How do online news organizations use hyperlinks? Judging from some websites, not very well. Several journalism researchers have noted that, compared to blogs, online news articles provide few hyperlinks. And most links that do appear in news articles lead to other pages within the organization’s site, which may not provide readers with the most useful information.
Why do news organizations resist linking to external websites? Researchers attempted to answer that question in a New Media & Society article published online in December. Tsan-Kuo Chang of the City University of Hong Kong and colleagues Brian G. Southwell (RTI International) and Hyung-Min Lee and Yejin Hong (University of Minnesota) surveyed US newspaper editors and television news directors about hyperlinks. The researchers concluded that decisions about whether and where to link had to do with economics: Providing a link to an outside website encourages readers to leave your site, while linking internally encourages them to click on more pages in your site, increasing your online metrics and, theoretically, your advertising revenue. The authors further concluded that “an ethnocentric journalistic practice” led news organizations to link more often to US websites than those of foreign media, governments, and organizations.
Economics and ethnocentrism are possible explanations for the hyperlinking choices that news organizations make. Still, it’s hard to see how the researchers drew those conclusions. For one, only 9.6 percent of survey respondents thought hyperlinks that led to external websites were “financially unwise”—a pretty low percentage to support Chang et al.’s economics explanation. And although the survey respondents indicated that they preferred to link to websites in the US, it’s not clear why. Perhaps the news managers see their reporting as superior to that of their foreign competitors. But maybe linking to foreign websites wouldn’t be appropriate for a news organization covering local crime, politics, or schools.
The survey also may not have asked the right questions of the right people. News managers are not necessarily the ones who make everyday decisions about hyperlinks. Our own informal queries of reporters and editors suggest that hyperlinking decisions might come from the top of the editorial hierarchy, the bottom, or anywhere in between; and sometimes, linking is done automatically with the use of computer software. Another problem with Chang’s survey was that it focused on opinions, not practices. The news managers were asked what they thought about hyperlinking generally, but not about how hyperlinks were used specifically.
Though CUNY journalism professor Jeff Jarvis advocates, “Do what you do best and link to the rest,” why would a news organization want to acknowledge that competitors have information that it does not?
Maybe because it provides readers who seek greater depth a richer array of information. Or because linking shows readers the extensive research behind the story they are reading. PolitiFact routinely links to the information sources it uses to arrive at judgments about the factuality of politicians’ assertions. In December, ProPublica initiated an experiment called “Explore Sources” with an investigative story by Marshall Allen, which includes in-text links to dozens of documents that Allen used in his reporting. Clicking a link opens a new window with relevant portions of the supporting document highlighted.
Publishing such heavily linked features holds reporters accountable by requiring them to show the sources of their work. It also helps readers find more information or deepen their research, if they want to. But features like “Explore Sources” also require a lot of work. Allen told the Poynter Institute that “there’s this whole meta-layer that goes into doing the story”—including posting the relevant documents online, highlighting the relevant sections, and linking the documents to the appropriate places in his article. Even if news organizations want to link to source material online rather than post it themselves, finding the best links can take time. Chang et al. did not offer their survey respondents the option of checking off “work-process-related issues” as a factor in decisions about linking. We suspect those issues matter as much or more to linking as economics and ethnocentrism.
Michael Schudson and Katherine Fink are contributors to CJR.