Last month, The Measurement Standard ran a review of author Seth Stephens-Davidowitz’s New York Times bestselling book, Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are.
We are thrilled that he agreed to an interview with TMS, as more and more communicators are now coming to terms with the fact that deriving insights from data sets, and building stories to convey information contained within that data, are now part of their jobs. There is much PR pros and those interested in measurement can take away from a read of the book. A notable takeaway (and thus the title) is that humans tend to be more open and honest with that anonymous white box on Google than they are in other settings—including social media. Analyzing the massive data set of Google searches performed brought many fascinating insights to the surface.
TMS – Thank you so much for agreeing to this interview. I really enjoyed your book, and want our readers to pick it up too. The case you have made for looking deeply into big data—particularly Google search data—is an important one for communicators to understand for a variety of reasons, so I’m looking forward to learning more.
TMS – There are a lot of Big Data sets to pore over—why do you find Google search data so compelling?
SSD – People tend to be extremely honest with Google. They tell the search engine things they might not say to friends, family members, social media, or surveys.
TMS – A lot of companies and PR and communications professionals are using data from social media—particularly Facebook, Twitter, and Instagram—to glean insights into consumers’ and target audiences’ thoughts and behaviors. If “Everybody Lies”—especially on social media—is it a mistake to base business decisions on this data? Would it be better to just rely on Google search data for consumer insights?
SSD – I think both social media and search can be informative. But you have to think through the incentives of the data. On social media, people present an idealized version of themselves.
TMS – Data analysis is becoming increasingly important across many work areas, businesses, and disciplines. What advice would you give to someone in college who might be pursuing a degree in a field like communications, so that they have the tools necessary to understand and analyze Big Data?
SSD – I think anybody should learn basic statistics and regression analysis. Learning the basics of machine learning probably is a good idea these days, too.
TMS – I completely agree—basic statistics is absolutely important, and regression analysis is getting more important for communicators to understand. Similarly, what recommendations do you have for someone who is mid-career and wants to understand how to examine and use Google data or other large data sets?
SSD – I recommend people start a blog. This will give a commitment to do many analyses.
TMS – I was particularly struck by the example in the book of the use of search data that appeared to show a higher rate of child abuse and neglect than was being reported during the Great Recession. That example seemed to highlight an important point: there are many streams of data, including the reported data and the Google search data. Realizing that collected data might not be reliable is a bit unsettling—how much weight should be placed on what Google data reveals?
SSD – Sometimes, you can compare Google data to a gold standard ground truth. For example, if you want to know how many people voted in an election, the best measure is actual government voting data—at least until Trump stops legitimate election. We can then see how well Google search data predicts this ground truth data. But with child abuse the story is different. The search data may be better than the government data because only a small percentage of child abuse cases are reported.
TMS – There are stark differences between the US and the EU in how data can be collected, stored, and used. Will this lead to a competitive disparity for researchers and businesses, given how much valuable information is contained in this search data—particularly because, as you note, Big Data can improve our decision-making?
SSD – Clearly, anybody who has better data will have a big advantage.
TMS – Thank you very much for joining us here on The Measurement Standard, it has been a pleasure to learn more about the book!