The Election Aftermath: Is Data Dead?


I didnít see any predictions yesterday for a Trump victory from the major news sources. Nate Silverís Five Eighty site said it was 71.4% sure that it was Clintonís victory. They were not alone. Nearly all the major media outlets were predicting the same thing. It seemed as if the only ones predicting a Trump victory was Team Trump, and oh and Katy Perryís parents.

Credit: Shutterstock
Credit: Shutterstock

By 3 AM EST it was clear the president-elect was Trump -- he had enough electoral votes. Based on the results this morning, looks like those predicting the Clinton victory were simply mistaken. Just like they flubbed the Brexit vote earlier in the year. Of course, just like the Republican primary, they got it oh so wrong.

As late as 2 AM Nate Silver was still giving hope to his forecast of a Clinton victory. He had noted earlier in the day about uncertainties with the predictions. He stated that polls were not perfect. They were just probabilities. Perhaps he was giving himself some wiggle room?

But who did get it right? Trumpís data team did. Wired magazine reported that Matt Oczkowski from Cambridge Analytica said their models were based on the rural voters. They were questioning him about data being dead. He noted that it comes down to knowing how to use the data. It was that simple.

My question though is, what happens to data now? Data and analytics are enjoying a popular surge. We can trust data when we want to make decisions. The data guides us. The data reveals the unknown. And my favorite quote, ďWithout data you are just another person with an opinionĒ. I donít know what to think. How did so many of them miss these results?

Should our argument be that they didnít have the correct data? Maybe the data collection process over sampled Clinton supporters? Another argument is that it was not a silent majority -- it was an invisible majority. Trump referred to them as shy voters who did not want to admit they supported him. The Clinton and Brexit forecast gave favor to the liberal viewpoint. Are they just so sure of their viewpoint that they donít even realize it is biased? Is it possible that the media and analysts are living in their own circle of affirmation?

But arenít data professionals (and journalists) just supposed to report the facts. Should we argue that these modelers with their education and experience were unable to do that? Should we instead say it is their very experience that should account for the data inaccuracies? After all, isnít it their job to put accurate models together that account for the invisible voters? Why was Oczkowski able to do that and Silver not?

Is data dead? Are the modelers not able to do their jobs? What do you think about dataís future?

Tricia Aanderud, Senior SAS Consultant for Zencos

Tricia Aanderud, Director, Data Visualization Practice at Zencos Consulting, provides SAS Consulting services to organizations that need assistance understanding how to transform their data into meaningful reports and dashboards. She has co-authored three books with her most recent title "Introduction to SAS Visual Analytics". She regularly shares data visualization tips and SAS knowledge through her BI Notes blog (http://www.bi-notes.com). Tricia has a background in technical writing, process engineering, and customer service. She has been an enthusiastic SAS user since 2002 and has presented papers at the SAS Global Forum and other industry conferences.

Predicting Traffic in the Sun's Shadow

How do you predict the popularity of a one-time event? A look at different ways to forecast road traffic along the path of the eclipse.

Where Sharks Attack in Florida

As Shark Week approaches, we've got some data visualizations to help you determine where it's safe to get in the water and where you will need a bigger boat.


Re: Catastrophic failure of polling
  • 11/29/2016 1:31:31 AM
NO RATINGS

They pointed to the the 1982(?) California governer election where everyone said they were voting for the then LA mayor. But come election day .... no they voted for the other guy.

When I do market research that involves predicting consumer behavior, especially when it comes to testing new products, I never include the question, "would you buy this?" It almost always yields inaccurate results. Predicting your own behavior well enough requires ample self-knowledge. As a marketer, I've grown to understand that consumers really don't know themselves well. What I do is I test for factors that will help me predict behavior, such as perceptions, past behavior, current consumption, etc. Or, if the brand could afford it, it would be best to actually sell the product in test markets, and then predict behavior in a more general market based on the data in the test markets.

Re: A Sea of Red
  • 11/29/2016 1:21:34 AM
NO RATINGS

I'd hate to say this amongst this wonderful group of writers, journalists, and data people, but it's clear that many US media outlets had been drowning in liberal ideology and in Clinton's "incentives." Being in such an environment can really convince you that there's no doubt Hillary would win. A Trump win just wouldn't make sense. Have that mindset in a tight race and it's easy for the data to bend according to your worldview.

Re: Fleeting Fame and Nate Silver
  • 11/18/2016 1:24:38 PM
NO RATINGS

Just to be clear I'm not implying that Nate was lucky in the past, in that last post I was agreeing that it's possible but he's been too close to actuals for it to be blind luck.  I also don't believe that his model is fundamentally flawed, it just didn't account for some things that very few if anyone saw coming.  It's part of working with data, sometimes new variables pop up and you have to adjust.  

