Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: Typical!



Bernard Cleyet wrote:
Actually three observations, minimum.

One the experimenters.
Two the photographer.
Three the web person

And likely more.

Let's not play word games. When I asked for more than one
observation, I wanted observations of the _distribution_,
i.e. multiple events separately drawn from the ensemble that
is supposedly being typified. Three reports of the *same*
event adds nothing to the statistical significance of the
event.

The null hypothesis in this case is M/F = 50/50 ratio.

You can't conclude that a coin is biassed based on one toss.

You can't say this event typifies the KSU data if you don't
have a statistically-significant amount of KSU data.

If this one event typifies your preconceived notions, it tells
us about *your* prejudices ... not KSU's prejudices.