A new kind of online survey
28/11/2012, 13:52
Filed under: Culture, Ephemera, Politics, Statistics

Today the CBC reported that, based on a recent online survey carried about by Leger Marketing for the Association for Canadian Studies (ASC), as many as two-thirds of Quebecers view a flag hanging in the provincial legislature as a source of pride.*

The article goes on to claim,

The findings shine a light on public opinion in a province that has been sending mixed political messages lately: Quebec recently elected the pro-independence PQ — but only with a minority, and at a time when polls suggest support for independence is low, while the PQ’s sister-party in Ottawa, the Bloc, was nearly wiped off the map barely a year earlier.

Setting aside what I think about whether the polling and voting habits of Quebecers actually send mixed messages,** my first instinct upon reading this bold statement and “online survey” in the same article was righteous indignation: “This study shines no such light!” Continue reading

Comments Off on A new kind of online survey

Geeky Graffiti
26/09/2012, 10:49
Filed under: Ephemera, Myself, Statistics

I got stats on the mind. The first thing I thought when I passed this graffiti this morning en route to my favourite daily coffee spot was, “R-Squared!” It does not help that right now I’m working on a paper in which the main hypothesis test consists of a comparison between R^2.

Comments Off on Geeky Graffiti

Big Island in actu
24/09/2012, 16:03
Filed under: Ephemera, Statistics

Sometimes when I read news stories out of China, the numbers reported make me think of the Big levels from the old Super Mario gamesContinue reading

Comments Off on Big Island in actu

More Albertans need to save (or vote NDP?)
14/08/2012, 14:06
Filed under: Economics, Policy, Politics, Statistics

In theory, a booming economy that supports higher incomes can increase lifetime well-being, but only if people are smart. In my work comparing poverty outcomes among the provinces, Alberta and Quebec stand out as having a couple of the healthiest poverty profiles. However, for very different reasons.

While Quebec has a fairly robust social security system (the kind that many Canadians think they have, and used to have, but don’t really anymore), Alberta does not.

Alberta, however, has a booming economy and a small population. The result is a labour market that favours sellers—i.e. one that provides a job and high wages for just about every able body.

This state of affairs is good news for most Albertans. In fact, Albertans are much wealthier than Quebecers today. But, what will happen if the Alberta economy tanks?  Continue reading

Comments Off on More Albertans need to save (or vote NDP?)

I am not…
13/08/2012, 13:00
Filed under: Ephemera, Myself, Statistics

I am not the same person I was four years ago. Yesterday evening, I found myself skimming the conference proceedings for a recent Stata conference.

With presentation titles like, “Custom Stata commands for semi-automatic confidentiality screening of Statistics Canada data” and “Binary choice models with endogenous regressors,” I chuckled to myself: How arcane this list would seem to the uninitiated? What about myself, only a few years ago? It seems like only yesterday that I’d have never guessed that I would be among the purveyors of such esoteric nomenclature.  Continue reading

Comments Off on I am not…

Counterintuitive results
06/08/2012, 09:59
Filed under: Statistics, Thought

Here is a general rule of thumb for cross checking statistical results: “Are they surprising? If so, you probably made a mistake.”

The world isn’t that random, and common sense ways of knowing are remarkably good at arriving at more or less accurate estimates of aggregate level patterns (you don’t need a meteorologist to tell you that summer tends to be warmer than winter). If your results don’t conform to popular prejudices you need to put in some extra work.

In my experience, 9 out of 10 times that I am surprised by a result, I made a mistake. If you still think your results are correct and surprising after double checking everything, it’s time to estimate them again using either a different approach, or different data. If they still stand, then you might have something interesting on your hands.

Making sense of nonsense
13/06/2012, 22:15
Filed under: Ephemera, Myself, Sociology, Statistics, Thought

I’m as excited as all hell about Maxis‘ decision to develop the next SimCity, SimCity 5. The only thing that I’m a little worried about is that I’ll probably never have time to play it. I’m an adult now—work first, right.

But, I might have found a loophole. I’m pretty sure I can define work broadly enough (I’m a social scientist—this is a very slippery slope, but bear with me) to rationalize me getting my jollies following the game development.

Recently, the development team started talking openly about the simulation engine, and my soci senses are on the brink of orgasm. They are calling it “GlassBox.”

Continue reading

Comments Off on Making sense of nonsense