I wasn't talking about the distribution per se, but those that make up the
distribution, but the distribution isn't normal either.

;)

As for sample size, it really depends on whom you're talking to and what
you're doing with the data. For instance a single case design only uses 1
subject (typically) and a hell of a lot of observations. A factor analysis
(Principle Axes, Alpha or Maximum Likelihood) most prefer having at least
500 to 1000. Getting into causal modeling/structural analysis can require
tens of thousands. It all depends.

larry

> -----Original Message-----
> From: Won Lee [mailto:[EMAIL PROTECTED]
> Sent: Monday, June 14, 2004 11:36 AM
> To: CF-Community
> Subject: RE: age, politics, and the numbers
>
>
> Yes you are correct.  It is not a normal distribution.  I
> guess the sample
> size is too small.
>
>
>
>
>
>
[Todays Threads] [This Message] [Subscription] [Fast Unsubscribe] [User Settings]

Reply via email to