We collected more data heading into this year’s Academy Awards than any event in our history, not including the 2012 election, which stretched over many months. Over the past few weeks, we collected over 220,000 survey responses including a range of questions about the show itself and predictions in most of the major categories (more on the predictions at the end). For now, we wanted to look at who will be watching the awards, who will be tuning in early for the pre-event festivities, and why.

Let’s start with the expected viewership. The first chart shows people we surveyed between January 22nd and February 22nd:

The second chart shows responses collected only in the past 24 hours:

We’ve seen a slight increase in expected viewers but overall the numbers have remained fairly steady. It looks like 13% of US consumers over age 13 consider themselves Very Likely to watch, with about 31% at least Somewhat Likely. That’s very close to 30% of consumers who were at least Somewhat Likely to watch the Grammy’s a few weeks ago.

As we delved into the profile of the most likely Oscars viewer, we found some fairly obvious correlations and a few that were not so obvious.

–  Women are more likely to watch. 36% of females are at least Somewhat Likely, compared to 25% of males.

–  Younger consumers are much more likely to watch. 49% of respondents aged 13-17 and 40% of respondents aged 18-24 are at least somewhat likely to watch. The least likely groups are those aged 35 to 44 (24%) and 45-54 (25%).

–  Black and Hispanic consumers should turn out in droves for the Oscars. A full 34% of Hispanic respondents say they are at least Somewhat Likely to watch. 33% of Blacks say the same.

–  People in the US West are the most likely to watch, followed by those in the Northeast, the Midwest, and then the South.

In general, the likely Oscars viewer is more likely than a non-viewer to live in an urban or suburban area,  have a bachelor’s degree or a graduate degree, and to be single and never married. She watches sports TV occasionally, prefers drama movies, and is, naturally, much more likely to go to the movies. When watching network TV, she is more likely than non-viewers to watch ABC most often. She watches CNN or MSNBC most for news, describes the National Economy and her own personal finances as good. She is most likely a registered Democrat and approves of how President Obama is handling his job. She prefers sedans over other types of cars. She’s more likely to be a Twitter user, smartphone user, and AT&T or T-Mobile subscriber.

Now let’s look at the die-hard Oscars fans. First, the people who attend some kind of gathering for the event:

5% of Americans will make the Oscars a special event with family and friends. 2% of people will attend a full-blown party with 10 or more people in attendance. Sounds fun.

Another group of super-fans will tune in early for the Oscar’s Red Carpet event to see their favorite stars and fashion. Here’s how that looks:

8% of respondents are Very Likely to watch the big pre-event and 24% are at least Somewhat Likely. Put another way, a full two-thirds of Oscar viewers will be tuning in early for the Red Carpet event. That’s more than we would have expected. Most of these will be women, who are 3X more likely to watch than men.

Finally, we asked 23,404 people what they look forward to most about the Oscars. Here’s what we found.

The largest group, 67%, don’t look forward to the Oscar’s for any reason, which matches up nicely with the 68% who said they were not likely to watch the awards. Of the remaining group, the largest number watch to see who wins the awards themselves. When we threw out the people wth no interest in the event, 44% watch for the award announcements, 31% watch for the fashion, 22% watch for the host and the performances. A measly 3% watch for the acceptance speeches (they had to be kidding, right?).

Among women, the most popular part of the Oscars is the fashion. Men watch for the awards competition and to see who wins. No shocker.

That’s it. Wait. What? Where are our predictions, you ask? Glad you brought it up.

It seems everybody and their mother has an Oscars prediction game this year. However, waiting to see how Golden Globes and other movie awards shook out before making a prediction is super-lame. (Note to Nate Silver: How about next year you wait until Monday morning to predict who wins? Talk about a look-ahead bias. Geez dude).

So, we went another route. We ran a multitude of experiments starting almost two months ago, including a stand-alone Oscars prediction website, where we would ask people who self-selected into the survey. We assumed these people were more interested in and more informed about movies. We also “intercepted” random respondents as they travelled the web. Some portion of these people were completely uninformed or, worse, partially informed about the nominees at all. Naturally, the self-selected group, at large, appears to have been more accurate in their predictions. Casual movie fans, for example, have a bias toward the movie Lincoln because they probably heard the most about it or even saw it. The late surge for Argo in recent weeks, however, came on the backs of movie enthusiasts and those who follow entertainment news most closely.

Across the board. our general population group and our self-selected group were in agreement, with the lone exception of Best Adapted Screenplay, where the experts see Argo winning and the laypeople picking Lincoln. But we don’t care.

Our goal is to study thousands of attributes of the people (especially the laypeople) who actually WERE right in all of their predictions. If we can figure out what they look like and why, we can get a lot smarter about predicting other things in the future. We’ll start all of that data mining tomorrow. Stay tuned.