Betsy DeVos was back before Congress Tuesday.

The education secretary, who drew national attention at her confirmation hearing, fielded questions from testy Democrats who pressed her on proposed federal education cuts, expanded charter and voucher programs, and school discipline. She was there to defend the administration’s budget proposal, which, if recent history is any guide, will largely be ignored by Congress.

The exchanges underscored long-standing differences between DeVos and progressives skeptical of her educational philosophy.

As DeVos restated her positions on school choice initiatives and skepticism of spending more money on education, she also reprised a number of misleading claims.

🔗DeVos suggests increased spending isn’t helping students, but recent research says otherwise

“Over the past 40 years, federal taxpayer spending on education has increased about 180 percent, amounting to over $1.2 trillion cumulatively. And yet, we’re still 24th in reading, 25th in science, and 40th in math when compared to the rest of the world,” DeVos said. “Doing the same thing — and more of it — won’t bring about new results.”

This idea, that more money won’t improve academic results, is a common theme for DeVos. But it’s hard to square it with a substantial body of recent research that has connected more spending to better academic outcomes, both nationwide and in specific states.

And while it is true that spending has increased in inflation adjusted dollars, K-12 education spending as a percent of the country’s GDP has been fairly steady over time, with occasional ebbs and flows.

Of course, that doesn’t mean that every program that costs money has been proven successful. DeVos is on firmer ground citing specific programs that have fallen short, like the Obama-era school turnaround program.

🔗DeVos says the vast majority of today’s jobs will be gone in several years, but that’s unrealistic — and based on very little evidence

In an exchange with Rep. Andy Harris, a Maryland Republican, DeVos described the “the fact that a kindergartener today can look forward to entering a workforce where 85 percent of the jobs don’t yet exist.”

We’ve looked into this one, which has been made by other prominent leaders, and found it lacks any firm basis. A report from Institute for the Future makes the claim, but it’s not based on any rigorous analysis. Other citations lead only to defunct websites, unsourced “expert” claims, or YouTube videos.

More credible estimates, based on historical data, suggest that by 2030 only 8 or 9 percent of occupations will be novel. Tellingly, DeVos has repeated different iterations of the same idea: In 2017, she said 65 percent of the jobs of today would vanish by the time kindergarteners reach the workforce. It’s not clear what made her up the estimate to 85 percent.

🔗DeVos says schools are “deteriorating,” but there’s not much data to back that up

DeVos also suggested that American education has gotten worse since the federal Department of Education was created in 1979, echoing a comment made by Rep. Harris. “I know we were much higher in the rankings and we’ve definitely continued to deteriorate,” she said.

It’s not clear what she was referring to. In her opening remarks referring to the U.S.’s middling education rankings, she relied on results of the PISA exam. But that was first administered in 2000. It’s unclear that the U.S. ever ranked highly on older international exams.

(A spokesperson for DeVos did not address specific questions about her testimony. “The Secretary made very clear, and the facts back her up, that U.S. students continue to lag behind their peers in Reading, Math and Science, and that’s unacceptable,” Liz Hill wrote in an email.)

Nationally, fourth- and eighth-grade test scores, as well as high school graduation rates, have generally risen. Twelfth-grade scores — as well as gaps between affluent and low-income students — haven’t changed much, though. (Scores have dipped somewhat on the most recent federal exams.)

The question of whether the federal government’s involvement has helped or hurt schools is tougher to answer. Certain federal programs, including No Child Left Behind and the teacher incentive fund, have boosted test scores. But others, such as the school improvement grant initiative and Washington D.C.’s voucher program, haven’t, according to federal evaluations.

Despite the D.C. program’s disappointing recent test score results, DeVos proposed expanding it in her budget request. Older research on the program has found it did increase high school graduation rates, though it did not improve college enrollment.