# Thread: A big enough sample size ...

1. ## A big enough sample size ...

Just off on a tangent here. I'm aware of the issues of risk assesment, business relevance, etc. when cmoming up with test cases.

Is there a quick shorthand method for determining if a sample size of a larger population is “statistically significant”? I haven’t opened a stats book since circa 1991. ;-)

I’m not looking for a very rigorous academic treatment, just a quick ballpark method, if one exists.

Thanks!

2. ## Re: A big enough sample size ...

In manufacturing we boiled it down to:
100 or less, 9 maximum (3 each from the beginning, middle, and end)
500 or less, 15 maximum (3 each from every 100)
1000 or less, 15 maximum (3 each from every 200)
3000 or less, 18 maximum (3 each from every 500)
Above 3000, 3 each from every 1000.

This was partly statistics, partly nature of the process. Getting a machine tuned in took 100 parts sometimes...running thousands, we only had to worry about slop as the machine warmed up after the initial set up.

3. ## Re: A big enough sample size ...

When I took statistics from this phenomenal man 15 or so years ago he said the number was 31 for any large quantity. I don't recall if or where he said the number came from but it was 31.

Go figure.

MArk B.

4. ## Re: A big enough sample size ...

Kind of competition to 42...?

Regards,
Juergen

5. ## Re: A big enough sample size ...

Originally posted by MArk B.:
When I took statistics from this phenomenal man 15 or so years ago he said the number was 31 for any large quantity. I don't recall if or where he said the number came from but it was 31.

Go figure.

MArk B.
<font size="2" face="Verdana, Arial, Helvetica">Now that you mention it Mark, I once saw that number mentioned too, but I have no idea where it came from or the thinking behind it.

Chris

6. ## Re: A big enough sample size ...

Originally posted by SteveC:
In manufacturing we boiled it down to:
100 or less, 9 maximum (3 each from the beginning, middle, and end)
500 or less, 15 maximum (3 each from every 100)
1000 or less, 15 maximum (3 each from every 200)
3000 or less, 18 maximum (3 each from every 500)
Above 3000, 3 each from every 1000.

This was partly statistics, partly nature of the process. Getting a machine tuned in took 100 parts sometimes...running thousands, we only had to worry about slop as the machine warmed up after the initial set up.
<font size="2" face="Verdana, Arial, Helvetica">Thanks! ;-)

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•

vBulletin Optimisation provided by vB Optimise v2.6.0 Beta 4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.