# Talk:Bernoulli trial

WikiProject Statistics (Rated Start-class, Mid-importance)

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

Start  This article has been rated as Start-Class on the quality scale.
Mid  This article has been rated as Mid-importance on the importance scale.

WikiProject Mathematics (Rated Start-class, Low-importance)
This article is within the scope of WikiProject Mathematics, a collaborative effort to improve the coverage of Mathematics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
Mathematics rating:
 Start Class
 Low Importance
Field:  Probability and statistics

## 'Bernoulli trial' with four possible outcomes

How would you call a 'Bernoulli trial' whose outcome is random and can be either of four possible outcomes, say "A1", "A2", "B1" and "B2". For example, a one-ball draw from an urn, in which you have the following balls: A1 = red/round, A1 = red/square, B1 = blue/round and B2 = blue/square.

## Are a person's eyes green?

"Are a person's eyes green?": This could arguably have more than two possible outcomes. I mean, my eyes have some green in them, yes, but they're not "green." Thoughts, anyone?

The question must not only be of the "yes-or-no-type", the chance (ok, Probability) for "yes" must be 50%, and the one for no must also be 50%. You could ask e.g. "Does the coin show heads or tails?" " Does the dice show 1,2,3 or 4,5,6?" "Is Schroedinger´s cat dead or alive?"... Christian, Germany 77.1.244.212 (talk) 21:37, 26 May 2010 (UTC)

(Adding to the above comment:) How about this (as a physicist friend pointed out to me): If you flip a coin 10 times in the exact same way, you are guaranteed to get 10 identical outcomes. This article, though obviously well-intended, is over-simplified. Shouldn't Bayesian probability be mentioned somewhere? --Smithfarm 16:05, 28 January 2006 (UTC)

Under the section "Mathematics", the expectation value of a Bernoulli trial should be 0.5, right? As far as I understand, if the expectation value were to be one, the proverbial flipped coin would always land heads up... Isaac, USA 14:25, 21 October 2010 (UTC) —Preceding unsigned comment added by 130.64.84.181 (talk)

No. That's an ideal coin or fair coin. The expectation of one side is p, the other is q and p + q = 1.  Randall Bart   Talk  17:45, 25 April 2013 (UTC)

## Was the newborn child a girl?

Depends on your definition of "girl" :) 88.111.132.121 (talk) 12:21, 26 February 2013 (UTC)

No it doesn't. It only requires that you have a firm definition of what is or isn't a girl.  Randall Bart   Talk  17:45, 25 April 2013 (UTC)

## Binomial trial

I added a redirect from binomial trial. I found a reference at answers.com, but it is mainly because of a story my father told me more than once.

In the 1960s my father worked for North American Rockwell. He was a mathematician working on reliability testing. He wrote a paper called the Bart Method, which at the time was a useful shortcut of an onerous calculation, but now we have computers to do the onerous calculations.

One day all the mathematicians were called to a seminar taught by a highly distinguished mathematician from England. He was talking about probability theory and statistics, and he kept talking about Bernoulli trials. After the speaker had used the term "Bernoulli trial" for the hundredth time, one man timidly raised his hand and asked "What are these Bernoulli trials you keep talking about?" The speaker paused then said "I supposed you could call them binomial trials." The entire room exhaled as the mystery was revealed. In the USA they were called binomial trials, and none of these American educated men knew the European term. This distinguished guest had spoken for an hour and none of his students knew what he was talking about.

The next generation of Americans grew up less cloistered, and the term "binomial trial" in now archaic, but we need a redirect in case anyone is reading a 50 year old US textbook.  Randall Bart   Talk  17:45, 25 April 2013 (UTC)

## Note on Poisson Trials

Section 'Definition': What about saying, "When multiple Bernoulli trials are performed, each with its own probability of success, these are sometimes referred to as Poisson trials" ? The 'own' sounds clearer. — Preceding unsigned comment added by Anant.sogani (talkcontribs) 13:21, 28 October 2013 (UTC)