How UT-Austin’s Bold Plan for Reinvention Went Belly Up

Gregory L. Fenves had heard the calls for change, and he pledged to act. Standing before a quiet auditorium in the fall of 2016, the University of Texas at Austin’s president detailed a path forward: Project 2021.

In five years, he said, the university wanted most of Austin’s students to be able to enroll in revamped degree programs. Project 2021, officials would later say, would incorporate state-of-the-art online classes. Redesigned curricula. An academic calendar that included short courses outside of traditional semesters. And researchers would dig into data to examine every aspect of the undergraduate academic experience — to measure what worked and adjust accordingly.

“Even the best,” Fenves said in his address, “can do better.”

Not two years later, Project 2021 was dead.

The story of the program’s rise and fall, based on more than 20 interviews and a review of emails, reports, and other documents, shows how universities too often pursue the elusive act of transformation: promising too much while investing too little. Campus leaders in Austin had used sweeping words to describe the potential of Project 2021: Futuristic. Next-generation. Bold. Higher education is “in the throes of a revolution,” one progress report read, and Project 2021 would meet those challenges.

The Austin president hadn’t been alone in promising to do things differently. Research universities across the country have announced their own projects with comparable pomp, pushed by politicians and parents to produce cheaper degrees that teach practical skills and use technology to reach students who balk at the traditional residential experience. Tight budgets have heightened the stakes.

Project 2021 was full of grand ambitions but had too little support to succeed.

In retrospect, even as Fenves gave his address, warning signs blinked red.

Project 2021’s executive director sat in the audience during the president’s speech after spending months trying to get the program off the ground. Navigating Austin’s maddening bureaucracy without much administrative experience had already proved challenging. Meanwhile, a team of Project 2021 staff members streamed the speech in an office conference room, hoping to glean some new bit of knowledge.

Fenves had a grand vision. These people just wanted to know how they were supposed to realize it.

Higher Education’s Manhattan Project

James W. Pennebaker liked a big audience. “One of the appeals of coming to the University of Texas was to be able to have 500 people laughing at my jokes,” he said at a 2016 SXSW EDU conference.

“There is no joy greater, as many of you know.”

The professor liked gizmos, too. Working with programmers, he tried experimenting in large classes. He and a colleague introduced software that allowed professors to quiz students on the material in every class. This repeated testing, his research would later show, narrowed grade disparities between students from different socioeconomic groups in an introductory psychology course.

This marked a crucial turn for Pennebaker. He started wondering what other pieces of undergraduate teaching could change — and how universities can measure the impact of a good course on a student.

Project 2021’s first seeds were planted.

Encouraged by the results of repeated testing, Pennebaker scaled up. Move the class online, he thought, and the audience of students goes from a few hundred to virtually limitless.

The course first streamed in 2012, airing live. But this wouldn’t be just any online course. Modeled off late-night talk shows, the class would use a studio, green screens, and costumes, with a spotlight trained on the professor. Pennebaker and his colleague dubbed it a “synchronous massive online class” — a SMOC.

Samuel Gosling and James Pennebaker taught “Introduction to Psychology” as a SMOC in 2013.

The development came at a crucial time. Buzz over massive open online courses had transfixed administrators across the country, and then-Gov. Rick Perry urged colleges in 2011 to develop a $10,000 degree, naming online instruction as a way to reach the goal.

In this moment, the SMOC was exciting. Starting in 2013, the university touted the effort as showing how Austin could “serve as a national model” for using technology to “help students learn more effectively.” The Wall Street Journal, citing an enrollment goal of 10,000 students in one course, reported on the SMOC with the headline: “Online Class Aims to Earn Millions.” Dotted across a university curriculum, SMOCs could free up classroom space and faculty time for smaller, more-advanced courses.

Pennebaker’s work on the SMOC pulled him into conversations with top administrators about the future of education. As Pennebaker remembers it, university leaders asked him to put together an online-learning project.

But Pennebaker, now 69 years old, likes big ideas — and he wanted to think bigger.

He envisioned a program that could be higher education’s Manhattan Project: a research team that would take on the future of learning. The group would come together, shake up campus bureaucracy, lay the foundation for long-term change, and then dissolve. Pennebaker, the project’s lead, initially eyed 2020 as the expiration date, but he said his wife raised an objection, noting that everybody used 2020. That decided it: 2021. The project launched in January 2016.

“I will report back in five years,” Pennebaker said at that 2016 SXSW conference, “and tell you how we have mastered education.”

‘20 Questions’

Outwardly the university was just as enthusiastic.

