You are not a statistically valid sample
Show of hands. How many of you have been in a meeting where someone presents an idea and someone else says “That won’t work”, or even “I don’t think that will work”? I see lots hands raised. Me too.
I took statistics in school and honestly, I don’t remember a lot of it (I do however remember the Mark Twain quote about lies and statistics). I also remember the textbook chapter on sample size. What struck me the most about sample size is that the sample size needed to get a 95% confidence rate with a confidence interval of say 5 (aka margin of error) with a population of 10,000 would be 370. Don’t believe me? Do the math yourself here. But, what about a population of 100,000? You’d need a sample of just 383, only 13 more despite a ten-fold increase in the population. Crazy, huh?
The point here is that the smaller the population the greater percentage of that population you’d need to have statistically valid sample.
Back to the meeting. Say you are in a meeting with 9 other people and using the parameters I laid out above, you’d need a sample size of…10. That’s right. Every. Single. Person.
Therefore when you’re sitting in that meeting and that certain person, aka Mr. Buzzkill, says “That won’t work!” remind him or her that they aren’t a statistically valid sample and if they are going to make a claim like that, they probably should have the data to back it up.
Now, just between us, if you happen to be that person, slow down. While your opinion certainly warrants consideration, there are others in the room that might have different opinions and they might like to be heard too.
Stay tuned for my next blog on the Binomial Theorem.
About the Author
Geoff Chinnock is a consultant with Morrison, working primarily in our Business & Accounting Advisory practice. To get in touch with Geoff, please find contact information for Morrison here.