In the field of research and data analysis, it is a well-known phenomenon that the outcome of almost any study is usually within the bounds of the investigators’ assumptions and expectations. This researcher bias has affected data collection and analysis since humans began collecting data about their environments, and it affects the data selected, sometimes resulting in less reliable conclusions.
It is human nature to “go with the flow” and sometimes take the path of least resistance, accepting time-honored observations in our fields or societal trends as near absolutes. But, with the mountains of digital data available today about billions of transactions and customer behaviors, it may be possible to see beyond assumptions into new vistas.
I ran into this idea head on while researching a very sizeable sample of more than 600 billion visits to banking websites and apps between January 2015 and March 2016. In addition to the analysis, Adobe Digital Insights surveyed 1,000 US consumers regarding their online financial planning and online-banking behaviors.
One thing I expected to find was the correlation between credit applications and interest rates. A common understanding in the banking industry is that, as interest rates fall, credit demand increases and vice versa. This has been the classic assumption for years.
However, that is NOT what I found. Rather, it was clear from the data that credit appetite in the US is not primarily driven by interest rates. It was pretty clear that there was a strong seasonal component to credit applications with a significant spike during the Christmas holiday season.
Now, the average consumer might say, “Well, of course.” However, the data analyst might say, “Well, that flies in the face of time-honored assumptions.” In fact, I suspect that this trend has been going on for quite some time, but it is only in the last couple of years since we have been gathering the huge amounts of digital analytical data that this trend could be clearly seen.
One approach would be to ignore this and continue with traditional thinking.
Chief Analytics Officer Spring 2017
15% off with code MP15
Big Data and Analytics for Healthcare Philadelphia
$200 off with code DATA200
10% off with code 7WDATASMX
Data Science Congress 2017
20% off with code 7wdata_DSC2017
20% off with code AIP17-7WDATA-20