Flipped Classroom with Interactive Simulations

A flipped classroom with interactive simulations moves the first meeting with a new idea out of the lesson. Students explore at home, commit to predictions and take one or two measurements. You read their answers before class and spend the lesson on the hard part. This guide shows you how to design the home task, read the results in ten minutes, group students by what they actually think, and run the lesson that follows. It includes two complete flipped lessons, one in math and one in physics, with the settings to pin, the questions, the answers and the class plan.

Why a simulation works better than a video for the home part

Most flipped lessons send a video home. A video can be watched with half an eye, and it tells you nothing about what students understood. A simulation has to be used. Students set values, run the model and read results, so they arrive with an experience to talk about.

The bigger gain is evidence. When the home link carries a question set, you know before the lesson:

• who did the task, and when;
• what each student predicted before touching the simulation;
• whether they changed their mind after it;
• whether they could read a value correctly.

That is enough to plan the lesson around the class you actually have, not the class you imagined.

Be clear about the limits. A simulation shows a pattern. It does not explain it, and the home task shouldn't try to. The explanation, the vocabulary and the hard problems stay in class, where you can respond to what students say.

What goes home and what stays in class

Split the work by what needs you in the room.

At home (15–20 minutes)	In class (the whole lesson)
Meet the phenomenon	Name and explain it
Commit to two predictions	Discuss why predictions were wrong
Take one or two readings	Use the idea on harder problems
Write what they noticed	Practice, a real practical, an extension

The rule of thumb: home is for noticing, class is for understanding. If a home task needs a paragraph of explanation before students can start, it belongs in class.

Design the home task in five moves

1. Pick one pattern students can find in ten minutes. "Compound interest pulls away from simple interest" works. "Everything about sequences" doesn't.
2. Pin the link. On the simulation's Share page, set the screen and starting values, then create one link per class. Every student opens exactly the setup you planned. The step-by-step guide to collecting answers walks through each screen.
3. Ask two predictions and ask them again. Set two questions to "Before, as a prediction" and tick "Ask again after the simulation". The simulation stays locked until students commit, and the results show a Prediction and an After column side by side.
4. Add one or two number questions. They prove the student ran the simulation and read it correctly. Numbers are checked automatically against your answer and tolerance.
5. End with a "notice" question, not an "explain" question. "What did you notice about the gap between the bars?" gets honest observations. "Explain why compound interest grows faster" gets answers copied from a search engine.

Put the steps in Instructions for students and add one line for students who get stuck: "If something confuses you, write it in the last answer. That is useful too." Confusion you know about is worth more than a blank.

Set the deadline for the evening before the lesson, so you can read answers in the morning. The homework guide covers deadlines, turning links off and students without a device at home.

Read the results before the lesson

Ten minutes before class is enough. Open the Share page and click View answers on the class link.

1. Read the percentage correct under each question. It tells you whether to keep your planned introduction or skip straight to the hard part.
2. Compare the Prediction and After columns for each student. Sort the class into three groups:
   • Group A: right before and after. They are ready for the extension.
   • Group B: wrong before, right after. The simulation changed their mind. They can explain it to others, which fixes the idea for them too.
   • Group C: still wrong after, or a wrong number. They need you.
3. Note who has no answer. Those students do the home task in the first ten minutes of the lesson, on a phone or class device.
4. Read the "notice" answers and pick two or three to quote, without names.

If one class does much worse on a shared set, create a separate question set for that class in the Questions tab (with an easier first question) and choose it for that class's link on the Share Links tab. The other classes keep the original set.

Example 1: simple vs compound interest (math, ages 14–17)

Simulation: Arithmetic and geometric sequences – simple and compound interest. Pin on the link: Simple and compound interest mode, first term u₁ = 10 (the principal, in thousand USD), interest rate 8% per period, number of terms n = 30.

Instructions for students: "Answer the two predictions first. Then drag the k slider. The line under the graph shows both balances 'after … periods'. Find when each balance first reaches 20 thousand USD. About 15 minutes."

Question set:

1. Prediction, multiple choice, asked again: "You invest 10 thousand USD at 8% per period. After 20 periods, compound interest gives you … compared with simple interest." Options: about the same / about 10% more / about 80% more / more than twice as much. Answer: about 80% more. After 20 periods the balances are 46.61 and 26.00 thousand USD.
2. Prediction, number, asked again: "After how many periods does the compound balance first reach 20 thousand USD?" Answer: 10, tolerance 0. At 9 periods it is 19.99, just short, which makes a good talking point.
3. Number: "After how many periods does the simple-interest balance first reach 20 thousand USD?" Answer: 13, tolerance 0. It is 19.60 after 12 periods and 20.40 after 13.
4. Number: "After 10 periods, how much more is the compound balance than the simple one?" Answer: 3.59, tolerance ± 0.01, unit thousand USD (21.59 − 18.00).
5. Short answer, read by hand: "What did you notice about the gap between the two bars as the periods go on?"

