Small Data

“Data can help compensate for our overconfidence in our own intuitions … but there are many things big data does poorly,” writes David Brooks in The New York Times (2/19/13). “Computer-driven data analysis,” David writes, “excels at measuring the quantity of social interactions but not the quality.” It can capture the number of times you interact with your co-workers daily, for example, but not “your devotion to the childhood friends you see twice a year.” Data can’t create context or tell a story, the way people can; it can’t “weave together multiple causes and multiple contexts.”

Even though big data enables us “to find many, many more statistically significant correlations,” most are “spurious and deceive us when we’re trying to understand a situation. Falsity grows exponentially the more data we collect.” The needle becomes more deeply embedded in the haystack. Big data isn’t very good at big problems. It can determine the most effective email campaign through “a randomized control experiment,” but you couldn’t apply the same technique to figure out the best way to stimulate the economy, since it would be impossible to create “an alternate society to use as a control group.”

Data, writes David, “favors memes over masterpieces. Data analysis can detect when large numbers of people take an instant liking to some cultural product. But many important (and profitable) products are hated initially because they are unfamiliar.” Finally, he writes, there is no such thing as “raw data” because it’s always structured according to somebody’s predispositions and values. The end result looks disinterested, but, in reality, there are value choices all the way through, from construction to interpretation.” It’s not “that big data isn’t a great tool,” David writes. “It’s just that, like any tool, it’s good at some things and not at others.”