Testing materials, and testing with users
Design & Technology · GCSE · The School
Testing a material fairly
One variable at a time, the rest held constant. Comparing two glues means the same materials, the same area, the same clamping time, the same temperature, and the same way of pulling them apart. Repeat and take a mean, because one joint tells you about one joint. Record the failure mode too — whether the glue let go or the wood tore — because those are different results.
Ask people to do, not to judge
Give the user a task and stay quiet: "please open it and put the batteries in." Then watch. Where did they pause? What did they turn over? What did they try first that did not work? Silence is the hard part — the moment you explain, you have destroyed the evidence, because you cannot ship an explanation alongside the product.
The people you test on matter
Testing on classmates who watched you build it gives you the most generous users you will ever find. Test on somebody who has not seen it, and, if the product is for a specific person, on somebody like them: a jar opener for arthritic hands must be tried by hands that struggle, not by a sixteen-year-old who can open anything.
PRODUCT: a storage box for a primary classroom. TASK GIVEN: "put these away and get the box onto the shelf." NO PROMPTING. WHAT WAS OBSERVED, not asked: three of five children tried to lift it by the rim and it slipped; two put the lid on the wrong way round because it fits both ways but only latches one; every one of them pushed it onto the shelf lid-first. WHAT THEY SAID when asked: "it's good." WHAT THE OBSERVATION CHANGED: recessed handles on the long sides, an asymmetric lid that physically cannot go on backwards, and a chamfer on the leading edge so pushing it onto a shelf works. Three specific changes, and not one of them came from the answer to "do you like it" — which is why the question is not worth asking.