From social-networking websites such as Facebook, where users build connections with friends and strangers, to sites where children create and care for virtual-reality pets, the internet offers tremendous potential for social-science research.
But scholars fear that growing ethical concerns about the collection of data could lead to restrictions that damage the web's usefulness as a vast laboratory for social science. They say there must be an informed, nuanced debate about privacy, consent and data sharing - and one that does not automatically lead to more limitations and fewer research avenues.
The warning came last week as social scientists gathered in Oxford for the third Economic and Social Research Council Research Methods Festival, which included a session on e-research ethics in social sciences.
"Overregulation is a widespread worry," said Luciano Floridi, who holds a research chair in the philosophy of information at the University of Hertfordshire. He told Times Higher Education that scholars were well aware of how some areas such as medical research had become a morass of ethical regulations and feared that similar red tape would swamp e-research.
"There is a huge wealth of information available now on the web that can be used without asking too much permission ... (But) it could be made expensive and hard to reach by the funding agencies," he said. "If you had to start applying for ethical clearance every time you applied for something, it would slow down - if not downright stop - the research."
Internet users were developing a more sophisticated understanding of what online information they wished to keep private, he said, and as a result they were putting on the web only what they were happy to share anyway.
The only guidelines specifically covering web research were drawn up in 2002 by the international Association of Internet Researchers.
Bill Dutton, professor of internet studies and director of the Oxford Internet Institute, based at the University of Oxford, said the new world of e-research offered all sorts of novel and inexpensive ways of exploring social behaviour, but it also created a whole set of new ethical issues where the "correct norms were still yet to be decided".
"What is the status of information in particular spaces? Can I collect data about people through Facebook and social-networking sites that they post, or should I ask their permission? Is a person's avatar (the character they create in the virtual world Second Life) private? ... And if they do (give permission for a study) what if someone else wanted to use the data for a different purpose?" he asked.
Internet research often fell into a category that did not require ethical approval. "It may be that some research goes on without proper scrutiny," Professor Dutton said.
He said he believed that there were ethical issues to be resolved because people still sometimes naively reveal private information about themselves online, although the situation is changing.
The answer, however, does not lie in the creation of standard ethical guidelines for e-research that cut across all disciplines, which the UK is moving towards at the moment, he argued.
"There are clear ethical guidelines for (non-internet based) social-science research that vary dramatically across disciplines.
"We need to enable academics to apply the appropriate codes of their discipline. Standardising them (for e-research) risks trampling over the codes of some disciplines ... and it could end up creating hurdles for new research approaches," Professor Dutton said.
POTTYMOUTHED BOYS AND GIRLS IN CYBERSPACE: THE RESEARCH FRONT LINE
Michael Thelwall of the University of Wolverhampton uses the web as his laboratory. The most recent published paper by the professor of information science is entitled "Fk yea I swear: cursing and gender in MySpace".
The research it details compares gender differences in the swearing of US- and UK-based users of the youth-oriented social-networking website MySpace.
To examine the differences, Professor Thelwall developed software to trawl 40,000 user-generated MySpace pages. Although the data, which are all available publicly, can be quite personal, he aggregates it all rather than homing in on any individual.
Using the software, he collects data including a user's reported age and gender, the date he or she signed up to MySpace and the number of friends that users report.
His research falls into a category that does not require ethical approval from his school and he did not ask any permissions.
"The information I am collecting is public; it is a web page that anyone can access. (Permissions) are not an issue because I am not analysing them (the users) as individuals and I am not analysing them intrusively," he said.
He is worried that internet research could become overregulated, particularly in social sciences such as psychology, where interviews are the norm and where research would usually be scrutinised by ethics committees.
"The analysis of internet documents is something that is much less ethically problematic than direct person-to-person communication ... but it might be hard for (ethics committees) to make the shift," Professor Thelwall said.
