Too Early to Tell About Tennessee Promise

Education officials in Tennessee have released data on the Tennessee Promise, a program designed to make community college free for graduating high school seniors in the state. The results were encouraging, but it’s too soon to proclaim a smashing success.

Of the more than 13,000 students who participated in the Tennessee Promise class of 2015, 21.5% graduated with a degree or certificate, an increase over the 13.8% of students who accomplished the same thing the year before the program started. The data also indicated that only 8.3% of students who didn’t enroll in the Tennessee Promise in 2015 were able to earn a degree or certificate in five semesters.

“I have my degree and zero student debt,” one graduate of the program said in a National Public Radio report. “You do have to pay for your books and your parking passes, but that’s a heck of a deal. You can’t beat that.”

On the other hand, the dropout rate for the first Tennessee Promise class was just 2.3% lower than the rate of freshmen who dropped out in 2014. Some in higher ed also argue the data don’t actually show what caused the increase in degrees and certificates attained.

“We don’t have a benchmark for success on this because Tennessee’s the first state,” said Joni Finney, director of the Institute for Research on Higher Education, University of Pennsylvania, Philadelphia. “I’m just worried about pulling the carrot out of the ground too soon to see if it’s full grown. I think these programs have to evolve.”

Competitors’ Data Keep Students on Track

College and university administrators are increasingly using data not only from their own institutions but also from other, potentially competing, schools to predict when their students might require an academic intervention.

Observing and understanding data on common factors that impact student retention and success—such as feeling isolated or overwhelmed, selecting the wrong classes, or being unable to afford the next semester—enhances administrators’ ability to proactively identify which students need help. For example, using predictive-analysis processes developed by the University of Texas at Austin, administrators at the University of Kansas discovered that 1,200 out of 1,500 students having difficulties on their campus hadn’t received any kind of intervention.

Both schools are part of the three-year-old University Innovation Alliance, a consortium of 11 research universities dedicated to raising undergraduate graduation rates. Since the group’s founding, its member universities have managed to increase the number of degrees awarded by 10%, with a 25% bump among Pell Grant recipients.

Iris Scans Could Be the Next Student ID

Identification cards have served college campuses well over the years, allowing students to do everything from gaining access to buildings to eating in the dining halls to charging purchases in the campus store. Soon, however, advances in technology may make the ID card obsolete.

With fingerprint readers and iris cameras already basic components in smartphones, some colleges and universities are working on ways to put those features to work. Campus stores are already experimenting with low-energy Bluetooth beacons to offer shoppers discounts as they enter. Advances in hand-geometry readers, which identify the shape of a user’s hand, are also on the horizon.

The University of Georgia, Athens, will allow students to enroll in a system that uses iris authentication to enter dining halls and the student center. Georgia Southern University in Statesboro has used iris cameras to control entry into dining halls since 2013 and has found data gained from the technology useful.

“If we have a freshman who’s living on campus and required to have a dining plan, and suddenly we see the student’s not coming in anymore—what’s going on?” said Richard Wynn, director of Eagle Card services at Georgia Southern. “We can actually alert housing staff and let them know we haven’t seen that student in a while and they can actually go check on them.”

Iris authentication could also be used for entry into residence halls, the library, and sports venues, providing the institution an idea of how individual students spend their day. That sort of information would be valuable in university marketing efforts, yet it also brings up privacy concerns.

“These aren’t scanners,” Bryan Varin, executive director of UGA dining services, said of hand-geometry readers and iris cameras. “Both of them are simply taking a picture and ending up with a mathematical equation that grants you entry.”

Hacking Worries Slow IoT Adoption

The Internet of Things (IoT)—a network of computing devices embedded in everyday objects and connected to the Internet—can encompass everything from vending machines to light posts to entire building automation systems.

