From skimming and scanning to (the ultimate) reading, a new paper by Nir Grinberg looks at the ways we read online and introduces a novel measure for predicting how long readers will stick with an article.
Grinberg, a research fellow at the Harvard Institute for Quantitative Social Science jointly with the Northeastern’s Lazer Lab, looked at Chartbeat data for seven different publishers’ sites — a dataset of more than 7.7 million pageviews, on both mobile and desktop, of 66,821 news articles from the sites. (To protect the publishers’ privacy, they aren’t named in the paper, but Grinberg looked at a financial news site, a how-to site, a tech news site, a science news site, a site aimed at women, a sports site, and a magazine site.)
Chartbeat, Grinberg said, already offers publishers pretty good tracking. “It’s one of the few companies that track what happens with a user after they click on a news article,” he told me. “Still, the actual measures it provides are kind of raw. It’ll tell you how much time a person has spent on a page, how far down the page they got, even something called ‘engaged time,’ which is the number of page interactions — mouse clicks, cursor movement, etc. But all of these are not particularly tailored to news; they could work on any web page.” Grinberg tailored these raw measures to create new metrics specifically for news articles.