“Move over, Enterprise,” read a 2016 pamphlet produced by Austin’s development office. “UT has launched a futuristic five-year mission of its own. Project 2021 will explore new technologies, seek out better ways of teaching and learning, and boldly go where no university has gone before in the development of next-generation undergraduate programs.”

But the project’s inner workings were less smooth than those proclamations indicated.

Top administrators had put globbed-together existing units under the Project 2021 umbrella: a center for learning sciences, rebranded as the Faculty Innovation Center; the Texas Extended Campus, which included courses for high schoolers and employees in the lucrative oil-and-gas industry; a research team; and a production studio that worked on SMOCs, among other things. Each piece of the project — dubbed by some former Project 2021 staff members as “orphans” or “misfit toys” within the university — had a different goal.

The innovation center would help departments revamp course offerings. The extended campus was expected to be the moneymaking arm, ideally enrolling non-degree-seeking students in droves. The film studio would churn out SMOCs — five dozen courses at once by 2021, according to one planning document. To Pennebaker, the heart of the project was the research team, which would assess courses or elements of campus life that altered a student’s path.

University leaders pledged that Project 2021 would shake things up, but they emphasized different facets of the program, including large and small online classes, simplified degree requirements, and experiential learning. Fenves’s stated goal — that half of students could enroll in a redesigned degree program in five years — was rarely repeated.

An Austin spokesman told The Chronicle that the different focus points reflected the different audiences these leaders were addressing. But even Pennebaker expressed some confusion.

“One of the issues that we will all have to work through is knowing what Greg’s vision is,” Pennebaker wrote to Maurie D. McInnis, the new provost, in the summer of 2016, referring to the president. “I sometimes feel as though I’m in the position of playing 20 questions.”

At first, Pennebaker’s team emphasized online classes. “The development of SMOCs at UT has been an instructive case study of innovation in teaching,” a planning document read, mentioning these courses more than a dozen times.

A revised version one month later stripped out nearly every mention of the SMOC. Highlighting the development studio’s role in online-course production “scared too many people,” Pennebaker wrote in an email to his colleagues. “The story we put together is now more compatible with Greg’s vision.”

Charged with realizing that vision was Pennebaker and his executive team. They had a small, tucked-away office. To Pennebaker, it called to mind a bunch of FBI agents holed up in a motel room — a far cry from the university’s Star Trek sales pitch.

“That group of five or six were relatively together and supportive of one another,” said Phillip D. Long, the project’s chief innovation officer. The group “banded together, as we started to realize the … currents that were impacting our naïve intentions.”

‘Not Even Close’

Early in Project 2021, Pennebaker approached the registrar with a question. What would it take, he wondered, to offer fractional-credit classes?

Such classes could make some students’ experiences more efficient. Say a student needed to take an introductory statistics course as a prerequisite for an upper-level class. Instead of delaying the advanced course by a semester, why not distill the statistics skills needed for that advanced class into a three-week, half-credit course?

The type of initiative would change the very definition of a class — a perfect fit for Project 2021.

Pennebaker quickly learned that such a change, though seemingly minute, would have far-reaching consequences. Updating student information systems cost other major research universities tens of millions of dollars each, and modernizing Texas’ aging system to handle the new classes would require a steep investment despite the various upgrades Austin had made over the years.

But it wouldn’t just take money. Making the change would touch many academic departments and staff offices, too. There could also be ramifications to federal financial aid, among other things.

This was a wake-up call for Pennebaker. Before Project 2021, he thought he understood how the university worked. He’d been a department chair for nine years, and creating the SMOC had connected him with deans’ offices and campus technology divisions. After hundreds of hours of meetings, he realized he was wrong. “I didn’t know anything about how the university functioned.”

Department-level changes were easier to make. Individual departments from journalism to French and Italian were making and assessing changes to their curricula, aided by Project 2021 money and staff. The history department, for example, added a digital tool that helped students visualize historical events and then surveyed them on their experiences using it.

But notes of frustration about the university’s bureaucracy began to seep into Pennebaker’s speeches about the project.

In a spring 2017 faculty-council meeting, minutes show, Pennebaker spoke of the problems with making the technological shift that would allow Austin to enroll 1,000 students in online classes, rather than just 999.

The fractional credit has not been implemented, he said. “Not even close.”

A Costly Hunch

Curricular change proved difficult, but the project was succeeding at another task: cranking out SMOCs, each of which could cost tens of thousands of dollars to produce.

Pennebaker told faculty and the public that his psychology SMOC worked. “We find that students learn better by using this online method,” he said at a 2017 conference. Not every class should become a SMOC, he said, but wider adoption could bring more efficiency. “We need to start experimenting.”

