• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

My Zener card experiment

Seismosaurus

Philosopher
Joined
Mar 15, 2003
Messages
6,092
Next September my school is having a day devoted to the topic of "Viewpoint". I've convinced the science department to do how we view pseudoscience as part of our contribution!

This is for 12 year old kids...

I'm kicking the lesson off with some clips of science fiction and fantasy where people use magic powers - Yoda lifting Luke's X Wing, that kind of thing. That will be used to spark a discussion of what might be possible in the way of mental powers. I might also try to include a clip of somebody like Gellar spoonbending and Randi debunking him if I can find it, copyright allowing.

Then the experiment. We're having kids in groups of three; a sender, receiver and judge. The sender and receiver will be sitting at opposite ends of a table; the receiver will be facing away from the sender. The judge will sit at the side of the table.

The sender will have a set of 20 Zener cards, laminated, and will have thirty seconds to try and send each one to the receiver; the judge will time this with a stopwatch.

After thirty seconds the receiver will mark down what he or she thinks is being sent; the judge will then record the right answer. After all twenty are done, the judge will collect the form from the receiver and mark it using his own to give a score out of twenty.

Afterward the kids will do a bar chart of their scores, and we will do a chart of how many kids got each score so that hopefully we will get a nice bell curve.

At the end we will cap it off with a discussion of how doing tests like this can change our viewpoint on pseudoscientific topics.

Does this sound like a fair test? I've tried to make it fairly cheat-proof without going over the top - bearing in mind that these are 12 year olds, and that they won't have any prior knowledge of what it is they are doing, and so no way to plan cheating in advance.

Also, my probability math is not really up to this... I know you have a one in five chance of being right on any individual go, but what is the probability of getting each score? So what are the odds of getting one of the twenty right, two right, three, etc?
 
Next September my school is having a day devoted to the topic of "Viewpoint". I've convinced the science department to do how we view pseudoscience as part of our contribution!

This is for 12 year old kids...

I'm kicking the lesson off with some clips of science fiction and fantasy where people use magic powers - Yoda lifting Luke's X Wing, that kind of thing. That will be used to spark a discussion of what might be possible in the way of mental powers. I might also try to include a clip of somebody like Gellar spoonbending and Randi debunking him if I can find it, copyright allowing.

Then the experiment. We're having kids in groups of three; a sender, receiver and judge. The sender and receiver will be sitting at opposite ends of a table; the receiver will be facing away from the sender. The judge will sit at the side of the table.

The sender will have a set of 20 Zener cards, laminated, and will have thirty seconds to try and send each one to the receiver; the judge will time this with a stopwatch.

After thirty seconds the receiver will mark down what he or she thinks is being sent; the judge will then record the right answer. After all twenty are done, the judge will collect the form from the receiver and mark it using his own to give a score out of twenty.

Afterward the kids will do a bar chart of their scores, and we will do a chart of how many kids got each score so that hopefully we will get a nice bell curve.

At the end we will cap it off with a discussion of how doing tests like this can change our viewpoint on pseudoscientific topics.

Does this sound like a fair test? I've tried to make it fairly cheat-proof without going over the top - bearing in mind that these are 12 year olds, and that they won't have any prior knowledge of what it is they are doing, and so no way to plan cheating in advance.

Also, my probability math is not really up to this... I know you have a one in five chance of being right on any individual go, but what is the probability of getting each score? So what are the odds of getting one of the twenty right, two right, three, etc?

How are you with combinatorics? The general formula is:

let N = number of trials (20)
let X = number of correct picks
let p = probability of an individual pick being correct (0.2)

For any number x from 0 to 20, the probability of getting exactly x correct is:

p(X=x) = combin(N,x) * (p)^x * (1-p)^(N-x)

So for 0 correct, this works out to be combin(20,0) * (0.2)^0 * (0.8)^20 = 1%
For 1 correct : 6%
2 correct : 14%
3 correct : 21%
4 correct : 22%
5 correct : 17%
6 correct : 11%
7 correct : 5%
8 correct : 2%
9 correct : 1%
10 or more correct : 0.3% (pretty darn tootin' small)

What I may add is a round 2 for your experiment. As in, take all the kids who got 8 or more correct. In typical woo terms, these people will "show signs of ESP". Have only these few kids take the test one more time and then compare the scores.

I'd love to know how the kids do. I love these quick and easy tests as practical examples. :)
 
Instead of confusing the kids with math, repeat the experiment where the only outcome is the result of chance as your control group. This could be done by having the receiver shuffle another deck of cards and compare the order of these cards with those of the senders. If you have enough trials, the two curves should appear identical.
 
Ooo, thanks for that Samedi, I got it into excel and ran the numbers myself. 1e-14 chance of getting all 20 right!

I might offer a reward for anybody who gets over 12, say... my own mini version of the million dollar prize!
 
And don't forget to collect your finders fee when you send the kid who gets all 20 right to take the MDC ;)

Arthur
 
12 year olds can be pretty tricky. Someone has to judge, correct? Oh, the ways one can cheat. *cackles in glee*
 
On the question of cheating...

...is it ethical or useful to have a "cheat group" who get amazing results and a discussion on why students think they get these results.

That seems a bit involved and away from what you're doing but if somehow you can introduce how you can ignore some data and get better results - the sort of thing woos have to do.
 
Last edited:

Back
Top Bottom