Lies, Damned Lies, and Statistics

My mom was a statistics junkie, her favourite quote was  “There are three kinds of lies: lies, damned lies, and statistics.” which was attributed to the British Prime Minister Benjamin Disraeli, by Mark Twain, who really made the quote popular in the United States.

And it popped into my mind earlier this week when I was on Twitter and came across this chart:

Screen Shot 2017-04-01 at 11.49.53 PM.png

Put out by Statistics Canada, it shows the rate of unemployment, in relation to attained education level.

At first blush it would appear that the lower the education level on the east the more fully employed the population is. This appears to decrease as you move west! Well except for Nova Scotia, they are always an anomaly though…

The unemployment rate is calculated using several indicators in Canada. They use a combination of household and business indicators. So employment status from household surveys (from self reporting) and business (number of jobs) using government payroll reports. This is meant to be used to calculate how many people are employed, and how many jobs there are.

Part of the problem with Statistics Canada’s monthly Labour Force Survey is that it relies on self reporting largely. This is often a snap shot, a cold call, which surveys the public, to answer questions. Randomly selected people are asked by strangers what their current employment situation is.

I don’t know about you, but I simply do not answer calls from numbers that I do not recognize, I can’t even remember the last 1-800 number that I have answered….. I also probably would pause if I was asked questions over the phone; the issue is that there are simply so many scams out there. This of course creates an instant bias. The sample is not really “random” any longer, this can skew the results either way. It is also one possible reason for the wild ups and downs in the numbers that you see.

You see this with poll results during elections, the polls show one party being a sure loss, and then Election Day rolls around and the result is entirely different than the polls predicted.

Really the behaviour of people over time has changed, they are less likely to answer the phone and even more unlikely to feel comfortable revealing personal information over the phone. Likely the long form census being brought back to replace the voluntary National Household Survey should assist in some areas for clarity.

The other problem is that Statistics Canada measures unemployment choosing a representative sample of Canadians, then asks them if they did anything during a four week period to look for a job. So if you are not actively looking, then you are not considered unemployed.

According to Stats Can, http://www.statcan.gc.ca/pub/71-543-g/2016001/part-partie2-eng.htm

The definition of unemployment is the following:

“Unemployed persons are those who, during the reference week:

  1. were without work but had looked for work in the past four weeks ending with the reference period and were available for work; or
  2. were on temporary layoff due to business conditions, with an expectation of recall, and were available for work; or
  3. were without work, had a job to start within four weeks from the reference period and were available for work.”

 

Their definition of “Not in the labour force”:

 “Persons who were neither employed nor unemployed during the reference period. This includes persons who, during the reference period, were either unable to work or unavailable for work. It also includes persons who were without work and who had neither looked for work in the past four weeks nor had a job to start within four weeks of the reference period.”

This creates a large segment of the population which would not be accounted for, those on social assistance, homeless, and working precariously. In order to fully understand the national true unemployment rate you would need to cross reference other data to understand how many people are truly unemployed.

In the graph above, it would appear that your best opportunity in PEI, NS and NL to find work would be if you had less than a high school diploma! That or there is NO ONE in PEI, NS and NL that does not have an education of High School or greater. Where the reality is more likely that those persons are “Not in the labour force”, they simply are not being taken into account.

E-Joe Friday Books!

A few days ago I picked up a copy of 1843 a new “culture, lifestyle and ideas” magazine, published bi-monthly by The Economist. I am an avid reader of The Economist, so I was excited to pick this new magazine up. Yes, it was the print one, I like my magazines, I like them in paper form, I like the shiny pages, the smell, the ads that are posted throughout that do not pop up, blink or make strange sounds, which you can only get from a print magazine.

And before anyone asks, I am under 40 (just…) and do not wear bifocals. I cannot be alone on this! I love technology and use it everyday for business, I consider myself tech savvy. But there is this joy which you get from walking into a bookstore and walking out with something tangible, that is not data in a cloud.

So fast forward as I am reading the magazine on page 124, the deputy editor Jonathan Beckman had the most interesting article covering what can only be referred to as speed reading apps, crossed with Coles Notes. Sort of Joe Friday, “just the facts ma’am” snippet apps. A Readers Digest version of longer books that someone does not have the time or will to read and would like to have a shortened summary version of the work.

My first thought was, oh where were you my whole life in school, then I realized I had that… It was called Coles Notes, abstracts, and for law, headnotes….Except they were in print, you had to buy them, or go find them in a library. But, either way they were a lifesaver in the research phase of something. They could not be your stand alone source, but they made digesting and framing large amounts of information easier, they allowed you to sort out what you needed and did not need to read.

Jonathan covered several of the apps available but the one that stood out to me was Joosr. Joosr promises to allow you to read the ‘most inspiring’ non fiction titles in under 20 mins. They have a library of over 250 titles.  Essentially Joosr provides summaries of books, summarizing the concepts of the longer book for people to enjoy.

This has left me with several questions, can you really glean the knowledge of the book from someone else’s idea? Often in non fiction, what you get from the book is constructed from the sum of your life’s experience, your perspective and attitude. You apply that to the information you are reading and synthesize that information from your perspective.

