Re: Elementary Statistics Question

The exact answer uses the Binomial distribution; this is the approach 
that Wolfgang Kouker has used.  The stated problem uses a Normal 
approximation.  For large n, the Binomial distribution approaches the 
Normal distribution with the same mean and variance as the approximated 
Binomial distribution.  Simply sandardardize the variable
   [x-mean]/SD
to get the corresponding value for standard normal approximation.

To reproduce their value exactly, you need to think a little about 
continuous approximation to a discrete variable.  Since fractional 
children are frowned upon, >945 is equivalent to >= 946, so how should 
you allocate the probability of 945<x<946 in the non-standardized Normal 
approximation?

This should get you going, but leaves enough detail to require useful 
thought in case you are the student instead of the tutor.

Jerry

Michael wrote:

> I am trying to tutor a student on a basic Finite Math course.  We have
> a problem that goes as follows:
>
> Assume that 80% of all children who are exposed to chicken pox
> contract the disease.  If 1225 children are exposed to chicken pox,
> the probability that more than 945 of them will contract the disease
> is approximately the area under the normal curve ..... and the answer
> is given as: to the right of -2.46.
>
> I have looked at this from several different angles in order to
> explain it to this student but I just don't see how it can be done. 
> Seems like I would need sigma and mu in order to get a z score. 
> Anyone have any insight here?
>
> Thanks,
> Michael