I'm sorry that statistics course are in a 
confusing state of transition between the
two types of tables and the use of the
technology, and that you are a victim of
circumstances. 

It depends on what textbook you are using and whether 
your teacher is old-fashioned or modern.  Different 
textbooks have different kinds of normal distribution 
tables.  Some teachers require you to use tables and 
others allow you to use a graphing calculator such as 
the TI-83 or TI-84.

I'll try to cover all three methods below:

The answer by calculator is .0912112819. The
answer by tables is .0918. The table is not
as accurate as the calculator.

Here's how to find that on a TI-83 or TI-84.

normalcdf(36000,1E99,32000,3000)

Then press ENTER. You will get the answer
.0912112819.

Here's how to do that on your TI calculator:

To get:

normalcdf(

1. press 2ND 
2. press VARS
3. press 2

There is a key for the comma just above the
7 key. To get E (for exponent) press 2ND and 
then the comma key.

If you are required to use the tables, then 
you calculate the z-score for the left bound
x = 36000:

     x - m
z = -------
      s/Ön

where m = 32000, s = 3000, n = 1. (There is only one 
teacher in the sample.)

z-score for 36000 is calculated thusly:

     36000 - 32000
z = --------------- = 1.333333333 
        3000/Ö1
        
To use tables you will have to round that off
to 1.33 

Some tables have negative numbers and
and some don't.

If your table has negative z-scores,
then you will look up the area to the left
of 1.33 and find it to be .9082.

So you will subtract 1 - .9082 and get .0918

If your table does not have negative z-scores
then you ignore the sign, look up 1.33, and 
find .4082. Then you must subtract that from
.5000 and get .5000-.4082 = .0918.
 
Edwin