And that interconnection is growing, with the number of such devices forecast to surge from 13.5 billion units now to 38.5 billion by 2020. It’s already spreading into schools and across college and university campuses via online portals, digital textbooks, classroom devices, wearables, and other connections. Although such adoption of IoT has been slower at educational institutions than in the consumer market, experts predict that will begin changing this year.

However, one major roadblock to that growth remains the security of IoT devices. A sneak peek Verizon released ahead of its 2017 Data Breach Digest report recounts how one unnamed university was hacked via more than 5,000 connected devices on its campus.

“With a massive campus to monitor and manage, everything from light bulbs to vending machines had been connected to the network for ease of management and improved efficiencies,” the school’s incident officer at the time says in the report preview.

To regain control, the school had to shut down all network access to its IoT segments. “Short-lived as it was,” the incident commander says, “the impact from severing all of our IoT devices from the Internet during that brief period of time was noticeable across the campus.”

The preview identifies the underlying problem as “many IoT manufacturers are primarily designing their devices for functionality; proper security testing often takes a backseat.”

Survey: Americans Don't Trust College Value

Many Americans are unhappy with state of higher education and no longer convinced it’s worth the trouble. Nearly half of the respondents to a poll taken in August by Public Agenda said a college education is no longer a good investment because of the debt incurred, while 44% said they considered schools to be wasteful and inefficient.

About 60% also said institutions are mainly concerned about their bottom lines and that a college education is no longer necessary to be successful. Even more troubling is that the results were similar to a February poll conducted by the Gallup Purdue Index.

The outlook isn’t much better for students already on campus. An annual survey of more than 140,000 low-income students reported that because of costs only half actually enrolled into their school of choice. A third said they couldn’t afford their first choice and nearly 90% picked a school based on cost.

“Students are also less likely than in the past to go to college just to learn, the survey found,” education writer Jon Marcus wrote in an article for Hechinger Report. “A record 60% said they were pursuing degrees because they want to get good jobs.”

College graduates did admit to researchers that their education made an impact on their lives, but not necessarily in a good way. One poll found that about a third of grads who borrowed money for college had put off buying a house and a quarter postponed starting a business because of the debt.

U.S. Millennials Don't Have the Skills

The Program for the International Assessment of Adult Competencies (PIAAC) released global data in 2013 on how the U.S. population compared to those of other nations in terms of literacy and reading, numeracy, and problem-solving. The results weren’t pretty, especially for millennials.

The data revealed that the American millennial generation had the highest levels of education attainment of any previous generation, but that they demonstrated relatively weak skills in all the categories under consideration when compared to their international peers, according to a report in eCampus News. The research also found that too many millennials graduated without learning the right skills to enter a technology-based global workforce.

“These findings hold true when looking at millennials overall, our best performing and most educated, those who are native born, and those from the highest socioeconomic background,” wrote Irwin Kirsh, director of the Center for Global Assessments at Educational Testing Service. “Equally troubling is that these findings represent a decrease in literacy and numeracy skills for U.S. adults when compared with the result of previous adult surveys.”

PIAAC surveyed 5,000 people aged 16-65 in 22 countries for the study. U.S. millennials scored lower than 15 of the countries in literacy and ranked last in numeracy and problem-solving skills.

“If we expect to have a better-educated population and a more competitive workforce, policy-makers and other stakeholders will need to shift the conversation from one of educational attainment to one that acknowledges the growing importance of skills and examines these more critically,” Kirsch added. “How are skills distributed in the population and how do they relate to important social and economic outcomes? How can we ensure that students earning a high school diploma and a postsecondary degree acquire the necessary skills to fully participate in our society?”

Privacy Group Complains about Google

Google has become a dominant player in educational technology, with its Chromebook making up 51% of device sales to schools in the third quarter of 2015. Part of the reason for this rise is that Chromebooks are available to school districts at a fraction of the cost of other devices and come complete with free software used by more than 50 million students and teachers worldwide.

