So about two years ago, I perceived Behavioral Economics almost as a religion. I was obsessed with it because BE gives us guidance for being rational. So for a considerable amount of time range, I study and read anything about BE. You can actually take a look at my obsession in this blog because I upload an excessive number of articles about BE.
However, something did happen 10 months ago. I was at Data Analysis class, and we are required to crunch some data as our final assignment. It turns out that I love analyzing and crunching data. It felt like I was playing an open world game. Because you can do an abundance of activities only with data. Hence, I started reading and learning Data Science. I also land an internship at one of the biggest big data agency in South East Asia.
After studying a little bit of both about Data Science and Behavioural Economics, I could summarise that two of them are alike and congruent with each other. They have the same outcome which is to understand and predict human behaviours. In this post, I will compare both field of studies.
Behavioural Econ came from Richard Thaler in 1990s which one of the founding fathers of behavioural economics following by Daniel Kahneman, Dan Ariely, and other researchers. The field originated from the concern about the accuracy of classical economics. Because in classical economics, we make a lot of assumption that doesn’t really applies to the real world. Especially the assumption that we are always trying to maximize and optimize anything that would gain utilization or satisfaction. That’s why behavioural economics use psychology and economics and create a middle ground.
Similar to BE, Data Science also relatively founded in the 1990s. Because there was a need to process a massive and excessive amount of data so we can gain something from it, especially insight. One particular article that really shapes the Data industry that we know today is “From Data Mining to Knowledge Discovery in Databases” made by Usama Fayyad, Gregory Piatetsky-Shapiro, and Padhraic Smyth in 1996. That article introduced the world to the Data Science framework that we know today.
Method and Approach
Behavioural economics findings came from observations or curiosities about our true behaviour. For example, about a decade ago Dan Ariely observed that we would spend half an hour just to figure out which coffee shop has the cheapest coffee. But when it comes to $100 price gap when we looking for car across dealerships we would sacrifice that $100 for the sake of convenience. So he conducted more research about relativity.
Reversely, Data Science method is much more technical. The process usually follows a framework (simplified) such as:
Primarily, Data Preparation is everything that we do to make the data analyzable. Such as collecting, cleaning, and standardizing data. This process usually will take the most time, unless you work in a big corporation that entitled specific job desc to take care of the preparation process.
Data Analysis is the core of the process. It is the process when the analyst will run a lot of crunching, modeling, and predictive process. The popular ones are linear regression, black box model, or just a simple pivot table. Usually, the company doesn’t care what tools are you using, as long as you give a big impact to your product or company.
Finally, Presentation and Visualization is a process when the analyst will act as a liaison (middleman) for a more strategic or top management person by presenting complicated data and insights to a simple word and charts.
In my opinion, behavioural economics it’s like a complementary field of study. Because in my opinion, you can’t make a living for just mastering behavioural economics unless you are a researcher, teacher, or author. But BE could add a lot of value if you use it with other established field of study. For example, if you know finance and BE, then it will help you as a credit analyst. If you know marketing and BE, then you can leverage BE as a tool to make a great and effective marketing campaign that will communicate the audience better.
Before we go any further about Data Science career, here is the popular venn diagram for data science:
The diagram combines 3 skills that are crucial to Data Science. I would like to emphasize on Domain expertise. Which is a depth understanding of a specific field of study other than Data Science. This skill is prerequisite for any data scientist. Because like BE, you can’t make a living for just knowing the hard skill and technical skill of Data Science. If you are a data scientist for an agriculture company, you also need to know domain expertise at agriculture. That’s why Forbes recommend Data Scientists to learn more domain expertise and communication skills because it is not automatable.
I think we both know that behavioural economics and data science are complementing each other, although they have a deeply different method and background. Data science is relatively covered broader and macro topic compared to behavioural economics. This dynamic duo is currently emerging really fast. Even now in 2019, there is a couple of top universities that offer behavioural data science postgraduate major (UvA and Warwick).
REFERENCES AND SUPPORTING ARTICLES
Budihardjo, W. (2017, November 04). A Brief Introduction To Behavioral Economics. Retrieved from https://behavioralviews.com/2017/11/13/an-brief-introduction-to-behavioral-economics/
Domain knowledge. (2018, August 19). Retrieved from https://en.wikipedia.org/wiki/Domain_knowledge
Fayyad, U., Piatetsky-Shapiro, G., & Smyth, P. (n.d.). From Data Mining to Knowledge Discovery in Databases. Retrieved from https://www.aaai.org/ojs/index.php/aimagazine/article/view/1230
Gift, N. (2019, February 04). Why There Will Be No Data Science Job Titles By 2029. Retrieved from https://www.forbes.com/sites/forbestechcouncil/2019/02/04/why-there-will-be-no-data-science-job-titles-by-2029/#5936e0e3a8f6
Thaler, R. H. (2016). Misbehaving: The making of behavioral economics. W.W. Norton & Company.