Kagan Cooperative Learning People Hunt

The People Hunt Structure in Practice

Kagan Cooperative Learning People Hunt is a structured intergroup movement activity where students circulate the room to find classmates who match specific prompts or criteria on a grid or checklist. It looks deceptively simple on paper. You hand out a sheet, students stand up, they mingle, they return. The reality of running it effectively is a lot narrower than people give it credit for. I stopped treating it as a "fun icebreaker" and started treating it as a content delivery mechanism around three years in. That shift alone changed how I designed every single grid.

Kagan Cooperative Learning People Hunt

Here is how it actually works step by step. You prepare a grid, typically 4x4 or 5x5, with prompts in each cell. Prompts can be content-based, like "Find someone who can define osmosis," or personal, like "Find someone who has traveled outside the country." Students use a specific sentence frame to ask: "Excuse me, can you help me? Do you [prompt]?" If the classmate says yes, they sign that cell. If no, the student moves to someone else. The goal is to fill the entire grid or hit a set number of unique signatures within a time limit. The Kagan framework specifies structural elements that make this different from unstructured mingling: simultaneous interaction, positive interdependence, and individual accountability. Every student must actively participate, not just hang back while the louder kids fill the grid. The sentence frame enforces that. Without the frame, you get chaos. With the frame, you get a repeatable classroom routine that runs in about 10 to 15 minutes for a 16-cell grid with 20 to 25 students. I learned this the hard way during my second year. I designed a People Hunt grid with prompts like "Find someone who likes the same movie as you." In a class of 28 students, seven kids ended up signing the same three popular students repeatedly. The remaining 21 students were bouncing between each other in overlapping clusters. I had essentially created a popularity bottleneck. The fix was straightforward but not obvious at first: I started seeding the grid with prompts that required specific, uncommon answers, and I added a rule that each signature had to come from a different person. That eliminated the clustering problem entirely. After that change, completion rates went from roughly 60 percent to about 90 percent within two minutes of starting.

The deeper issue most people miss is that People Hunt only works well when the prompts have a high discriminative value. A prompt like "Find someone who has a brother" will get signatures from maybe 60 percent of the class, which means the grid fills unevenly and some cells become race-to-the-first-person-who-fits. Good prompts have answers that distribute more evenly across the population. For a 25-student class, you want each prompt to fit roughly 4 to 7 students. That means avoiding demographic-heavy prompts and leaning into content knowledge, opinions, or experiences that are genuinely variable. Another counter-intuitive point: People Hunt is not ideal for introducing new content. It is a review and reinforcement structure. If you use it to teach a concept for the first time, students will confidently sign off on incorrect information and walk away with misconceptions locked in. I used to do this on Monday mornings when I was trying to be creative. By Wednesday, the wrong answers had spread through three different groups. Switching to content review grids after direct instruction cut my correction time in half. There is also a timing ceiling. Beyond about 20 minutes, the energy drops and the structure becomes mechanical. Students stop engaging with the content of the answers and start treating it as a ticking clock exercise. I learned to keep grids to 9 to 12 cells maximum for a first pass, then run a second shorter round if needed. Two quick rounds beat one long dragging round every time.

The activity fails completely in classes smaller than 10 students unless you modify the structure. With 8 students, you will run out of unique partners before anyone fills the grid. The workaround is a modified People Match where students rotate through a circle and each person presents one fact, and others record it. It is not a true People Hunt, but it preserves the cooperative structure without the math breaking down. You do not need special software for this. Most teachers I know build grids in Google Sheets or Microsoft Excel and print them as PDFs. Some use Canva for nicer templates, but the tool does not matter. What matters is the prompt design. I spend about 20 minutes per grid, sometimes more for content-heavy units. The reward is usually 10 minutes of highly structured student interaction that doubles as an informal assessment. If you want a ready-made template to start with, a basic 4x4 grid generator in Google Sheets is easy to build yourself. Set up columns for the prompt, signature space, and date. Share it with a copy for each student. That is the simplest version. Paid resources and premade collections exist, but the real differentiator is always your prompt quality, not the template itself.

I stopped using this strategy for large lecture-style classes with 35 plus students. The logistics of movement and monitoring break down past that number. For those situations, I switch to Quiz-Quiz-Trade or Numbered Heads Together, which scale better. People Hunt has a clear upper limit.