L5 · Asking Good Questions; Interpreting Data Clinic
“I can tell a closed-list question from an open-ended question, write a good closed-list question, sort open-ended answers into categories - and use the mode to answer a statistical question with evidence.”
The statistical question is: "Which fruit should our school store sell?" Here are two ways to survey it. (a) Open-ended: "What fruit should the store sell? Write your answer: ________" Students can answer anything - apple, but also mango, kiwi, or "the red kind." The answers cannot be tallied until someone sorts them into categories. The surprise answers are valuable: they can reveal a favourite nobody thought of. (b) Closed-list: "Circle ONE: Apple / Banana / Orange / Grapes." The choices are given, so the answers are already in categories and can be tallied immediately. But a fruit that is not on the list can never win - if half the school loves mango, this survey will never show it. Say it plainly: neither kind is "the good one." Closed-list is fast and clean; open-ended can discover things - but its answers need categorizing work before they mean anything.
Why can closed-list answers be tallied right away?
Which of these is a closed-list question for finding out how students get to school? (a) "How do you get to school?" (b) "Circle one: Bus / Walk / Car / Bike" -> (b) - it provides a list of possible responses to choose from; (a) allows any response, so it is open-ended
"An open-ended pet survey collected these 8 answers: dog, puppy, cat, kitten, dog, cat, goldfish, dog. One way to categorize them is by animal, counting puppies as Dogs and kittens as Cats. Dogs = ___ (dog, puppy, dog, dog), Cats = ___, Fish = ___. Check: the categories total ___ answers." -> 4; 3; 1; 8
10 scored questions with a 3-rung hint ladder. Mastery needs 8/10 plus both gate questions.