One of the shortened books is “The Seven Principles for Making Marriage Work”, John Gottman, I chuckled out loud at this particular title. I can picture it “Honey I am home, I listened to the shortened version of this on the way home tonight during my commute, I think I have this down pat now”, oy nothing could go wrong here.

These types of book are the ones I wonder about. There would be a natural bias from the person who is doing the summarizing, It’s not intentional. The person summarizing is reading the book, then applying their perspective based on their life experiences. What do you cut?  Something has to be shortened. How do you interpret and summarize that concept? If 10 people were to read the same book cover to cover, they would each likely have different parts that resonated to them. No two people experience the same book the exact same way.

What do you think?

For me, I know that I although I love to surf the web and read a good article or blog, and that I enjoy using apps like GoodReader to digest and annotate PDF’s; I still much prefer my print book or magazine, a cup of coffee and a fire to read beside, then an electronic version. Maybe I am old-fashioned…

reading a book.jpeg

There is nothing like a good book on a rainy day!

 

Language In The Media

Point of viewHow media frames the message is an interesting perspective for analysis.  The choice of the words used to report news has an incredible ability to influence the way that an issue is perceived. This of course, should not be a surprise, as Marshall McLuan observed over fifty years ago in his seminal work Understanding Media: “the Medium is the message”.

Take for example this article: POLL: 25% of Canadians Want Travel Ban Similar To What Trump Proposed.  Imagine if the same article was titled “75% of Canadians do not want a travel ban similar to what Trump proposed”?  That likely would not have the same click-through rate…

The article reports several other figures and never really discusses the travel ban issue again. But it sure opened a debate, most commenters focusing on how they thought the number must be ‘higher’. “The Beaverton”, a satirical website picked up the story, also pointing out the 75%, Poll suggests 75% of Canadians sick of other 25% always embarrassing them

Here is another example from USA Today: One in five say women are inferior to men, global study reports

What if it said: “4 out of 5 respondents believe Men and Women are equal”?  The sample consisted of 17,551 adults from 24 countries from around the world, given that information it is even more remarkable that 4/5th in a “Global study” responded the way they did. The method for choosing the sample, how the information was gathered, and whether the sample was representative was not covered. If you go to the Ipsos Site and head down to the technical section you see that:

Technical note

  • In total 17,551 adults aged 18-64 in the US and Canada, and age 16-64 in all other countries, were interviewed between January 20th – February 3rd, 2017. The survey was conducted in 24 countries around the world via the Ipsos Online Panel system (Argentina, Australia, Belgium, Brazil, Canada, China, France, Great Britain, Germany, Hungary, India, Italy, Japan, Mexico, Peru, Poland, Russia, Serbia, South Africa, South Korea, Spain, Sweden, Turkey and the United States of America). Approximately 1000+ individuals participated in each country with the exception of Argentina, Belgium, Hungary, India, Mexico, Peru, Poland, Russia, Serbia, South Africa, South Korea, Sweden and Turkey, where each have a sample approximately 500+. The precision of Ipsos online polls are calculated using a credibility interval with a poll of 1,000 accurate to +/- 3.5 percentage points and of 500 accurate to +/- 5.0 percentage points. For more information on the Ipsos use of credibility intervals, please visit the Ipsos website.
  • Data are weighted to match the profile of the population. In countries where internet penetration is approximately 60% or higher the data output generally reflects the overall population. 16 of the 24 countries surveyed generate nationally representative samples in their countries (Argentina, Australia, Belgium, Canada, France, Germany, Great Britain, Hungary, Italy, Japan, Poland, Serbia, South Korea, Spain, Sweden, and United States). Brazil, China, India, Mexico, Peru, Russia, South Africa and Turkey produce a national sample that is more urban & educated, and with higher incomes than their fellow citizens.”

Although the data was weighted to match the population, by virtue of gathering the data through an online platform, the results in many of the countries would have an artefactual bias as a result of over representation of the wealthier and more highly educated segment of the population.

As you read through the article though you see the challenge with the headline, it cites both country specific attitudes and many other statistics which point to greater systemic issues, but from the headline you would assume that the issue of inequality and gender equity was globally much better than ever before.

The travel ban article of course came out during the heat of the Trump administrations court battles and the USA Today article came out on International Women’s Day.   As Amin Maalouf wrote in The First Century After Beatrice: “People sometimes imagine that just because they have access to so many newspapers, radio and TV channels, they will get an infinity of different opinions. Then they discover that things are just the opposite: the power of these loudspeakers only amplifies the opinion prevalent at a certain time, to the point where it covers any other opinion.”

Welcome to Juxtaposed Thoughts

Welcome to Juxtaposed Thoughts! This blog covers, news, politics, international relations, governance and current events.

Analyzing news articles, current events and international debates, this blog strives to provide impartial analysis of events that shape our political landscape.

Respectful debate is encouraged. Please feel free to comment or email the author at comments@juxtaposedthoughts.com