statistics

4 Out of 5 Dentists Recommend Sugarless Gum!

Will you need to ask more questions to validate that, – or do you just believe it because “numbers don’t lie?”

  • How many dentists were surveyed?  Mathematically it could be as few as 5
  • Do they recommend it because it is less expensive?
  • Do they recommend it because it’s healthier?
  • Do they recommend it because they think it tastes better?
  • Were the dentists being paid by the Gum Lobby?

We are a stat-happy society – we always have been but today it’s a little different.  Like so many other things, technology has changed – or even eliminated – the confidence we have when we hear numbers. Even though we know this, we still have a vestigial affinity to believe things when they are attached to a spreadsheet or pie chart.  It is not uncommon to ask the same question to two different sources and come up with different answers

More often than not, it’s not really the statistics that are different but the underlying questions upon which the data was compiled before it was calculated as a statistic.  Here’s a simple example from the Real Estate world.

You’d like to know whether home prices in your neighborhood are dropping, staying flat or rising.  Simple question.  So you go to one of the Real Estate websites and do some searches until you can narrow it down to the geographic area you want.   You select “Sale Prices for April and May”  The computer then asks if you want Median or Average price.   You’re not sure which to pick.  You figure that they can’t be too much different and your just trying to get a rough idea.   So you choose Average.    The computer spits out a chart listing all the sales in each month and the average selling price. There were 5 sales in each month so it’s going to be a good comparison for you.

You look at average sale price for April and you look it for May.  The prices went up by $15,000.  You’re good now and go onto something else.

A week or so later one of your neighbors tells you that he was thinking of selling his house but he’s waiting because prices are flat – he wants to wait until they go up a little.  You said that you had found that prices took a good size jump in the last couple of months.  He said he got his information from a Realtor that he knows and he was assured that prices are flat.

So – two different conclusions from the same data and they are both correct.   The difference is that you asked for  Average Price and the Realtor gave your neighbor Median Price.  Without getting too mathy, here’s the difference.

  • The average of a set of numbers is the sum of the numbers divided by the quantity of the numbers.
  • The median of a set of numbers is the number that falls exactly in the middle of the group – it has the same amount of numbers above it and below it.

Depending on what you want to study, you would look at one or the other or both.  Here’s what those results look like

Average Vs Median

Average Vs Median

You can see that the Average Price rose from 127  to 142  while the Median Price remained exactly the same.  Both of those numbers are accurate.  They just answer different questions.

So it doesn’t matter what you’re discussing, if you want to be able to throw “facts” or “statistics” around to prove a point, keep in mind that if you don’t know the underlying methodology, your argument is basically worthless and doesn’t prove a thing.

“Lies, Damned Lies and Statistics.” There’s a 79% Chance That’s a Mark Twain Quote.

by Rick time to read: 3 min
0