Where did you catch the debate Tuesday night? Broadcast television? Cable news channel? Public broadcast system? Streaming online?
How about Xbox LIVE?
If you don’t have an Xbox 360 or didn’t watch the first debate streaming on Xbox LIVE, where Microsoft and YouGov have been hosting both the presidential and vice-presidential debates, you may be surprised to learn that on Tuesday night, over 100,000 Xbox LIVE members tuned in via their video game consoles to watch President Obama and Governor Romney square off.
And that’s not all these viewers did: Microsoft’s been up to something unusual and arguably groundbreaking as each debate airs — allowing viewers to interact through real-time polling.
As CNN’s Candy Crowley, the debate moderator, was delivering her opening remarks, a dark blue strip at screen bottom displayed the question: “Should the electoral college be abolished?” and displayed live, shifting percentile results as viewers clicked “Yes,” “No” or “Don’t Know.” A moment later, another question: “Do you think journalists are generally biased?”
Over the course of the debate, which lasted a little over 90 minutes, Microsoft says viewers responded to some 70 polling questions in real time. That’s a lot of interaction — almost a question every minute. And here’s where the math gets interesting: Microsoft says individual responses to poll questions totaled over two million, each question averaging 35,000 to 40,000 responses apiece.
The company says all of that makes the second presidential debate on Xbox LIVE “one of the largest single-screen interactive TV experiences in history.” I don’t doubt it.
Let’s answer the question I’ve posed in the title. Microsoft says that “the aggregate raw responses to the poll questions clearly favored Obama.” But the company then drills deeper, shifting the focus to “undecideds and leaners,” a group that may determine the election in just a few weeks.
Their take on who won the debate? Says Microsoft:
Obama carried this 2nd presidential debate with the “Xbox Undecided/Swing Voters” — respondents who identified themselves as undecided or leaning toward one candidate or the other, prior to the debate.
The percentile breakdown: 51% of undecideds said Obama won, 17% said it was Romney and 32% of undecideds were — wait for it — “undecided.”
Granted, the demographic we’re talking about isn’t representative of the general electorate. I’m lucky enough to be married to someone who deals with polls in a professional research capacity, and she tells me this is what’s called a “non-probability sample.” That essentially means it’s not concerned with being representative, or — if you want to read more than you probably need to about probability sampling — as Wikipedia (correctly) puts it, “Nonprobability sampling techniques cannot be used to infer from the sample to the general population.”
Let’s break down the Xbox LIVE audience quickly, so we know more precisely who we’re talking about. According to Microsoft, 66% of Xbox LIVE users are male, while 34% are female. Eighty percent are aged 18 to 44, 51% are married and 31% are single. Fifty-nine percent are college-educated, 81% are employed, less than half make $75,000 or more a year, 77% are white (versus 18% non-white), and — here’s the surprise, or at least I found it surprising, given conventional assumptions about younger voters — 29% identify as Republican, 25% as Democrat and a whopping 42% identify as Independent.
Still, it’s interesting to see some of the pullouts here. Take “truthfulness”: Microsoft reports that 52% of responders who identified as “undecided” declared Obama the most truthful, while 17% said it was Romney. Or “specificity”: 56% of undecideds said Obama was “specific enough” about his plans, while 17% said as much of Romney.
Walking away from the debate, Microsoft reports 28% of undecided responders “either leaned or decisively supported Obama,” while 10% “either leaned or decisively supported Romney.”
Sidenote: Microsoft wisely requires respondents to click an answer before showing the poll results in real time, eliminating that as a potential biasing factor. But there’s something else the company is doing by simply asking questions at all: By crafting the poll questions and timing when they appear — often contextual, but sometimes not — Microsoft is, to a greater or lesser degree, shaping the way viewers experience the debate itself.
Seeing poll results in real time may influence your perception of the debate — a little like instantly accessible exit polling. If 52% say Obama was more “truthful,” does that make you more inclined to think it’s so, reinforcing your belief and potentially coloring your responses to subsequent related questions? It’s worth considering. After all, how, when and in what context you ask a question, as well as how you allow respondents to answer, is at the core of crafting meaningful poll questions.
In any case, three down, one to go. If you have an Xbox 360, Microsoft will stream the final presidential debate free through Xbox LIVE on Monday, Oct. 22, 2012, 9:00pm–10:30pm ET. If you’re an Xbox 360 owner and haven’t done this yet, you might want to tune in just to see what all the “live interactive polling” hullabaloo’s about. Whether through a game console or a simple set-top device, you’re looking at what may well be the future of the form, where you’re able to view live results as they happen instead of waiting for day-after results.