Huge z-values

What to do for “huge z-values”? They do sometimes turn up, especially when the evidence for a hypothesis test is very strong one way or the other. Anyway, just use a spreadsheet to find your p-values instead of the standard normal table. The following formulas work for finding p-values using a spreadsheet (and work for any size z-values actually, not just “huge” ones):

1) left tailed test, use =text{NORMSDIST}(z)

For example if z=-3.8

p= text{NORMSDIST}(-3.8)=0.000072348261104

2) right tailed test, use = 1.0-text{NORMSDIST}(z)

For example if z=4.4

p=1.0 - text{NORMSDIST}(4.4)=0.000005408458265

3) two tailed test, and z is negative, use = 2* text{NORMSDIST}(z)

For example if z = -4.8

p=2*text{NORMSDIST}(-4.8)=0.00000171710376

4) two tailed test and z is positive use = 2* text{NORMSDIST}(-z)

For example if  z = 5.1

p=2*text{NORMSDIST}(-5.1)=0.000001969336413

Of course any time the z-value is “huge” that means the p-value is going to be small.

And remember a very small p-value means very strong evidence against the null hypothesis H_0.

You still want to find the actual p-value though and state what it is when you do a hypothesis test even if you do know because the z-value is “huge” it is going to lead to very strong evidence against H_0.

Advertisement

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Connecting to %s


Follow

Get every new post delivered to your Inbox.