Teacher evaluation overhauls were supposed to reshape the teaching profession. New research suggests they may have had an even greater impact on what it means to be a school principal.

As policy makers overhauled teacher rating systems in the last decade, principals began spending much more of their time watching teachers in action and talking to them about how to improve. But the shift also overwhelmed them with work, stopped them from fulfilling other responsibilities in their schools, and weakened their relationships with teachers.

Those are some of the takeaways of a study based on interviews with dozens of principals in six districts that revamped how teachers were evaluated in the last decade. Typically, the new evaluation systems — often put in place at the urging of the federal government and influential philanthropies, particularly the Gates Foundation — incorporated student performance for the first time but were driven mostly by teachers’ scores on rubrics that attempt to spell out good teaching practices.

While the systems’ adoption often precipitated significant pushback and in some cases have been rolled back, they remain in force in many places — and their long-term consequences are only now becoming clear.

The study, published last week in The Elementary School Journal, focuses on six large districts that embarked on dramatic evaluation changes: Baltimore, Denver, Hillsborough County (Florida), Houston, Memphis, and Metro Nashville. The researchers interviewed 10 principals in each district for about an hour during the 2012-13 school years, selecting a group at random while also ensuring a combination in different grade and performance levels.

One finding stood out. “The role of administrator is drastically changing,” one principal said. “You cannot be just a manager of a building. … You have to be an instructional leader first.”

The study examines only a handful of the many districts that reworked evaluations in the last decade, so it’s unclear how widely the findings apply. But the researchers detected themes across the six locations, which all had somewhat different evaluation systems and strategies for rolling them out. Here’s what they heard.

The new evaluations take way more time than principals feel they have.

The most consistent concern shared with researchers centered on the time it takes to complete the new evaluations. The responsibilities didn’t just include observing teachers more frequently, but also conducting pre- and post-observation conferences, completing additional paperwork and data entry, and using final ratings to make human resources decisions.

“I mean, honest to God. I just can’t do it by myself,” one principal told the researchers. “If I’m running my school, something’s going to lose out, either academics or your school because you’re just one person.”

“We need to [have] a preconference, an observation, and a post conference, you’re talking about, even though a preconference and postconference may only take 20 to 30 minutes and then an observation, you’re looking at two hours,” another principal said.

“It took us like an hour and a half to input the evaluations per teacher, and I have 82 people. So it was like a nightmare, literally,” said another.

In New York City, where the recently proposed teachers union contract reduces the number of required observation from four to two for most teachers, principals say the blanket requirements of the new system impeded their ability to make good choices about how they spend their time.

“We’re better off spending more time with the teachers who need our help than with the teachers who are doing a bang-up job,” the president of the union that represents school administrators told Chalkbeat. “To say time is at a premium for our members is an understatement.”

Principals are spending more time with teachers — and many say that’s a good thing.

Although the use of test scores has driven much of recent public debates about evaluations, the principals interviewed in the study often focused on the role of observations. That’s in line with past research suggesting principals trust that measure more than test scores; it’s also not surprising since principals conducted observations themselves, and those scores accounted for the lion’s share of a teacher’s overall rating.

While some frameworks for evaluating instruction have gotten mixed reviews, principals in the study said they found them helpful. Having an “objective” rubric, the principals told the researchers, allowed them to both give specific feedback to teachers who were improving and to build evidence to dismiss teachers who persistently struggled.

“For me, as a leader, it has given me a tool that I can use to help teachers grow, more than I’ve ever been able to do in the past, ever, because it’s a lot more specific than it ever has been in the past,” one principal said.

“The … rubric is the single most solid evaluative tool that I’ve used in my career,” said another.

“The conversations and the reflection … have gone from, ‘Oh, I thought it went great,’ to a really robust conversation.”  said a third principal.

Whether observations are really as “objective” as principals seemed to believe is not clear. Other research has shown that classroom observations are consistently biased against teachers who serve more students of color and students who start the year with lower test scores.

Principals’ relationships with teachers worsened.

Though there has been modest uptick in teachers who received low ratings, the vast majority of teachers across the country were still rated effective under the new systems.

Still, principals said the new evaluations damaged their relationships with teachers, who weren’t always sold that the new approaches were fair. They cited this dynamic as a major downside of new evaluation systems.

Some principals framed the tension as a lack of understanding on teachers’ part. “It’s very difficult for some teachers to understand that you really can quantify what they’re doing in the classroom, and many of them don’t want you quantifying it,” one said.

Others said the culture in the school had changed as teachers became more fearful of high-stakes evaluation, and thus less open with their principal.

“I feel like every time I walk in, a teacher’s like automatically on pins and needles thinking I’m there in an evaluative capacity,” a principal told researchers. “I just want to go in and see what the kids are doing.”

Principals ended up being less visible in their schools.

One surprising finding is that even as the new evaluation systems required principals to spend lots of time in classrooms, they said they ended up being less visible to educators and students.

“I’m in more classrooms, but I’m not walking through lots of classes. I’m mainly only in four classes or three classes a day,” one principal explained.

“I don’t do lunch duty as much,” another said. “I loved lunch duty because it’s a time to interact with the kids where you can just communicate with them.”

Trisha Arnold, a New York City teacher who helped negotiate the contract terms that reduce required observations there, told Chalkbeat earlier this month that she had experienced this phenomenon in her school.

“Kids aren’t sure who the principal is,” she said, “because they’re bogged down with paperwork.”

What to do with the research is unclear.

How do the benefits and costs of new systems weigh out? That remains a frequently debated question among policymakers and researchers. A recent analysis focusing on the Gates Foundation’s evaluation reforms in a number of districts yielded disappointing results, with little if any gains in achievement. (Two of those districts were the same ones featured in the latest study, which was also funded by the Gates Foundation; additionally, Gates is a supporter of Chalkbeat.) Another study found that evaluations systems likely deterred prospective teachers from entering the profession.

Other research is more upbeat, linking evaluation to test score gains and teacher improvement.  Studies of evaluation efforts in Cincinnati, Chicago, Washington D.C., and Minnesota have all found benefits, as well as national research on performance pay and feedback connected to classroom observations.

The latest research suggests that finding ways to lighten principals’ workloads might be worthwhile. That’s already happened to a degree, with some principals saying they placed more responsibility on assistant principals or other administrators. A separate Washington State study found a hiring spree of assistant principals soon after evaluation systems were put into place. That might be a smart move to help principals — but it also costs money, which could have been spent elsewhere, say, increasing teacher pay or reducing class size.

The new research also underscores how significant the choice of observation rubric is for districts and schools that adopt new evaluation systems. “Policy makers may also want to consider the enormous weight that is being placed on instructional rubrics as a means of guiding principals’ understanding of teacher performance,” the authors say.

Perhaps most importantly, the authors say, there should be more attention on principals’ role in enacting the raft of procedures connected to new evaluations.

“Without a careful examination of how to address principals’ concerns about lack of time,” they conclude, “the risk of principal fatigue, frustration, and eventual turnover remains substantial.”