[...] This is a cross post from Talking Philosophy. [...]

Jeremy,
I took your arithmetic quizzes. i made a mistake on the first question of the second quiz. After correcting the answer I was told the test was ended. I never got to do the remaining 9 problems. Is this what you wanted?
Also, about the Monty Hall problem, if you remember, I suggested a solution which you were kind enough to say convinced you as being correct, although you wondered if it could be considered a proof. Although I’m a mathematician, probability has always been sort of a mystery to me. When I taught it, I just knew enough to keep my students amused and bemused. I don’t really know what would constitute a proof in a problem such as the MH one, or indeed, many other probability problems.
The major difficulty people have with the MH problem is that they are sure that when the person must make his/her final door choice, what preceded that moment is inconsequential. If you can their conviction in this, they have a chance of understanding why the solution is what it is, viz. 2/3.
I’ve come up with what I think goes a long way in helping. Instead of 3 doors say there are a million. After the first door is chosen, do away with all but the chosen door and one other. It should be far clearer in this case to make the switch.

Thanks Ralph.

No, the quiz shouldn’t have ended. Did you hit the “Complete!” button or press Return on your keyboard?

Actually, it shouldn’t have allowed you to end the test until you’d completed all the questions. I’ll see if I can replicate the problem.

The million doors thing is the standard way of illustrating the problem. I’m impressed that you came up with it independently of all the existing material! :)

Ooh I do like mental arithmetic. The competitive streak in me wants to know average times though, any such information available?

This might not be relevant but on the second test the auto-complete came up with the answers to the first, which was a little off-putting. Obviously this can be turned off but people might not think to and I’m not sure if it will skew results maybe?

Hi Dave

I’ll probably post the averages here early next week.

The raw data is skewed by a small percentage of people who have taken much, much longer than everybody else to complete each test (i.e., 6 to 7 minutes per test!).

The auto-complete thing is… well annoying. I’m not sure I can turn that off client-side. I’ll probably warm people about it. Thanks for pointing it out!

One thing I noticed myself is that during the first test I was warming up, so in the second test I was much quicker since I had already done some math.

I do not know if the test appear in random order or not. But I think they should, since other people might experience the same thing; or might get tired after the first test.

Thanks Mats. Yes, that kind of complication is known as an ‘order effect’. I am contemplating how to deal with it!

Also, in the second there was an element of ‘the race is on’ which the arning about the clock running right away execerbates.

The tests are about the same difficulty. I just did them at half past midnight so I was a little slower than usual, but they were relatively easy. If I were less tired, they would have taken about half the time.

I was able to just hit complete without entering any answers and it finished the test.

i don’t know will it help or not, but i know that some meditation technics can help

The fear of death how interesting.

When will the fear of death test be complete? Seems fascinating.