It’s not that extracting information from your kids is hard; but parents know that extracting high quality, useful information is almost impossible. My offspring can talk continuously1, but when it comes to a direct request for information the clam up, like clams with dimples. Their vocabulary is reduced a repetition of”yes” or “no”. How can I know if my questions are generating reliable information? How can I know if “no” really means “no” when I ask if anyone needs to use the potty2?
Me: Did you eat lunch?
The Frogger: Yes.
Me: What did you have for lunch?
The Frogger: Yes.
Me: Did you have a sandwich?
The Frogger: Yes.
Me: Did you have a hot dog?
The Frogger: Yes.
Me: Did you have pizza?
The Frogger: Yes.
I don’t remember what is on the school lunch menu everyday, but I do know that they do not try to give my child a sandwich and a hot dog and pizza all on the same day. This suggests that I cannot trust our question and answer system to
One way I can test the reliability of my system, is to try a question where I know the answer is “no”:
Me: Did you have deep fried elk?
The Frogger: Yes.
It is also useful to try a question where I know the answer is “yes”:
Me: Do you want a lollipop?
The Frogger: Oui.
I now use these control questions at the beginning of such conversations with my children to determine whether we are in an informative state of mind or not. I also find that eliciting a “yes” and a “no” right at the start calibrates the conversation to allow more accurate responses (hypothesis: children need to be reminded that both words exist).
Scientists use the same concept in the lab when running tests. You want to know the answer. In order to do that, you need to know that your test works. How do you do that? In addition to your experimental samples, you add a sample that you know should work (eg, lollipops) and you add a sample that you know should not work (eg, deep fried elk). If you get a positive result for the sample that shouldn’t work, you know your test is borked. If you get a negative result for the sample that should work, you know your test is borked. Such results mean that you need to improve your test and try again. We call this process graduate school.
NOTES
1. I suspect that, due to small body size, they can efficiently oxygenate their internal organs by diffusion through the skin2.
2. From a practical, parenting perspective, the answer is always “yes”. You may do the Ghostbusters line now if you wish.
The Finch and Pea 