The lesson (50 minutes):

• 5 minutes, whole class. Put the percentages for question 1 on the board, before and after. Quote two "notice" answers.
• 15 minutes, groups.
  • Group C works with you at the projector. Step through periods 0 to 2 together. Simple interest adds 0.8 each period: 10.8, then 11.6. Compound interest multiplies by 1.08: 10.8, then 11.66. The first gap appears in period 2, because the second period's interest is paid on 10.8, not 10.
  • Group B writes both formulas, A = 10(1 + 0.08n) and A = 10 × 1.08ⁿ, and checks each against three values from the simulation.
  • Group A tests the "rule of 72": change the rate in the starting values under the simulation to 6%, 9% and 12%, and compare the first doubling period with 72 ÷ rate. The simulation gives 12, 9 and 7 periods; the rule predicts 12, 8 and 6.
• 20 minutes, everyone. Mixed problems on paper, from "which account would you choose?" to finding a doubling time without the simulation.
• 10 minutes, exit. Students answer one new problem. For a stretch, compare loan repayment methods in amortized loans.

Example 2: charging a capacitor (physics, ages 16–18)

Simulation: Charging and discharging a capacitor – measuring the RC time constant. Pin on the link: source EMF 6 V, resistance 10 kΩ, capacitance 470 μF, theoretical curve on. With these values τ = RC = 4.7 s.

Instructions for students: "Answer the predictions. Press Charge (K → 1) and watch the voltmeter and the graph. The simulation marks the moment the voltage reaches 63% of the way to 6 V and measures τ. Then press Discharge (K → 2). Finally, change Resistance from 10 to 20 in the starting values under the simulation and charge again."

Question set:

1. Prediction, multiple choice, asked again: "You double the resistance. The capacitor will charge…" Options: twice as fast / at the same rate / twice as slowly / four times as slowly. Answer: twice as slowly. τ rises from 4.7 s to 9.4 s.
2. Prediction, multiple choice, asked again: "Just after you close the switch, the current is…" Options: zero, then it rises / largest, then it falls / the same all the time. Answer: largest, then it falls.
3. Number: "With 10 kΩ and 470 μF, what time constant does the simulation measure?" Answer: 4.7, tolerance ± 0.1, unit s.
4. Number: "What voltage marks the 63% level while charging?" Answer: 3.79, tolerance ± 0.02, unit V.
5. Short answer, read by hand: "What did you notice about how fast the voltage rises at the start and near the end?"

The lesson (50 minutes): start from question 2. A common wrong prediction is that the current builds up slowly, like water filling a pipe. Run the charge on the projector in present mode and point at the current reading. Then derive u = E(1 − e^(−t/τ)) with the class and check it against the measured 3.79 V at t = τ. Groups then design a circuit with τ = 10 s (for example 10 kΩ with 1,000 μF) and test it in the simulation. If you have the equipment, finish with a real capacitor and a data logger, and compare the real curve with the model.

Make it a routine

Flipping works best when students know the pattern. Use the same day, the same length and the same structure every time: predictions, a short task, numbers, one "notice" question.

• One link per class, one set for all. Several links can share a question set, and All classes on the Questions tab compares the classes.
• Count your answers. The free plan collects 100 student answers a month across all your links. Three classes of 28 use 84 with one flipped task, so a monthly flip fits the free plan and a weekly one needs Pro ($7 a month or $59 a year; see pricing).
• Grade completion, never predictions. A student who predicts wrong and changes their mind did the task well. Grading predictions teaches students to look things up first.
• Keep the privacy rules. Students type a first name or nickname, nothing else. They don't need an account, and student pages set no cookies.

Flip a short unit, not just one lesson

Once the routine works, chain two home tasks around two lessons. The second home task checks whether the first lesson worked.

1. Home task 1: the first meeting, as above.
2. Lesson 1: groups, explanation, practice.
3. Home task 2: the same simulation on a new screen or with new pinned values, and predictions that need the idea from lesson 1. Create a second link with its own question set.
4. Lesson 2: start from the home task 2 results, then move on to harder problems or a test.

In Example 1, home task 2 could pin the Geometric sequence mode with u₁ = 10 and q = 1.08 and ask: "Which term equals the compound balance after 10 periods?" The answer is u₁₁ = 21.59, the same number students met at home in task 1. Students who see that compound interest is a geometric sequence with q = 1 + r have understood lesson 1. Students who don't will show you in the results.

Mistakes that sink a flipped lesson

• A home task that tries to teach everything. Twenty minutes of exploration can't replace a lesson. Aim for one pattern.
• Teaching the home task again in class. If 85% of students got it, a ten-minute recap wastes their time. Go to the hard part and support the rest in Group C.
• Not reading the results. Without the morning check, a flipped lesson is just homework plus the usual lesson.
• No plan for students who didn't do it. Give them the link at the start of the lesson. Don't let them stall a group.

For the whole-class demo in present mode, see teaching with a projector. For more ways to use simulations in lessons, start with how to use interactive simulations in the classroom. If you want students to argue about a single prediction in class instead, use Predict–Observe–Explain.

FAQ

How long should the home part of a flipped lesson be?

Fifteen to twenty minutes. Long enough to meet the idea and answer five questions, short enough that most students actually do it.

What if students don't do the home task?

The results table shows who answered and when. Students without an answer do the task in the first ten minutes of the lesson on a phone or class device, then join a group.

Can several classes use the same flipped task?

Yes. Create one link per class with the same question set. Answers stay separate per link, and All classes compares them.

Do I need a paid plan for a flipped classroom?

No. Links, question sets and predictions work on the free plan, which collects 100 student answers a month. If you flip every week with several classes, Pro removes the limit.