Re: Fleeting Fame and Nate Silver
  • 11/18/2016 12:18:35 AM
NO RATINGS

In Nate's defense -

It's difficult to remember now, but Nate Silver was taking heat in the days before the election from people who thought his analysis was overestimating Trumps chances.

I don't think his prior success was luck at all. He's a skilled analyst. Who, like most of us, is not entirely unaffected by the chorus of other voices around us.

Re: Fleeting Fame and Nate Silver
  • 11/17/2016 2:37:36 PM
NO RATINGS

@Louis Watson, it is possible that Nate was just lucky in the past but I have looked at his results and he was very accurate in the past two Presidential elections down to the county level.  I think we'll hear more about the flaws in many of the models out there, not just his.  I have heard about exit polling issues, phone polling issues and of course the always sketchy web based poll.  If we learn anything from this election cycle maybe it's that the hivemind who upvoted Boaty McBoatface into the winner of a government run name our new vessel poll isn't the most predictable group. 

Re: Fleeting Fame and Nate Silver
  • 11/17/2016 2:37:35 PM
NO RATINGS

@Louis Watson, it is possible that Nate was just lucky in the past but I have looked at his results and he was very accurate in the past two Presidential elections down to the county level.  I think we'll hear more about the flaws in many of the models out there, not just his.  I have heard about exit polling issues, phone polling issues and of course the always sketchy web based poll.  If we learn anything from this election cycle maybe it's that the hivemind who upvoted Boaty McBoatface into the winner of a government run name our new vessel poll isn't the most predictable group. 

Re: Rural vs. Urban Perspective
  • 11/17/2016 4:07:00 AM
NO RATINGS

Hopefully the pollsters learn their lessons from this election, but given the realities --- the increasingly complex problems of contacting people who want to be reached, let alone those who don't --- I think they should quit while they are behind. As for the protestors who didn't vote, I am not sure they will learn their lessons. They are worse than sore losers who claim to be exercising their right to free speech but who refuse to protect that right at the ballot box.

Re: Can Any Analytics Be Trusted Now?
  • 11/16/2016 9:34:49 PM
NO RATINGS

Thanks @Bryan - I missed that one.

The US presidential process is long and tedious. People in Europe can't understand how it was "too late" for Biden to throw his hat in the ring when the election was still about a year away.

Despite being tiresome, I think it's a good process. People, especially politicians, can act like something they're not... for a few months. The year+ process means that all of us get to see our candidates for who they are - the best parts and the charactor flaws.

 

Re: Can Any Analytics Be Trusted Now?
  • 11/16/2016 9:19:45 PM
NO RATINGS

@PC

There is an urban radio show, called The Breakfast Club - very popular and ton of videos on You Tube.  She was a guest on their show (4/18/16) and made reference to having a bottle of hot sauce in her bag.  One of the culinary practices in the black comunity is to put hot sauce on foods such as fried fish.  And there are some folks who actually carry small bottles with them.  This was an attempt to pursuade the listeners that she 'was down with the homies'. In fact if you enter 'clinton hot sauce purse' into a Google search, this event comes up. This act, in addition to getting a shout out from Jay Z and Beyonce, was designed to get the black vote; the premise was that any one who carried hot sauce in their purse understood the needs of black americans and was fully prepared to adjudicate their claims for justice, equality and economic empowerment.

Hence from a socio-political perspective, she tried to identify with people who did not look like her and told the people who looked like her that they were a basket of deplorables - think about that one for a minute ...

She believed that she could sacrifice the votes of the deplorables (and the pollsters supported her in that belief) in order to get the black vote. She should have read Acts 1:8 in the bible - start in Jerusalem, move through Judea, Samaria and then the entire world.  In short Christ told his disciples to start with the base, then expand to those outside of the base and then expand beyond your region. HRC skipped her base in Jerusalem and Judea, took her hot sauce to Samaria and never looked back. If there was ever a time to use the bible for directions in running a campaign, this was it. 

Hence PC - HRC demonstrated that she was more in touch with black people than Bernie. Even though he marched with MLK, she carried hot sauce and was therefore the better candidate. And she also knew Jay z and Beyonce - so that proved that Bernie was not in her league.

And yes - we the analytics community affirmed her hot sauce by only polling the people who used it.

Re: Can Any Analytics Be Trusted Now?
  • 11/16/2016 9:18:16 PM
NO RATINGS

Oh, she carries a bottle of hot sauce in her purse?

And she tells people about this to establish that special bond - that we both like spicy food.

I can see why that didn't come off as real. Sheesh.

Page 1 / 10   >   >>
INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +