Find the Latest Resources in Education Today
May 20th, 2008
Study probes RFID use in schools
Radio-frequency tracking technology would be ideal for equipment but could violate privacy laws if applied to people, researchers say
Computer-science researchers at the University of Washington (UW) are tracking students and faculty in an experiment they hope will show the benefits and drawbacks of using radio frequency identification (RFID) technology to bolster school safety and security initiatives.
RFID systems have begun cropping up in schools in the past few years—and they’ve generated controversy over student privacy. The UW research is intended to give educators more information to help them make better decisions about when, how, and whether to use RFID technology in their schools.
Participants in the experiment carry RFID tags throughout the university’s six-story, 85,000-square-foot computer science building, constantly tracked by 150 antennas that record and store where and when people move from room to room. Gaetaneo Borriello, UW’s associate chair for research and a computer science professor who heads the RFID project, said 12 people have been tracked over the last six months, and he hopes to have between 100 and 150 subjects by next school year. Participants can have their information deleted any time they feel uncomfortable with what the RFID technology has recorded.
"We wanted to create an environment at scale to explore what the utility is … and how to approach those privacy issues," Borriello said, adding that the public has been slow to recognize RFID—already widely used to monitor the location of things such as cattle or commodities—as a useful tool for tracking valuable assets.
"The general public is worried about being tracked and what limits are on [the technology]," he said. "They don’t have a sense of the positive side of this stuff, what it can be good for. It’s kind of a knee-jerk reaction."
Most people are not aware of the ubiquity of RFID technology, Borriello said, which can be used in driver’s licenses, passports, and keys.
"This is a technology that is now going to reach widespread deployment, because it has gotten cheap enough and easy enough to install," he said. "We’re trying to get a little bit ahead of the curve and see what the implications of this might be."
Those involved with the UW experiment are testing for potential drawbacks, but some schools have seen these pitfalls firsthand.
At Brittan Elementary School in rural Sutter, Calif., parent complaints nixed a deal three years ago between school officials and local firm InCom Corp. to provide the district with RFID tags to track students’ attendance. InCom pulled out of the deal after some parents complained to the American Civil Liberties Union. (See "RFID spells trouble in tiny school district.")
And earlier this year, a technology company’s plans to help a Rhode Island school district improve bus safety by putting RFID tags on grade-schoolers’ backpacks were slammed by the ACLU as invasive and unnecessary. (See "ACLU rips district’s student-tracking pilot.")
At Enterprise Charter School in Buffalo, N.Y., students were issued identification cards with RFID microchips imbedded in each card in 2004. Every microchip contains the student’s personal information, making it easier and less time consuming for teachers to take attendance and creating an automated system for school administrators. Enterprise teachers also wear an RFID badge that allows them to open any locker in the school.
Enterprise officials said it cost about $25,000 to install the RFID system—a reasonable cost that will only get cheaper, Borriello said. It cost UW about $100,000 to set up the infrastructure for its RFID experiment, he said, adding that the project had to be approved by the university’s internal review board, which evaluates the validity of on-campus scientific undertakings.
As tracking technology becomes more common, there remains firm opposition to its use on people. Although RFID technology might be useful in some cases, people’s actions and movements can be carefully recorded and documented, sometimes without consent, said Lee Tien, a senior staff attorney with the Electronic Frontier Foundation, an organization that cautions against the intrusion of technology into private life.
"Remember that this is relational—widespread tracking means knowing who is with whom when and how often," Tien wrote in an eMail message to eSchool News. "If you tracked me and my wife, you couldn’t help but know that we’re together at night."
Experiments like the one at UW could help researchers learn more about the possible pitfalls of tracking technology, but Tien said personal rights would be violated if RFID was used to track students, employees, or anyone else in their daily lives.
"So the big trick is that the technology must be designed so that the user … is in control of who [is doing the tracking], when, [and] for how long, as well as all of the data," he said.
Asked what he expects UW researchers to conclude from their RFID project, Tien said, "What we already know—tracking is creepy, it makes you vulnerable."
Evan Welbourne, the lead graduate student in the RFID project, said despite meticulous efforts to create an "RFID ecosystem" in UW’s computer science building, results are often unclear. For example, if two people appear to be next to each other, there is a 90 percent chance they are standing in the same room, talking to each other. But there is a 10 percent chance that one person is in a room next door, Welbourne said.
"This has been a very difficult, iterative process, but it has also been a great way to structure our thinking about privacy and security in next-generation RFID systems," said Welbourne, who is using the RFID experiment as a thesis project. "We’d like to evaluate whether user-centered RFID systems can be built to be useful and secure enough to justify the potential loss of privacy."
Borriello said the project results will include some guesswork to compensate for the unreliability of using RFID to track people, not boxes or equipment.
"We sometimes have to fill in missing data," he said. "The technology is pretty robust, but the reading is less than perfect."