The idea of a "digital divide" -- describing those who can or cannot get on the Internet -- has been around since the 1990s. Although, it used to refer mostly to access, now it often also means the quality and speed of Internet access -- and the skills to make use of the technology.

But who is actually responsible for closing the digital divide? It depends how the issue is framed, says a new Cornell study published in The Information Society, 27(2).

If presented as a problem of access, people tend to say the government or corporations are responsible, but if presented in terms of technical skills necessary to navigate and put information to use, then they are more likely to say the onus is on individuals and educational institutions, according to the researchers.

"The way you talk about the issue changes peoples' view of who is responsible for resolving it," said Dmitry Epstein, a doctoral student in communication who wrote the study with Erik Nisbet '94, Ph.D. '08, an assistant professor at Ohio State University, and Tarleton Gillespie, Cornell assistant professor of communication. "This issue has been around for years, but its meaning is in constant flux and is manipulated by political agendas."

Using data collected by the Cornell Survey Research Institute to test two frames, the researchers found that among such factors as age, gender, race, household income, education and how often people use the Internet, two factors had the most influence on who people said were responsible for closing the digital divide: how the digital divide was described and political affiliation.

"A conservative audience tends to view the individual and private enterprise as responsible for resolving the digital divide, whereas liberal respondents would allow government a greater role," Epstein said. "That sounds intuitive, but it was interesting to see actual empirical support for this assertion. Ironically, talking about the digital divide in terms of skills shifts the public perception of responsibility to individuals who are least capable of helping themselves in this situation. We can see this kind of unintended consequence also in discourse about poverty."

And while the term digital divide is still widely used, its meaning has changed. "How you talk about it depends on what idea you're selling," Epstein said. "Policymakers deploy it strategically, and it means different things to different stakeholders."

Recent events illustrate how being plugged into or cut off from the Internet can change history. "For some, access means buying something on eBay. Others use the technology to challenge regimes," he said, referring to Egypt.

The study has implications for policy communication and implementation. "We found that framing matters. The way you talk about the digital divide does affect how people respond to policy initiatives. A government policy focused on access may be perceived by the public as a responsibility of the private sector. A more efficient way to communicate such a policy would be in terms of public-private partnership."

"Governments and corporations can help to move us toward greater access and ability to take advantage of the Internet as a civic resource," said Gillespie. "There's clear support, here and abroad, for this to be a global project. But the partisan gap in how responsibility is perceived that we found may be its biggest obstacle. How the issue is framed for the public can really affect whether this becomes a public mandate or stalls on the legislative table."

Close to 80 percent of the U.S. population has some kind of Internet, less than in South Korea. Once everyone is connected, it is assumed people "will figure everything out by themselves," Epstein said. "But studies show that different socio-economic groups use the Internet differently. Richer people use it for capital enhancement" -- online banking, political participation -- but those less privileged "use the Net for recreation -- games, gambling." Whether more access would translate to more productive use "is a question," he said.

The study was funded by Cornell's Department of Communication.