That’s an issue for the Electronic Frontier Foundation (EFF), a nonprofit privacy organization, which filed a complaint with the Federal Trade Commission accusing Google of collecting and data-mining Internet searches by children. The group claimed Google is using some of the collected information to sell targeted ads.

In addition, the EFF said Google is tracking user activity through the Chrome Sync feature, which automatically is turned on in Chromebooks. That tracking function is being used without parental consent, the group argued.

“In some of the schools we’ve talked to parents about, there’s literally no ability to say, ‘No,’” Nate Cardozzo, staff attorney for the EFF, told The Washington Post.

Google countered with a blog post claiming its educational apps comply with the law, while admitting that it does collect data of student activities to improve its products. In addition, both the Future of Privacy Forum, which authored the student privacy pledge that was signed by Google, and the Software and Information Industry Association sided with Google.

“We have reviewed the EFF complaint but do not believe it has merit,” said Jules Polonetsky, executive director of the Future of Privacy Forum.

Along with lack of parental consent, school districts are also not informing parents of online service policies. A 2013 study of found that 95% of the surveyed school districts relied on an online cloud service, but only 25% informed parents.

“Despite publicly promising not to, Google mines students’ browsing data and other information, and uses it for the company’s own purposes,” Cardozo said in a statement to CBS News. “Making such promises and failing to live up to them is a violation of FTC rules against unfair and deceptive business practices. Minors shouldn’t be tracked or used as guinea pigs, with their data treated as a profit center. If Google wants to use students’ data to ‘improve Google products,’ then it needs to get express consent from parents.”

Video Lectures Are Not That Effective

Researchers at Carnegie Mellon University, Pittsburgh, PA, have found that online lectures are not really very good teaching tools. Ken Koedinger and his team studied results of nearly 28,000 students who took massive open online courses (MOOCs) on the Coursera platform and found that video lectures were actually the least effective way to learn.

Students who only watched the provided video lectures did the worst on the 11 quizzes and final exam given during the 12-week course. Those students who used a combination of reading and video lectures did only slightly better. Students who did nothing more than complete the interactive exercises used in the course did about the same as those who combined reading and watching the lecture videos.

“People have such strong intuitions about their verbal learning experiences: ‘I learn from listening to a lecture,” Koedinger said in an article that appeared in U.S. News and World Report. “It sure feels that way. But in fact, lectures do very little to change your brain.”

The study recorded every mouse click to calculate how often a student read a passage, watched a video, or completed an interactive exercise, according to the report. Findings from the report were presented by Koedinger at the Association for Computing Machinery’s Learning at Scale conference.

“We have all kinds of assumptions about what works in learning and education,” Koedinger said. “But what we find out, not just in this field, but in almost every field, is that when you put those assumptions to the test, they’re more often wrong than right.”

University Servers Could Be Targets

Hackers may be turning their attention to data stored on college and university IT infrastructures. A hacker gained access to data from an unnamed U.S. university early in 2014, according to a warning issued by the Department of Homeland Security (DHS).

The hack initiated a denial-of-service attack against the servers and used about 98% of the school’s bandwidth. The DHS memo warned that government-funded research programs are appealing targets and university networks can offer hackers a way in.

“University networks, which often have multiple levels of connectivity and accessibility to fuel collaboration, may present easier targets for cyber-espionage actors than sensitive government or private-industry networks,” the memo said.

The memo also warned that less sophisticated cybercriminals may look to hack university networks to carry out phishing scams, insert ransomware, or create havoc with student financial information. The university network can also be used as a base for cybercriminal attacks on other IT systems because constant use by students can mask the criminal activity.

E-Reading Tracking Not as Invasive

After some slightly scary reports that Adobe was monitoring the reading habits of those using its e-reader application, it turns out the reality is not so Big Brother after all.

