Question regarding the approximation to the normal distribution for W.

I am running a very small sample : 8 matched pairs, using the Wilcoxon Signed Rank Test and of course, anytime

one uses a small sample, below 10, there is the issue of not being able to calculate an accurate p-value. The approximation

to the normal distribution for W is used in such cases, but I am unclear on how one does that. Some refer to a simple table, constructed

on the exact distribution. Is there a quick and easy way to accomplish this on SPSS 14? I ran my data, and obtained a W of -0.421 but the normal

approximation of z, according to a supplied key for this specific problem says that the normal approximation of z is -2.38. Where am I going amiss on this one?