Still, the idea that the heavily produced SMOC is a superior tool of pedagogy is far from proven.

To Daniel H. Robinson, who analyzed SMOC data on the research team, it was clear: There was absolutely no advantage.

Project 2021’s research team studied two aspects of the SMOC. In one review, the team found that instructors in SMOCs received slightly lower teaching evaluations than those leading in-person classes, though differences were not statistically significant.

In a second, researchers looked at whether SMOCs could make students more broadly into better learners, using repeated testing. They assessed how students enrolled in eight SMOCs did in their other courses that semester. In half of the SMOCs, enrolled students performed slightly worse than peers who had enrolled in comparable face-to-face classes. Students in just one of the eight SMOCs examined — the psychology course initially piloted by Pennebaker — outperformed face-to-face students in their other classes.

Neither study analyzed the part of the SMOC that was so expensive — the production quality.

Jason D. Ferrell, one of Pennebaker’s graduate students, wrote his dissertation on SMOCs and found them to be about as effective as face-to-face classes. He told The Chronicle he did not believe the SMOC’s calling card — the studio quality — enhances learning over early low-quality attempts. “Somebody with a $100 webcam in their office could do the same things, as far as getting the material across,” he said.

Pennebaker said the research team hadn’t studied this because of its small staff and limited budget. He says he has long tracked student performance in psychology SMOCs and feels confident that students learn effectively — and he says he still believes SMOCs work when they test students repeatedly.

It’s a hunch, though an intuitive one, that the highly produced, snappy segments are better than a cheaper alternative. “This gets to the central question of what works and what doesn’t,” Pennebaker said. A static shot of a person’s head lecturing “is real tough to maintain attention.”

Joey Williams, a university spokesman, said students rate SMOCs highly and faculty say they are effective.

The costs of filming a new SMOC are between $9,000 and $66,000, according to data provided by Williams, with about two-thirds of classes between spring 2015 and fall 2018 costing more than $20,000. These figures don’t include instructional costs.

The SMOC development team says comparisons between in-person and online classes aren’t black and white. Consider the costs of constructing and maintaining auditoriums, said Marla Gilliland, director of course development with the online-learning program, that can house only a few hundred people for in-person classes.

During Project 2021, administrators expanded SMOCs “in a way that we hadn’t before,” Gilliland said. “Our primary role was taking this foundation that we built and trying to expand … it across campus.”

‘We’re Screwed’

Project 2021 faced a slew of problems from its early days: lack of direction, a bureaucratic quagmire, and doubts about the SMOC. Money wasn’t supposed to be on that list.

Project 2021’s budget would rest on three legs: the university system, which would contribute a one-time $16 million; the extended campus, with revenue from non-degree-seeking students and other operations; and the provost’s office, which would kick in a few million dollars annually.

Shortly after Fenves’s fall-2016 speech, the first leg started to tremble.

McInnis had started as provost shortly after Project 2021 launched. Early in her tenure, she was wary about giving Project 2021 the system’s $16 million. She had seen universities “pour tens of millions of dollars into the fad of online, of MOOCs, of a variety of different things,” with “little to show for it.”

She said in an interview that the plan for how Project 2021 would spend the money “was not concretely articulated.” Pennebaker had sent her plans for the project. But she didn’t have a clear sense, she said, of how Project 2021 would use that money to drive change.

Her office, then, would not release the money.

Pennebaker said he was not shaken in October 2016 when he got the news. He hadn’t touched that money yet, and the other pots were reliable.

But in December, the second leg cracked. Jamie Southerland approached Pennebaker in his office, Pennebaker recalled, after digging through the books of the extended campus.

The extended campus was supposed to bring in $3 million annually. But Southerland, Project 2021’s chief business officer, had realized that the program, which predated Project 2021, had been actually losing about that amount. (Southerland did not respond to requests for comment.)

Within the extended campus, the petroleum-education group had started shedding business when oil prices dropped in 2014, and the division that allowed non-degree-seeking students to take online courses experienced a sharp enrollment drop, according to a summary document sent to the provost.

Pennebaker recalled that Southerland’s message was clear. “We’re screwed.”

Project 2021, then, wobbled on one leg: the university’s own budget. That too was under strain.

Texas’ legislative session started in early 2017, and the UT system was in the spotlight for high spending. That semester, Pennebaker learned that Project 2021 would receive smaller distributions from the provost’s office. The cuts would be substantial, totaling about $1 million and amounting to a 40-percent cut from the previous year.

In a recent interview with The Chronicle, Fenves said the money that Project 2021 received was “enough to get started.” He didn’t want to make a big, up-front commitment to something in innovation. Instead, the university wanted to take an iterative view. “Sometimes, things don’t work, and you pivot.”

