Happy Numbers

Send to a friend

Send this article to a friend.

But other truths lie hidden in the statistics

October 6, 2012 - Statistics don’t lie, but sometimes they can be confusing.

Check this story that appeared Friday on VOCM, under the headline “Big Drop in NL Unemployment Rate”:  

“The unemployment rate in Newfoundland and Labrador has dropped by almost a half percentage point. The rate, which stood at 12.7 in August, dropped to 12.3 in September. Nationally, the domestic economy is coming off one of the strongest job creation months of the year.”

While statistics may not lie, they can certainly deceive. If you look at the Statistics Canada chart from which this news item was generated, you see that the news for Newfoundland and Labrador isn’t nearly as happy. Here’s the link:


In actual fact, the Newfoundland and Labrador labour force shrunk by 1,400 and the number of jobs fell by 200. And because the pool of labour contracted more than the number of jobs, the unemployment rate actually appears to go down. This is not cause for celebration and certainly not a “big drop” in the unemployment rate.

The story quickly moves to the national figures, where they use the same math – in reverse – to describe the situation. Here’s the rest of the story, copied verbatim from the VOCM site:

“A surprise addition of more than 52,000 new jobs last month was five times what economists predicted. The gain was the third biggest of the year and was way better (than) the 10,000 jobs analysts were expecting. It wasn't enough to put a dent in the unemployment rate, which actually edged up one tenth of a point to 7.4 per cent. That's because thousands of Canadians found work in September, almost 73,000 joining the labour force.”

In other words, the number of jobs grew by 52,000 while the number of job seekers grew by even more. Outside of Atlantic Canada, many provinces experienced huge growth in their labour pools, resulting in a gain of 72,700 people nationally looking for work. This is obviously higher than the new job creation numbers and accounts for the rise in the national unemployment rate. VOCM used those raw numbers to explain away the national unemployment rate, but ignored the very same – yet opposite trending – factors that are driving the provincial rate.

To put it another way, they used background data to identify good news in the national figures, but ignored the same data to put a positive spin on provincial numbers.

I would dismiss this as just not reading the chart properly – I despise charts too – but they seem to have fully understood the factors in the national instance. Why not the provincial numbers too?

So what’s up? An honest error committed in “cut and paste”? Or is VOCM, known for touting “good news stories,” trying to make a silk purse with a sow’s ear?

  • 1
  • 2
  • 3
  • 4
  • 5

Thanks for voting!

Top of page



Recent comments

  • Geoff Meeker
    October 07, 2012 - 12:34

    Nope. If you look at the fine print, here’s what StatsCan says: "Unless otherwise stated, this release presents seasonally adjusted data, which facilitates comparisons by removing the effects of seasonal variations." Either way, the media outlet is still covering the data in contradictory ways. That’s my point.

    • John
      October 09, 2012 - 12:56

      The whole thing about jobs is a joke in NL. Ask any recent MUN graduate what they are doing now, that is key. I was back there two years ago for the first time in a number of years and even in St. John's, things were not pretty. Optimism among youth seeking work was about the same when I lived there, ZERO. A quick look at internet job posts showed me the same old minumum wage BS that was there 20 years ago. Thank God I moved to western Canada where the real action is.

  • Don
    October 07, 2012 - 03:41

    I checked the stats. Thousands of part time jobs disappeared and almost the same number of full time jobs were created. So what really happened was Newfoundland and Labrador has 3800 more full time workers and the kids are back in school. You are basing your argument on the number of workers but ignoring whether they are full or part time. The whole article is based on nothing.