Marijuana: the Democrats' secret weapon in 2014?

Politics Editor Katie Riley delves deep into whether or not marijuana ballot initiatives can save the Democratic Party in the midterm elections.

For background on America’s midterm elections, check out Pi’s primer here.

The big prize for the Democratic Party in November’s midterm election is retaining control of the Senate. Although gaining a majority in the House of Representatives is almost a statistical impossibility for the Democrats, the Senate is still on the table. They haven’t lost it. Yet. (That doesn’t mean they’re not trying to make it as hard as possible for themselves, though.)

It’s common knowledge that the Democrats are in trouble. They’re not only fighting against an electorate that historically favours their opposition, but the leader of their party – President Obama – is also clearly unpopular.

If the Democrats are going to hold onto their weak grip over the Senate, they’re going to have to get those who often stay at home during the midterms to the polls – specifically in states with close senate races, like Alaska, because control of the chamber could come down to one seat.

One group of people which historically opts-out of voting in midterms, but could help Democrats win these key races, is young people. And what is one thing for which young people will be excited to vote? Marijuana, of course! At least, that’s the plan.

There are marijuana legalisation referendums on the November ballot in three states – including Alaska – plus the District of Columbia. The question is whether or not marijuana referendums actually draw young voters to the polls.

Research done by three political scientists in 2001 concluded that states with frequent ballot initiatives seemed to have higher average voter turnout. Obviously, this 13 year-old study that considered all sorts of referendums in numerous states does not hold the answers to this year’s specific question about marijuana and young voters. But, it serves as a good starting point because studies like this exist in the water supply of common knowledge.

In the age of data journalism, though, this really does seem to be an ideal question – it’s straightforward and (at least theoretically) involves lots of data. So, will marijuana referendums bring young voters to the polls, or not? Like most questions in politics, there’s no clear answer (and anyone who says otherwise is probably trying to impress you).

A recent poll conducted by two pollsters, one Democrat and one Republican, found that 69% of respondents would be “much more likely” or “somewhat more likely” to vote if there was a ballot proposal for the legalisation of marijuana in their state. But, there’s a fundamental problem with putting too much emphasis on this fact or this poll, and it’s three-fold.

First, they polled people all over the country, not just in states with ballot initiatives in 2014. There’s no state-by-state breakdown, so it’s impossible to tell exactly how just people from any one state, say Alaska, responded. Second, they surveyed people of all ages. According to this poll, people in all age groups are more likely to vote in the event of a marijuana referendum. (In fact, young people did not even lead in the category.) Third, it’s not an exit poll – you’re not asking them how they actually voted and why. These are people randomly called on the phone and asked hypothetical questions about a hypothetical future event. Yet, that is not to say that exit polls are so cut and dry either.

Take recent history: in 2012, Colorado, Washington and Oregon all voted on marijuana legalisation measures, and the exit polls from these referendums have been scrutinised by those trying to determine if similar ballot initiatives will drive young voter turnout this year. News outlets such as governing.com and ABC News are citing these elections as evidence that putting marijuana on the ballot gets more young people to the polls.

Yet, looking at the same election, other sources – most notably Nate Silver’s statistics super-blog FiveThirtyEight– have found exit polls to be an unreliable source. Instead, Silver’s team used data from the government’s Current Population Survey and found that the increase in young voters from 2008 to 2012 in Washington and Colorado was nowhere as large as the exit polls imply. In fact, it was pretty much the same as the average national increase in young voters between the two elections.

In the age of data journalism, especially when considering politics, we like to think that data can take us all the way – that numbers are like a crystal ball at which our predecessors just didn’t bother to look. This question just illustrates that even data journalism can only go so far, that people – and numbers – can only be so sure.

All of that being said, however, I would side with Nate Silver. Historically, ballot initiatives have not been enough to persuade young people to get to the polling booth – at least, no more than they would have done, anyway.