Pivoting here meant layoffs and cutbacks. Pennebaker spent his first year as Project 2021 director trying to understand the units he now oversaw. He spent the second year trying to stop the bleeding.

By the summer of 2017, Project 2021 cut staff in the extended-campus unit and pawned the petroleum group off to the university’s engineering department. In one summer update, leaders predicted that they could shed the high-school classes, too. The extended campus’s director, Stephen Walls, tried to mitigate layoffs, believing the deficits were smaller than initially indicated, before he too resigned in September.

“I couldn’t be an effective leader of extended campus as long as it was part of 2021,” he said, adding that the larger administrative structure made it tough to get a complete picture of his unit’s work.

Other arms of the program weren’t doing much better.

Project 2021’s leaders said they would pare back curriculum-redesign efforts. They would no longer try to “change most departments” and would offer guidance only to deans and department heads who asked, according to one 2017 report. Even some of those requests were rejected. Of the three full-time analysts who made up the research team, one was laid off and one resigned. The third required “significant retooling,” according to the same report.

Emails flew between Project 2021 staff members that year, announcing departures. One person took a job at the university’s Moody College of Communication. Two left for its Dell Medical School.

“Gentle reminder,” one person wrote after a departure. “P2021 is short staffed … things are about to get interesting.”

Pulling the Plug

Pennebaker had thought about how this all might end.

In November 2017, he wrote to McInnis that “almost two years into the existence of Project 2021, only minimal progress has been made.” He presented three possible options for the program’s future in an email.

First, Project 2021 could scale back, with layoffs and limited potential for innovation. Second, the university could turn Project 2021 into a consulting group for the president and provost.

Third, the university could shut down the project. “Indeed, completing a project three years ahead of schedule,” Pennebaker wrote, “could be viewed as a remarkable success.” (He told The Chronicle the comment was tongue-in-cheek.)

Austin pulled the plug soon after.

Project 2021 had been based on the idea that a new group could measure learning and drive change in the classroom and across the university. But McInnis says that’s not entirely how she sees things.

Data can’t entirely capture many aspects of “those complex ways in which you touch students,” McInnis says. And instructional changes at a decentralized place like Austin should be left to the departments or individual colleges, she said, not a separate program like Project 2021.

Fenves’s grand goal — that half of the university’s students would be able to take redesigned degree programs by 2021 — had faded. But that hadn’t been widely discussed around the university, McInnis said, nor was it something Austin was shooting for. It was instead a useful bit of messaging, “a great rhetorical claim to get a campus energized about the work.”

McInnis doesn’t say Project 2021 was a failure. She points to departments changing pieces of their curricula, particularly in experiential learning, and the SMOC’s expansion as important outcomes. Some of that work is continuing. But taking the university’s budget into consideration, she said, she’s glad the guillotine dropped sooner rather than later.

Today, Fenves doesn’t think the experiment was fruitless. They needed to try, and in any innovation, he said in an interview, “you don’t usually know where you’re going to end up.”

The Project 2021 team has varied explanations for what went wrong.

Hillary Hart, the Faculty Innovation Center director, said it was the combination of not enough time, resources, buy-in, and attention — “that’s what doomed” Project 2021. Long, the chief innovation officer, now more clearly sees the challenges of focusing a public research university on undergraduate reform — the research budget and research agenda are top priorities.

Pennebaker, who has since returned to the psychology department, hasn’t ruled out being involved in a project like this again. He still thinks a lot about the possibility of change writ large.

He’s considered leaders like Michael M. Crow, the Arizona State University president who has expanded enrollment by tens of thousands of students. Perhaps large institutions will only change with such a charismatic figure, he muses, or when their peers do, or when they face existential threats.

“Perhaps universities like UT and other large universities simply won’t change,” he said. “The general conclusion you have to come away with is, large institutions virtually never do.”

Public research universities certainly say they will. Several have set 2023 as a deadline for change. By then, expect the University of Massachusetts at Amherst to “Be Revolutionary” and Oregon State University to have achieved “Transformation, Excellence, and Impact.” Others look farther ahead still. Ball State University: “Destination 2040: Our Flight Path.”

Austin rolled out Project 2021 with fanfare in 2016. Its eulogy came in a short note from McInnis in June 2018. The decision was made, she said, “after evaluating early progress and taking into consideration the need for budget restrictions.”

One employee realized that some of the Project 2021 staff may not have received the email, as the email list hadn’t been kept up since an administrator left the university.

But Fenves got the note. Less than two hours later, the president responded:

“Nice message.”