The phrase “big data” is often used in business, but many consumers have not been exposed to this new vocabulary. It is used to describe information from a variety of sources, which are normally processed by various software programs. A company with a large database may use big data to identify trends or behavior in their customer base. They may then use this information to take advantage of those trends to target their specific customers. For example, a retail outlet may use big data to find out what types of shoes its customers prefer, what brands they buy the most often, and which ones are more likely to make them happy if they buy those items.
Another application is to improve the performance of an organization by collecting and organizing much data about its employees. If you work at an office, then you are probably already familiar with the benefits of big data. Companies need to know which employees are doing well and which ones are underperformers; they need to optimize training and development budgets accordingly, and even use big data to determine the productivity of its employees. Today, big data has become almost a staple of modern business.
However, there is another potential application for big data. This one concerns the study of human vices and habits. In fact, there is now much empirical and experimental work on this topic, all of which indicates that there are significant relationships between people and certain habits or attitudes. This paper will review some of the applications of correlations in social science and address current questions concerning correlations and social influence. The paper will end with recommendations for future research on correlation and social influence.
The first application of big data is to investigate relationships among people. Correlations are easy to understand: two people might be at x and y for all possible combinations of factors that are measured. For example, suppose one person smokes and wears red shoes; does this mean that person is more likely to smoke than another person? Of course not. But it does show that the correlation is positive and could be predictive. By calculating the probability that a set of variables will cause a given result, researchers can improve their models and thus make them much more accurate.
Another application of big data is to investigate the economic value of different actions. We all have outdoor activities; some are better than others. Those that we enjoy are likely to remain with us longer and exert greater economic value. However, we often forget that these outdoor events have a social dimension as well. The same is true of the habits and attitudes that we have about the world in general.
Social networks may play a larger role in affecting our behavior and in creating and reinforcing different values and attitudes than previously thought. This is because people tend to trust those whom they see as having similar traits and beliefs as themselves, regardless of race or gender. So if people are already using social networks to create and reinforce their own value chains, it is not surprising that they will be willing to spread that information throughout the economy. This is just one more way in which big data-driven economic value chains may prove beneficial to society.
A final example of how big data can be used for improved decision making came from an unexpected source. In the context of economics, the impact of increased automation on work has been discussed extensively over the past decade. However, one small aspect of the automation revolution is the impact it has had on decision making itself. Researchers Vitaly V. Tsiveriotis and Alex lanes recently published a paper detailing how decision making when confronted with multiple choices was made more difficult as automated systems took over more of the role of deciding.
In short, Tsiveriotis and lanes showed that it is possible to observe large-scale patterns in data through statistical analysis and use that information to create improved policy and practice. The paper is timely, as recent advances in statistical analysis have afforded us the opportunity to not only detect trends, but to identify relationships too. V.V. Tsiveriotis and A.L. lanes’ work may open the door to better decision making by giving decision makers the statistical tools they need to make decisions grounded in fact.