Re: virus: alpha and beta errors

Eric Boyd (
Mon, 15 Mar 1999 14:59:46 -0500


From: KMO <>

Do you think the sentence "I have faith that I will intuitively know when to tip the scales and which way to time them" invokes "evidence of things unseen" any more than either of the other two sentences? That's not a rhetorical question.

Certainly. The word "faith" has connotations of much more dogmatic belief than either confidence or trust. That is why it is used in religious areas -- people don't just "trust" in the lord, they *have faith* in him, which is clearly something stronger and more immune to negation.

Why use a stronger word when the weaker one is more accurate?

Do you keep written records of your hunches and wether they panned out? Do you have "data" or just "impressions" based on your memory of a few experiences that stand out in your memory?

Well, my usual system two fold. I make constant guesses during class -- you know, when the prof asks those (semi) rhetorical questions. I always answer based on a quick intuition, and then, when the prof gives the answer, I have instant feedback into the effectivness of my intuition. Depending on how new the material is to me, my effectivness varies from perfect to one or two mistakes in an hour long lecture. The *really* important thing I do, is when I guess wrong, I *find out* why, and I learn the new model, so that next time, my intuition will be correct. I credit much of my acedemic success to this lecture learning strategy.

Another exercise I have used is on assignments is attempting to guess the range of the answer (the form of the answer is usually obvious), and seeing how I fair. I haven't done this often lately, becuase now, at the end of every question, I do a quick "does it seem reasonable" check, which essentially preforms the same service. (interestingly, this latter method actually puts stronger weight on the intuition than on the work done, since if it *doesn't* seem reasonable, I go back and check the work; often finding a mistake or two. This method has triumphed becuase the nature of the questions has changed -- they are now big long mathematical questions, with a high chance of making an error, but a very physical answer, which can be easily checked for reasonableness)

So, in final answer, no, I don't have collected data -- I do the analysis so soon after the data collection that saving the data is unnecessary.

I hope everyone realizes how much I have to overstate my thoughts on the matter of faith and the limits of rationality in order to maintain a sense of contention in this discussion.

I suspect you and I have very similar outlooks.