A TechWorld article reported the Electronic Frontier Foundation investigated Adobe’s actions and found the application only tracked e-book titles with digital rights management (DRM). Even then, Adobe only checked the first time the e-book was opened to make sure the copy was legitimate and not pirated. For books being used under a metered pricing model, Adobe was also measuring the length of time or number of pages consumed.

Readers of Fifty Shades of Grey can rest assured Adobe wasn’t looking to see how long they lingered over the juicy bits. In response to the reports, though, Adobe did make changes in how the application reported information, switching from easily hacked plain text to an encrypted file.

However, an app that follows e-reading patterns would actually yield a lot of useful data for publishers and would probably result in better books. For instance, publishers would be able to determine how many readers finished the book, where they tended to abandon it, whether they flipped back to previous pages, or if they seemed to skim through some parts.

edX Data Shed Light on MOOCs

Harvard University and the Massachusetts Institute of Technology released data last month on the first 16 massive open online courses (MOOCs) offered by their edX partnership. It turns out that while millions may register, far fewer actually complete the class and most people taking MOOCs already have a college degree.

While that comes as no surprise, the report may provide developers with valuable information for the future.

“There was a lot we didn’t know, especially about who took different types of MOOCs and how much of the course content they viewed,” wrote Jonah Newman and Soo Oh in The Chronicle of Higher Education. “This information may be valuable to those looking to design and lead successful MOOCs.”

For instance, while 57% of students on campus are female, 76% of students taking MOOCs are male. The median age of the participants is 24, with a third coming from North America. More than 20% of the students come from South Asia and students from that part of the world. They are most likely taking engineering and computer science courses, while Africans enroll in social science courses at nearly twice the rate of any other offering.

Students with a doctorate viewed the most course materials and students who took more than one MOOC were the most engaged in the course material, unless they took more than five classes, which is when engagement fell off.

Perhaps the most troubling trend is that nearly half of all registered students never look at any content associated with the class.

Universities Create Tech Consortium

While universities offer different programs to attract students, the technology infrastructure necessary to deliver them is often the same. That led Colorado State University, Indiana University, the University of Florida, and the University of Michigan to join forces.

The schools created a consortium called Unizin to negotiate contracts with technology vendors for products and services they purchase individually. Schools are currently paying millions to build and buy systems for storing and managing digital content, but Unizin is designed to allow its members to purchase a set of services in an effort to create a common infrastructure.

“Right now, that stuff is going into digital shoeboxes, one campus at a time,” Bradley C. Wheeler, vice president for information technology at Indiana, told The Chronicle of Higher Education.

The goal of Unizin is to allow member institutions to share data and learning materials through the common infrastructure without losing control of student data or intellectual property. The effort follows a 2013 position paper from the Committee on Institutional Cooperation that suggested institutions needed a better way to make smart technology decisions.

“It’s not about owning everything,” Wheeler said. “It’s about owning the rights to make choices that are in the interest of universities.”

Data-Mining of MOOCs Has Begun

Experts can debate the effectiveness of massive open online courses (MOOCs), but there’s no question about the amount of student information provided to register for the class. Now, colleges and universities are starting to work with technology vendors on ways to capture that data to use for marketing and recruitment.

“Showcasing the institution to capture prospective students is a rising reason [for offering MOOCs],” Katie Blot, president of the education services division at Blackboard, told University Business magazine. “There are more people talking about doing it than actually doing it.”

Institutions see MOOCs as a way to interact with students and spotlight the quality of their online classes. That leads to the question of whether to offer the courses for credit, with many only giving credit when the participant enrolls in a paid program, according to Blot. Administrators are also looking at the MOOC data to find out what assignments and discussions students participate in, when students stop participating, and why they stop.

“As long as MOOC courses are still dabbling in the credit process, accrediting bodies don’t feel the need to get into the mix,” said Rick Tomlinson, manager of academic solutions for Jenzabar, a provider of software and services for higher ed. “But I think that will change significantly once this whole process becomes more disruptive and we get into that credit-bearing mode.”