Group Activities with Simulations: Roles and Routines

Group activities with simulations go wrong in a familiar way: one student holds the device, two watch, and one drifts off. The simulation isn't the problem. The structure is. This guide gives you a structure that works in a normal classroom with three or four students per device. You will set up one link per group, hand out four roles that give every student a job, and choose from five routines (jigsaw, pooled data, station rotation, claim boards and the prediction showdown). Two worked examples follow, in photosynthesis and ecology sampling, with the numbers each simulation produces so you can check groups at a glance.

For the whole-class demo that often comes before group work, see teaching with a projector. For the bigger picture, start with how to use interactive simulations in the classroom.

Why simulations suit group work

A simulation gives a group something real to argue about. When the graph goes the "wrong" way, four students have four explanations, and the conversation does the teaching. Three features make simulations especially good for groups:

• Cheap trials. A group can test an idea in ten seconds, so the quiet student's suggestion gets tried instead of dismissed.
• Shared evidence. Everyone sees the same screen, so disagreements are about what it means, not what happened.
• One device is enough. You don't need a laptop per student. One phone, tablet or laptop per group of three or four works well, and fewer screens mean more talk.

The risk is passengers. Roles and a clear product, such as a table, a graph or one submitted answer, are what stop them.

Set up: one device and one link per group

Spend ten minutes before class and the lesson runs itself.

1. Open the simulation and set the starting values you want groups to begin from.
2. Click Share. On the Share Links tab, under New link, create one link per group and name them "Group 1", "Group 2" and so on. Each link can pin different starting values, which is how you run a jigsaw or a station rotation from one simulation.
3. On the Questions tab, build one question set and choose it for every group link. Use the instructions box for the task and the role reminders.
4. Print or project each group's QR code, or post the links in your LMS. Students don't need accounts.
5. Tell groups to submit once, as a group. In the name box, they type the group name and first names only, for example "Group 3 Ana Ben Kai". Never full names.

One submission per group has two advantages. Each group must agree on its answer before it sends anything, which is where the arguing happens. And it keeps you well inside the free plan's 100 student answers per month: six groups use six answers, not thirty.

After the lesson, open the question set on the Questions tab and choose All classes. Because each group has its own link, the % correct by class table becomes a % correct by group table. You see at once which group needs a visit next lesson. View answers on a single link shows that group's full answers, and Download CSV exports them.

Four roles that give everyone a job

Assign roles before anyone touches the device. Rotate them after every run or every 10 minutes, so nobody is the permanent recorder.

Role	Job	Says things like
Driver	Controls the device. Changes only what the group agreed.	"We said one thing at a time. Which one?"
Recorder	Writes every reading in the table, with units.	"Wait, read that again. Was it 8.7 or 8.1?"
Skeptic	Checks the method and challenges conclusions.	"Did we change two things? How do we know it's not chance?"
Reporter	Types the group's answers and explains them to the class.	"Our claim is... and our evidence is..."

With three students, merge Skeptic and Reporter. With pairs, use Driver and Recorder and swap halfway.

Two rules make roles stick. First, the Driver can't decide alone: no change until another role agrees. Second, anyone can be asked to explain. When you visit a group, ask the Recorder or the Skeptic, not the Reporter. Groups quickly learn that everyone needs to understand the result.

Five routines for simulation group work

1. Jigsaw by variable

Each group investigates one variable and becomes the class expert on it. Then students regroup with one expert from each group and combine findings. Use it with any simulation that has several independent sliders. The photosynthesis example below is a full jigsaw.

2. Pooled data

Every group runs the same experiment, and the class pools the results. This works best with simulations that contain chance, such as sampling, genetics or radioactive decay, where one group's result is unreliable and the class total is not. The quadrat example below shows how big the difference can be.

3. Station rotation

Set up four stations, each with a different simulation or a different link to the same one, and rotate groups every 10 to 12 minutes. Give each station its own short task card. Name the links after the stations, not the groups, so each station's answers collect in one place.

4. Claim boards

Each group writes a claim, its evidence (two or three readings from the simulation) and its reasoning on a large sheet. Groups then walk around and leave one question on every other board. This routine works well after an inquiry-based lesson, when groups have reached different conclusions.

5. Prediction showdown

Add a prediction question to the set. The simulation stays locked until the group commits, so they have to agree first. Groups that disagree must record the minority view too. After running the simulation, the same question is asked again. This is Predict–Observe–Explain in small groups, and it gets far more students talking than a whole-class vote.

Worked example 1: a photosynthesis jigsaw

This lesson takes about 50 minutes for ages 15–17. It uses the photosynthesis rate simulation: pondweed under a lamp, with sliders for light intensity, CO₂ and temperature. A line under the graph shows gross photosynthesis, respiration and net photosynthesis in mg CO₂/dm²/hour.

Pin on every link: light 500 µmol/m²/s, CO₂ 400 ppm, temperature 25 °C. At these values the simulation shows gross photosynthesis 10.7, respiration 2.0 and net 8.7.

Instructions for students (paste into the set): "Driver: change only your group's variable. Recorder: write net photosynthesis from the line under the graph for each value. Skeptic: check that the other two sliders haven't moved. Reporter: answer your group's question below. In 20 minutes you will teach your result to a new group."

Expert groups (20 minutes). Each group changes only its own variable, records net photosynthesis at five values, and switches the graph axis to match.

