention, and demands that you rectify this betrayal. When you are the parent of a small child, you are definitely the Smartest One in the Room. You have to constantly teach the toddler, which makes you unable to get anything else done. Just like in Dr. Jungerman’s research, but in dramatic and messy fashion. The main difference is that, unlike salaried employees, you aren’t paid for it. Instead, you are repaid with cuteness.

If you think about school group work in light of Dr. Jungerman’s “learning on the job” study, you would expect that group work would have three effects: a) slower learners would benefit from their quicker peers and b) the quicker peers would feel less productive and get frustrated. But do you see the missing piece? In the workplace, companies pay their more-experienced workers for teaching the less-experienced ones. This typically doesn’t happen with student group work.

Poor group leaders! They just want to get the job done! Maybe the group project gets slowed down, but the students that are doing the slowing are benefiting disproportionately. I’m not saying the top students shouldn’t be mad about it. But if they demanded some sort of compensation for it, I wouldn’t blame them.

Here are a few takeaways:

1. Teachers: in group work, target both group goals and individual responsibility. Emphasize that the ultimate goal is to learn. Make sure all group members have a new skill to apply, rather than making the project about task-completion.⁴

2. Less-experienced workers: being around more-experienced workers is like getting a free education. Take the opportunity to learn. Stay away from toxic coworkers. They will drag you down.

3. More-experienced workers: you deserve a raise for training the Young Guns. Bring people along with you.

4. Parents of young children: teaching them is the best thing you can do. I know, you can’t get the chores done. But being the Smartest One in the Room means that they are learning so much from you, even more than group work leaders or more-skilled workers.

1. I do not put myself in this group. Once, in a teacher education class, I had a group leader assign me the task of getting The Giving Tree from the library and delivering it to her while she did the entire group project. That’s how much she trusted my abilities.

2. Of the University of North Carolina-Chapel Hill, the University of Edinburgh, and the United States Census Bureau, respectively.

3. The authors used machine learning to connect a mind-boggling number of data points. After they set up their model and pressed “Enter,” it took two days for their program to crawl through the data like a hulking digital spider before returning with the neatly bundled results you find summarized above.

4. Badly designed group work gets bad outcomes. Good group work depends broadly on two things: group goals and individual accountability. Badly designed group work can lead to what John Dewey called “aimless chit-chat.” See here for research on how to do group work well.