Friday, October 18, 2013

Graduation Day for the Emerson College Polling Society

In Tuesday's Democratic primary in MA-05, more than just State Senator Katherine Clark emerged with a win. It was also a big night for nascent pollster the Emerson College Polling Society (ECPS). The outfit was the only public pollster to dip its toe into the congressional special election, and its October 2–8 poll correctly predicted Clark's 10-point margin of victory (albeit over the wrong runner-up). The firm has also polled the closely contested Virginia governor's race and June's Massachusetts Senate special.

But while these and other of ECPS's polls have been remarked and reported on, there has also been a fair amount of skepticism toward the pollster in the political sphere. This famously insular community was never going to simply welcome a newcomer like ECPS with open arms, of course; in a town like DC, you get credibility only by earning it. And rightfully so—it's good for people to rigorously question and test unknown commodities, especially when they're playing with live ammo to the degree that pollsters are.

But with ECPS, I question what people are questioning. The doubt that seems to surround ECPS in political circles is not a natural, healthy skepticism, but rather a dismaying bias against it for its unique status as a student organization. I've had debates with other political analysts about ECPS in which my peers were wary of believing its polls simply because they were conducted by students. Then, in the wake of the MA-05 primary, I read this well-meaning yet offensive post on the blog Blue Mass Group:
"This outfit, which appears to be a student-run organization at Emerson College, seems to be the only independent organization to have polled the race (more on that in a sec). I didn’t post on the poll they released last Friday because it’s hard to tell if they’re just a student club having fun, or if they are running real polls. … I still wish I knew more about how they’re set up … like, are any grown-ups involved with them? :)"
To be clear: I don't believe we should accept every new outlet that walks up and calls itself a pollster. But a student group is every bit as capable of producing high-quality, scientifically rigorous work as a professional firm, and it's insulting to say otherwise. ECPS has proven as much with the accuracy of their polls to date, exactly nailing the final margin of the Markey/Gomez Senate election as well as the this week's congressional primary. In the case of MA-05, ECPS especially impressed because of how notoriously difficult it is to poll special House elections (they deserve credit for being the only public pollster to even try).

Just as relevantly, their conduction by students does not imply a sloppier or less precise methodology to ECPS's polls. Even in this day and age, there remains a notion that party-hardy college students can't take things seriously or be responsible members of society. But like the ubiquitous columns that bemoan the laziness and entitlement of the Millennial generation, anyone trafficking in false and ignorant assumptions about students' capabilities is guaranteed to have no understanding or first-hand knowledge of the subject. This is not just a bunch of students playing around with phones and numbers. Just the opposite. Students don't often create polling societies, or stick their necks out into the adult world like this, every day. I can tell you from experience that the students who undertake something like the ECPS do so because they are uncommonly serious about learning—and about their craft. In fact, student organizations founded to compete with the big boys have a thirst to be taken seriously that is unmatched by anyone else.

A slightly more valid complaint would be that the people running ECPS don't even have college degrees—even though the same is true of countless accomplished individuals. But it is also an arbitrary line to draw, since a college graduate doesn't go from uneducated to educated the minute he receives his diploma. In fact, ECPS workers are totally immersed in what is probably the best possible influence on a data cruncher: academia. The pollsters of ECPS may not have that degree in hand yet, but they have more immediacy to the science of polling because they are taking classes and learning about it in the moment. (In fact, in my experience, student organizations are often book-smarter and more rigorous in their standards than professional firms because of this.) These classes, of course, are taught by universally renowned and credentialed experts in the fields of polling, statistics, political science, and more. Any gaps students have in their still-developing knowledge (important caveat: even as a 50-year-old, if your knowledge has stopped developing, you're doing it wrong) are more than made up for by the fact that those gaps can be filled by these unimpeachable authorities. Every student group I've ever heard of has had a faculty adviser, and a quick Google search reveals ECPS is no different. This is not "just a student club having fun." It's a responsible collection of students and their mentors who feel their skills are ready to be tested.

Not every student group should be taken seriously, just like not every company run by "adults" should be taken seriously. But college students founded, and in many cases still run, many businesses that are widely accepted and trusted. Let's Go, the internationally known travel guide on six continents, has been since its founding 53 years ago an entirely student-run organization—one I am proud to have written, edited, and overseen for. The social-media analytics firm Syndio Social started as a student group at Northwestern but now is an industry leader with huge corporate clients like Procter and Gamble. The Statler Hotel, the only luxury hotel in Ithaca, NY, is run by students at Cornell's hospitality school. Countless radio stations, including several that dominate their otherwise-underserved market, are run by students. And now, the Emerson College Polling Society simply seeks to join in the tradition of serious student contributions to the "adult" business world.

At 18, 20, even 22 years old, college students not only have "grown-ups involved with them"; they are grown-ups. Election law treats them this way; the criminal-justice system treats them this way; we should treat them that way. ECPS should be held to the same standards and subjected to the same scrutiny as any other pollster.

I'm no expert on polling, but to me ECPS passes this test. It has a detailed website, including a methodology page disclosing its vendors and laying out a polling procedure that, as far as I know, is accepted and standard in the industry. It releases the full scripts, raw counts, and crosstabs of its polls. If someone with more knowledge of polling wants or needs to see more to be convinced, cool—ask ECPS to show you more and I'm sure they'd oblige. But if what ECPS currently provides isn't enough, that further inquiry should be the default reaction—not a lazy dismissal of its chances as a serious firm without giving it a chance. If a new professional polling company emerged with the same level of public disclosure as ECPS, it would probably get the benefit of the doubt—and if not, it would be administered a proper and fair test to prove its worth. The industry's perception of ECPS should be based off the same.