‘Controlling’ for confounding variables graphically

As we’ve learned (ad nauseum) basing causal claims on a simple bivariate relationship is fraught with potential roadblocks. Even though there may be a strong, and statistically significant, relationship between an independent and dependent variable, if we haven’t controlled for potentially confounding variables, we can not state with any measure of confidence that the putative relationship between the IV and DV is causal. We should always statistically control for any (and all) potentially confounding variables.

Additionally, it is often desirable to dig deeper into the data and find out if the units-of-analysis are fundamentally different on the basis of some other variable. Below you may find two plots–each of which shows the relationship between margin of victory and electoral turnout (by electoral district) for the 2017 British Columbia provincial election. The first graph plots a simple bivariate relationship, while the second plot breaks that initial relationship down by political party (which party won the electoral district). It could conceivably be the case that the relationship between turnout and margin of victory varies across the values of political party. That is, the relationship may hold in those electoral districts where party A won, but not hold in those in which party B won.

We can see here that there is little evidence to suggest a difference in the relationship based on which party won the electoral district. Can you think of another `third’ variable that may cause the relationship between turnout and margin of victory to be systematically different across different values of that variable? What about rural-versus-urban electoral districts?

Here are the plots:

Advertisements

You may not be registered to vote even though you think you are.

I read a somewhat troubling story this morning about Canadian citizens who have previously not only been registered to vote, but who have voted, and are no longer registered with Elections Canada. Here is an excerpt:

Delaney Ryan is a 23-year-old anthropology student at Simon Fraser University. Unlike many of her contemporaries, she voted in both the last provincial and federal elections. She meets all the criteria for voter registration, having her driver’s licence and maintaining the same address since voting in those elections.

She can’t understand why, then, this time, she wasn’t registered on the voters list.

“I had heard rumours about people not being registered, and a friend’s Facebook page had postings on it of other people finding out they were suddenly no longer registered. A lot of these people seemed to be in the same demographic as me. So I went online (to the Elections Canada website) and checked, and I wasn’t on it, either.

“So I had to reapply for registration.”

So, go to the Elections Canada website and verify that you are registered to vote this October 19.

Political Psychology–How the gut, head, and heart decide our political views

David Moscrop of Maclean’s magazine, has been writing a series of articles on the psychology of politics. Why do we have the political opinions that we do? The answer, as you now know, is a combination of environmental, demographic, and personal characteristics. More and more, political scientists and psychologists have been researching the importance of psychology as a factor that influences political beliefs and behaviours.

moscrop-graphic

At the beginning of his article, Moscrop reveals the inspiration behind the series:

But the truth is that your gut is as much a source of your political decisions as your rational brain, and much of the time your gut—emotions, feelings, intuition—does its work outside of your awareness.

Yep. The faces of the candidates, the pitch of their voices, their gender and ethnicity and height; whether or not you believe in God, whether or not you’re hungry, whether you’re a lawyer or dock worker or school teacher; the effects of political advertising, the effects of issue framing or priming, the effects of your peer group; your partisanship, your family, your fears.

Here we see an image taken from the article, that demonstrates the various parts of our bodies that are involved in the making of political decisions.

Political Ideology and Political Parties

The Conservative Party of Canada website in the weeks leading up to the 2015 federal election.

The Conservative Party of Canada website in the weeks leading up to the 2015 federal election.

Today, we continue to look at Canada’s glorious Prime Minister, Stephen Harper. Besides being the current PM of Canada, Stephen Harper is also an honourable MP (member of parliament–we Canadians are forever grateful to the glorious denizens of Calgary Southwest*) and the leader of the federal Conservative Party.

On the short quiz that I gave on Tuesday, you were asked to write three words or phrases that are usually associated with conservatism, as a political ideology. Did any of you use the word “protect?”

Below you’ll find some screenshots of the scrolling images on the home page of the Conservative Party website. You’ll notice that the political campaign operatives have decided to portray PM Harper is our daddy, whose job it is to protect us in a dangerous world. Protect our jobs (presumably from being snatched away by foreigners); protect our children; protecting our economy, and protecting us from terrorists and others who would do us harm. Will it work?

