Globally, ‘big data’ has recently emerged as a major enabler of various business processes. The term refers to the vast amount of data that a firm interacts with on a daily basis. It also encompasses three main features namely volume, velocity and variety. The amount of data available to an organization does not matter. The most important thing is how the available data is utilized . Big data, if well analyzed can give insights that facilitate better decision making and strategic planning. Due to these fundamental roles, it has become an essential component of political processes all over the world. One of big data’s notable uses in recent times was the 2016 United States (U.S) presidential election. Nonetheless, the final results contradicted what the data indicated. Thus, this essay seeks to explain why big data failed to predict the 2016 U.S election correctly .
In the 2016 U.S elections, all forecasts favoured Hillary Clinton. Consequently, the entire media fraternity was convinced that Donald Trump was bound to lose the elections (Bloomberg, 2016; Flores, 2016; Rutenberg, 2016; Stelter, 2016). Despite the predictions, Trump won the elections, becoming the 45 th President of the United States of America. The media has since admitted that the “ The numbers weren’t just a poor guide for election night — they were an off-ramp away from what was actually happening ” (Rutenberg, 2016). In this case, most analytics failed to capture the mood of a significant number of the electorate, who were dissatisfied with the government, Wall Street as well as the traditional media. It is based on this realization that CNN’s John King admitted that “We were not having a reality-based conversation” (Rutenberg, 2016). The value of modern polling, which is based on big data, has thus been highly questioned. Consequently , t his has elicited a debate on whether or not big data is dead.
Delegate your assignment to our experts and they will do the rest.
According to Steve Hilton, the CEO of Crowdpac, two big data failures were experienced in the 2016 U.S Presidential elections. The two failures were associated with the polling models and the voter turnout (Bloomberg, 2016). The polling model failed to incorporate two major groups of voters . The first one was Trump’s ‘secret voters,' which comprised of those that could not publicly declare their support for Trump for fear of moral shaming from the Democrats . The second group was composed of ‘invisible voters.' These were the first time voters who believed that Trump could change the status quo. Secondly, the voter turnout significantly affected Clinton’s chances of winning. Her campaign was backed by an “incredibly sophisticated data-driven machine” (Bloomberg, 2016). However, Hilton argues that this was not supported by a leader and a message that people could believe in . In support of this assertion, Keith Rabois, a partner at Khosla Ventures cites that while the Democrats had a superior marketing strategy, their candidate wasn’t as good.
The failure could also have been attributed to subtle features in the available data. For instance, according to Steve Huffman, the CEO of Reddit, Trump’s supporters were more active and engaged on Reddit compared to Clinton’s (Bloomberg, 2016). Therefore, the same engagement could have been converted into actual votes, fuelling a Trump win, and contradicting all the forecasts. Matt Oczkowski, the Product Director at Cambridge Analytic a, argues that the polling was also a bit off. In this case, the polls failed to identify Trump’s actual voters and thus could not predict his win. This group comprised of the disenfranchised citizens , those who had not voted in the last elections, and th ose who felt that the system was against them. Thus, Trump ’s supporters were “a bit older, a bit more male, a bit more white than the traditional Republican. A bit more rural ,” (Flores, 2016) . Additionally, the existing data showed an increase in the number of Hispanic/Latino voters. However, it did not identify who their preferred candidate would be. Eventually, this group voted for Trump as opposed to Clinton. Other factors that may have offset the forecasts included the final presidential debate which greatly favoured Trump, and Clinton’s E-mail investigation by the FBI (Flores, 2016) .
The failure of big data in the U.S election may be a reflection of the changing society. Informed by this, t he electoral process should not be viewed as the use of sophisticated targeting and soul-less mechanistic approach es to obtain votes. R ather , it should be about inspir ing people and develop ing p lan s that people can believe in (Bloomberg, 2016; Rutenberg, 2016). Therefore, overreliance on data-driven techniques may have contributed to the failure of big data. Conversely, Oczkowski argues that “ The science was not wrong, but the way people interpreted it was all wrong ” (Flores, 2016). When viewed from this perspective, big data is not dead. However, to understand and use it correctly, one has to go against normal political trends. In conclusion, the witnessed failure by big data to predict the 2016 U.S election was not a result of data flaws but rather its flawed interpretation.
References
Bloomberg (2016, November 10). Why Big Data Failed to Predict the U.S. Election [Video File]. Retrieved from https://www.bloomberg.com/news/videos/2016-11-10/why-big-data-failed-to-predict-the-u-s-election
Flores, R. (2016, November 10). Trump data analyst on rural voter turnout, Hispanic voter [Video File]. Retrieved from http://www.cbsnews.com/news/trump-data-analyst-on-voter-turnout-hispanic-voters/
Lapowsky, I. (2016, November 9). Trump’s Big Data Mind Explains How He Knew Trump Could Win. wired.com. Retrieved from https://www.wired.com/2016/11/trump-polling-data/
Rutenberg J. (2016, November 9). A ‘Dewey Defeats Truman’ Lesson for the Digital Age. nytimes.com. Retrieved from https://www.nytimes.com/2016/11/09/business/media/media-trump-clinton.html?_r=0
Stelter , B. (2016, November 10). 52 questions about the media and the Donald Trump presidency. Cnn.com. Retrieved from http://money.cnn.com/2016/11/10/media/reliable-sources-donald-trump/index.html?iid=hp-toplead-dom
