Coding robots for clubs and groups

Running a coding club with floor robots is easier than it looks, and the things that make it work are organisational rather than technical. Almost none of them are about the robots.

This is what tends to work, from people who run them.

Three children per robot

One child per robot sounds ideal and is worse. Groups of three produce discussion, disagreement and explanation, which is where most of the learning happens. Above four, somebody is standing back.

Three is the number almost every club settles on.

Card systems beat button systems for groups

When the program is a row of physical cards, three children can all see it, point at it and rearrange it. When it is a sequence of button presses, one child is holding the robot.

This single difference does more for group dynamics than anything else.

Start with a challenge, not a lesson

“Get it to the red square” produces more programming in ten minutes than an explanation of sequencing does in thirty. The concepts are learned in service of the goal rather than in advance of it.

Challenges also scale naturally — the same task can be made harder for a group that finishes early.

Challenges that need no preparation

Draw a square. Visit three points in a fixed order. Get out of a maze made from books. Deliver something to another group’s robot. Each of these works with any robot and any mat.

Keeping five in your head means a session never runs dry.

Predict before running

Ask each group to say where the robot will end up before pressing go. This is the single most effective teaching intervention available and it costs nothing. It turns running the program into a test of a prediction rather than a reveal.

It also makes debugging obvious, because the child already has an expectation to compare against.

Let it fail

The instinct to correct a program before it runs is strong and should be resisted. A robot doing the wrong thing is the most information-rich moment in the session.

Adults helping in a club are usually the biggest obstacle to this.

Have a non-robot task ready

Groups finish at different rates, and a group with nothing to do gets loud. Paper mazes to design for another group, or writing a program on paper before entering it, absorbs the gap productively.

Designing challenges for each other is usually the most popular part of a session.

Session length

Forty-five minutes to an hour is the practical range: fifteen minutes of setup and a challenge, half an hour of work, ten minutes of showing each other. Longer sessions do not produce more.

Ending while it is still going well is the right instinct.

Mixed ages in one room

Floor robots handle this better than almost anything else, because the same challenge can be met with a five-step program or a twenty-step one. Nobody has to be given a different task.

Grouping by age rather than mixing tends to work better for the discussion, though.

Storage and setup time

A box per robot containing its cards, mat and cable turns setup into handing out boxes. Anything shared from a central bin costs ten minutes at the start and ten at the end.

More on that in batteries, charging and classroom use.

Budget order

Robots first, then mats and cards, then spares, then anything else. A club with four robots and enough mats runs better than one with six robots and no mats.

Expansion accessories are the last thing to buy, not the first.

Where to look

Screen-free coding robots are the shelf built for this; app and remote robots and robot building kits suit older groups.

We are happy to talk through a set for a club — the number is on the contact page, and the shop is open seven days.

Running one with no technical background

A great many good coding clubs are run by people who do not program, and the robots are chosen precisely so that this works. The adult’s job is not to know the answer — it is to ask what the group thinks will happen, to insist the program is run rather than corrected, and to keep the challenges coming. None of that requires knowing anything about programming.

It is worth saying that out loud to anybody hesitating to start one. The children will find things the adult did not know the robot could do, and saying “I do not know, try it” is a better response than an explanation would have been. The technical knowledge that matters here is knowing where the spare cards are.

What does help is doing each challenge once yourself beforehand, not to have the answer ready but to know roughly how long it takes and where a group is likely to get stuck. Ten minutes of preparation is enough, and it is the difference between a session that fills the hour and one that runs out after twenty minutes.