Group	Variable	Values the simulation gives (net)
A	Light	0 → −2.0, 250 → 6.1, 500 → 8.7, 1000 → 10.9, 2000 → 12.3
B	CO₂	50 → 0.7, 200 → 5.5, 400 → 8.7, 800 → 11.7, 1500 → 13.7
C	Temperature	15 °C → 2.5, 25 °C → 8.7, 28 °C → 9.0, 35 °C → 4.1, 40 °C → −1.5
D	Lamp on and off	Lamp off: −2.0. The readout gives the light compensation point: about 35 at 25 °C.

Each group answers one question in the set:

• A: "By what percentage does net photosynthesis rise when light doubles from 500 to 1,000?" Answer 25, tolerance ± 2, unit %. From 1,000 to 2,000 it rises only about 13%: the curve is saturating.
• B: "By what percentage does it rise when CO₂ doubles from 400 to 800 ppm?" Answer 34, tolerance ± 2, unit %.
• C: "At which whole-number temperature is net photosynthesis highest?" Answer 27, tolerance ± 1, unit °C. Net photosynthesis peaks at about 9.1 there. At 40 °C respiration (5.7) is larger than gross photosynthesis (4.2).
• D: "What is the light compensation point at 35 °C?" Answer 122, tolerance ± 3. It is about 58 at 15 °C and lowest, about 34, near 23 °C.

Home groups (20 minutes). Students regroup with one expert from A, B, C and D. Their task: "A greenhouse grower can afford one upgrade: brighter lamps (500 to 1,000), extra CO₂ (400 to 800 ppm) or heating from 15 °C to 25 °C. Which gives the biggest gain in net photosynthesis? Use every expert's data." Each expert has only part of the answer. Groups then test all three options from a 15 °C start: net photosynthesis goes from 2.5 to 3.2 with brighter lamps, 3.5 with extra CO₂ and 8.7 with heating. The reason is the most useful part: at 15 °C, temperature is the limiting factor.

Close (10 minutes). Reporters give one claim each. Finish with a question no single group can answer alone: "Why does the compensation point go up at 35 °C?" Respiration doubles every 10 °C in this model, so the plant needs more light just to break even.

Worked example 2: pooled quadrat data

This lesson takes about 40 minutes for ages 13–16. It uses the quadrat sampling simulation: a 20 m × 12 m field with daisies, clumped dandelions, plantain and ferns. Students place 1 m quadrats at random and the simulation estimates the population as mean per quadrat × 240 m² ÷ quadrat area.

Pin on every link: habitat "Woodland shade to open sun", random placement, quadrat side 1 m, 10 quadrats, species Dandelion. Every group gets the same field when the page opens, so tell groups not to press New field.

Each group (15 minutes):

1. Press Sample automatically. Record the estimate for dandelions and for daisies (change the species to read it).
2. Press Clear quadrats and repeat twice more. Record all three runs.
3. Note the true count the simulation shows. Real ecologists never get this number, which is exactly why the activity is useful.

The field has 142 dandelions and 141 daisies, almost the same. The estimates are not alike at all. With 10 quadrats, 80% of dandelion estimates fall between about 24 and 312. Daisy estimates are much tighter, between about 72 and 216.

Pool the data (15 minutes). Collect every group's estimates on the board and average them. Six groups × 10 quadrats is 60 quadrats, and the spread shrinks: about 92 to 224 for dandelions and 108 to 172 for daisies.

Discussion question for the set: "Both species have about 140 plants. Why are the dandelion estimates so much more variable?" Model answer: dandelions grow in clumps, because their seeds fall near the parent plant. A quadrat either hits a clump or misses it. Daisies are spread out, so each quadrat gives a similar count. Clumped species need more quadrats.

For more biology activities, see interactive biology lesson ideas. To turn either example into a full write-up with a data table and analysis questions, use the steps in how to create a virtual lab activity.

Keep groups on task

One student hogs the device. Put the device in the middle of the table, not in anyone's hands, and rotate roles on a timer.

Groups change two things at once. Make the Skeptic sign off each change in the table. If a group gets an odd result, ask "What else changed?" before you explain anything.

Fast groups finish early. Give every task a stretch question on the card: a second variable, a harder value, or "predict, then test".

Slow groups never reach the discussion. Set a visible timer and say when to stop collecting. Half a table and a good discussion beats a full table and none.

You can't see what groups think. Read the answers as they arrive: View answers on each group's link. Visit the group with the oddest answer first.

Individual accountability. Group answers show what the group agreed. End with a one-question exit ticket on paper that each student answers alone.

FAQ

How many students should share one device?

Three or four. With more than four, someone always ends up watching. Pairs work well too, using Driver and Recorder roles only.

Should each student submit answers, or one per group?

One per group for the group task, so the group has to agree first. It also uses fewer of the free plan's 100 student answers per month. Check individual understanding with a short exit question.

Can different groups start with different settings?

Yes. Create one link per group and pin different starting values on each. Several links can share one question set, and All classes compares them group by group.

Do students need accounts for group work?

No. Each group opens its link, types a group name and first names, and answers below the simulation. Student pages set no cookies.

How long should a group simulation activity take?

Plan 15–20 minutes of group work and 10 minutes of sharing. A full jigsaw, with expert and home groups, fits in a 50-minute lesson.