Some Prep Programs Do A Better Job Of Building Teachers’ Knowledge

Some teacher prep programs are more effective than others in ensuring their graduates have relevant content knowledge, judging from newly released data on teacher licensure tests. Figuring out why could help improve K-12 education outcomes.

How many prospective elementary teachers manage to pass a test on subjects they’ll be expected to teach—required for a teaching license—on the first try? And which teacher prep programs provided their training?

Those questions should be easy to answer. In other professional fields—like nursing and law—that kind of data is routinely used to evaluate the quality of professional schools. In education, it’s been surprisingly difficult to obtain. But today, pass rates for programs in most states are being released by the National Council on Teacher Quality (NCTQ).

The organization’s president, Kate Walsh, said that in some respects, “we’ve been working on [getting the data] for 20 years.” Congress passed legislation in 1998 requiring states to report pass rates for institutions within their borders, but Walsh says the requirements were interpreted in varying ways, making the data inaccurate—and allowing some states to exclude from their results candidates who failed the test.

NCTQ launched its own efforts by making overtures to the two companies that administer teacher licensing tests. But states need to request their data from the testing companies before it can be made public. About two years ago, NCTQ started approaching state agencies that oversee teacher prep programs. So far, 38 states and Washington, D.C., have cooperated.

Walsh is surprised the number is so high. One possibility, she said, is that states actually want to have the pass rates made public but don’t want to release it themselves because they’re under pressure not to do so from teacher prep programs. Some, including states that still haven’t cooperated, have been convinced by the programs that the data isn’t meaningful—or simply intimidated by them, Walsh said.

The data show that overall, only 45% of candidates in states that have a “well-structured licensure test” pass on their first attempt. By “well-structured,” NCTQ means that candidates are tested separately on different content areas, so that a high score in, say, English language arts isn’t allowed to make up for a low score in math. The overall pass rate has been public for at least a couple of years now, and it compares unfavorably with other fields. In nursing, for example, the first-time pass rate averages about 85%. Aspiring teachers can take the test repeatedly, but that costs money—and NCTQ says that 22% of those who fail the first time never take the test again, a figure that rises to 30% for test-takers of color.

The new wrinkle is the ability to connect pass rates to specific schools. Using three years’ worth of data—2015 to 2018—NCTQ has found that 29% of institutions had pass rates of less than 50%. In six states, there was at least one institution where no one who took the test passed. But there were also schools where pass rates were far better than the state average—including, in some cases, for students of color and those from low-income families, who fail the tests in disproportionately large numbers.

Some teacher-educators and others argue that licensure tests don’t predict classroom performance and only serve to limit the number of Black and Hispanic teachers, who have been found to have a positive impact on students of the same ethnicity. At the same time, overwhelming majorities of teachers, prep program leaders, and leaders of state education agencies agree that teachers should have to demonstrate their content knowledge before entering the classroom, according to NCTQ. The organization also says that, based on 11 of the 15 relevant studies it reviewed, the tests do predict teacher effectiveness. There’s also evidence, NCTQ says, that holds true for teachers of color.

One key question is why teacher candidates at some programs do much better than those at others: there’s an average 56 percentage-point difference between programs with the highest and lowest first-time pass rates within the same state. Walsh said NCTQ doesn’t have the capacity to answer that question, but she hopes states will pursue it. “Many states we’ve shown the data to have found it fascinating,” she said. “Their eyes have been opened to what they might do.”

Still, NCTQ’s report includes case studies that shed some light on measures that seem to work. At Western Kentucky University, aspiring elementary teachers take two courses in writing and language and three courses in relevant math content as part of the requirements for their major. They’re also guided to take a general U.S. history course rather than a course in Kentucky history. (Many students never get an adequate grounding in U.S. history before entering college, and when left to their own devices they may take courses that have little relevance to the elementary curriculum.) In their teaching methods courses, candidates use diagnostic tests and study guides to prep for the licensure test. The result: 60% of the program’s candidates of color passed the test on their first try, as compared to a statewide average for all test-takers of 46%. NCTQ’s report also describes apparently successful initiatives at institutions serving large numbers of students from low-income families that are aimed at boosting their content knowledge.

The report highlights efforts by a few states to use test data to improve teacher education. When programs in Florida opposed the state’s plan to publish pass rate data by institution on the ground that some scores were erroneously attributed to specific institutions—another frequent objection—the state worked with the testing company to get better data. Texas has made its data public and can withdraw state approval if a program’s pass rate is below 75% for three years in a row.

Still, the data needs to be used with care. While candidates of color did better than the state average at Western Kentucky, the overall pass rate for the institution was actually lower: 41% as compared to 46%. And some schools where candidates of color did well had few such students; a pass rate of 100% doesn’t mean much when it’s based on one test-taker.

More fundamentally, the failure to ensure that prospective elementary teachers have adequate content knowledge is only one of many problems with teacher education. Ed school faculty are generally unaware of or hostile to the body of scientific evidence on how people learn, and candidates are routinely encouraged to use techniques that contradict it. And many prep programs provide inadequate training in early reading skills. Walsh says that next spring NCTQ will release pass-rate data on licensing tests focused on that body of knowledge.

But pass rates on content knowledge tests, tied to specific institutions, can serve as an impetus to improving at least one part of a seriously flawed system. NCTQ’s evidence suggests that if programs identify gaps in candidates’ content knowledge early and take steps to address them, they can help ensure that those charged with teaching children about social studies, math, science, and English language arts will have a grounding in those areas. That doesn’t seem like too much to ask for.