“Fear makes man unwise in the three great departments of human conduct: his dealings with nature, his dealings with other men, and his dealings with himself. Until you have admitted your own fears to yourself, and have guarded yourself by a difficult effort of will against their myth-making power, you cannot hope to think truly about many matters of great importance … .”

—Bertrand Russell (“Outline of Intellectual Rubbish” in Unpopular Essays 1950)

*Calgary Southwest is no longer a federal constituency. It has been dissolved into Calgary Heritage and Calgary Midnapore.

Should Canada have Mandatory Voting?

When political scientists engage in studies of political phenomena there are many approaches that they may take. One oft-used approach is the so-called most similar systems design. This approach tries to “control for concomitant variation.” What does that mean? In social research it is difficult to clear cause-and-effect relationships because phenomena are complex and multi-faceted. Thus, if we wanted to determine why, for example, Canada is a relatively enduring and stable democracy, and Azerbaijan is not, one potential reason could be the relatively different histories of the two countries–Canada is a former British colony, while Azerbaijan was a former republic in the Soviet Union (which was a communist state).

Figure_3_3

Could this be the reason? Possibly. But, there are so many other differences between Canada and Azerbaijan that could also be the cause of the divergent outcomes regarding present political regime. Which one of these myriad differences, then, is the true cause of the difference between Canada and Azerbaijan regarding the level of democracy in each? (Indeed, the answer may not be mono-causal, but more complex and multi-causal.)

This is why many comparativists use the most similar systems design. By selecting units (countries) that are as similar as possible, they can control for many other potential causes for the alleged divergence in outcomes across the political phenomenon of interest.

So, let’s look at Canada and Australia–two countries that are quite similar in many respects: former British colonies, large land masses with relatively small populations, multi-cultural, constitutional monarchies, parliamentary democracies, economies reliant on natural resources, neither of which has won FIFA’s World Cup (men or women), etc. The two countries, differ, however, in levels of voting participation. Whereas barely 60% of eligible Canadians vote in federal elections, the corresponding figure for Australia is well over 90%. Do Aussies simply value political participation more than Canadians? Hardly! Australia has a mandatory voting law, which penalizes (monetarily) those who do not vote.

Should Canada enact a mandatory voting law? What do you think? Is it anti-democratic to force citizens to participate in the democratic process?

US Midterm Election Results and Climate Change

Election results in the United States are mostly final and the Republican Party has had a big night, capturing control of the US Senate, which combined with a Republican-controlled House of Representative means that President Barack Obama will face a united (in party name, at least) Republican Congress upon the opening of the new Congressional session–the 114th–which meets for the first time in early January of next year.

The New York Times has a handy graphic, summarizing the disconcerting results (from the perspective of climate change politics) of exit polls earlier today. This seems to be disheartening news to those who wish to see the United States government become more proactive in the are of climate politics and climate change. As you can see, while six in 10 voters said that climate change is a problem, fully 83\% of the partisans of the majority party in Congress believe the same.

us_midterm_elections_exit_polls_climatechange

Domestic Politics and Climate Change

Last week we discussed the role of domestic politics–institutions, electoral systems, partisanship, etc.,–on national political leaders’ attitudes towards and policies on climate change. We noted that the Canadian federal governments stance toward mitigation and adaptation changed dramatically upon the ascension of the Conservatives to power in 2006 (a minority government). The majority government that Harper was able to win in 2011 signalled the death knell for Canada’s involvement in the Kyoto process as Harper’s government reneged on Canada’s obligations quickly thereafter.

The United States, meanwhile, enters the final week of the biennial “midterm elections”, with most candidates (and the public) focused on issues other than climate change. When climate change is mentioned, however, the candidates responses are not reassuring. Have a look at this video for an impressive compilation of candidates’ responses to whether they believe in the existence of anthropogenic climate change. Incidentally, for a comprehensive debunking of Representative Steve Pearce’s claim that 31,000 scientists signed a petition claiming that there was no global warming, click here.