The number of new teachers completing science education certification programs fell by nearly half from 2011 to 2021, even as states continue to push students to take more science, technology, engineering, and math courses.
Experts at a National Academies of Science meeting last week highlighted the need to increase the flow and preparation of new K-12 STEM teachers.
In a longitudinal analysis of federal teacher data, Richard Ingersoll, an education professor at the University of Pennsylvania, found that the number of new science classes and staff positions is vastly outpacing the number of teachers being prepared for them. In the last 30 years, the “average” STEM teacher has moved from a 15-year veteran to a novice, with double the number of new teachers now as there were in the 1980s. That “greening” of science teachers comes both from ongoing retirements of veteran teachers and states’ rapid expansion of science and math course offerings, increasing the number of positions to fill.
“We have a severe problem with people assigned to teach things who don’t necessarily have very much background in the field,” Ingersoll said.
‘A worrisome development’
Michael Marder, a physics professor at the University of Texas-Austin, agreed. Since 2022, in Marder’s home state of Texas, the number of science teachers who are entering classrooms with no science certification—including traditional college-based or alternate preparation—now make up the largest share of the teaching workforce. Uncertified science teachers for the first time outnumbered the combined total of all teachers who have completed science teaching programs from state colleges and universities. Uncertified science teachers also outnumber the science teachers trained via the state’s extensive alternative certification programs, according to an analysis by UT’s Educational Research Center.
“This is a very worrisome development,” he said. “The whole process of certifying teachers seems to be crumbling under the weight of a vacancy crisis.”
Staffing classrooms with a combination of younger and less-prepared science teachers can exacerbate a cycle of teachers entering and quickly leaving the profession, the researchers said. While about a third of all new teachers leave in their first five years, novice STEM teachers have an even lower retention rate, Ingersoll said, with as many as half leaving the profession within their first five years.
“It’s something of a revolving door” for school districts, Ingersoll said.
Teacher recruiting issues also are likely to worsen opportunity gaps for historically disadvantaged students. Federal civil rights data show nearly 1 in 3 Black and Hispanic students attend a high school that does not offer computer science compared with 1 in 4 white students.
The researchers called for better preservice training and recruitment of science majors into teaching, as well as a better transition for new STEM teachers entering the classroom.
“If we want [students] to be scientists, they have to be engaged in science, right? If we want them to be mathematicians, they have to be doing mathematics, not hearing about it,” said Megan Franke, an education professor at the University of California, Los Angeles. “These same things are true for teachers. … If we’re going to work on teacher education, we have to find ways that teachers, pre-service teachers, are having opportunities to learn in and with communities.”
For example, Marder co-founded , a secondary science teacher preparation program now used at 43 universities, with a statewide initiative set to launch in Alabama. The program creates partnerships between colleges of science and education, as well as with school districts. These partnerships actively recruit science and engineering and mathematics majors into STEM-specific degree programs and student teaching opportunities.
Randomized controlled evaluations of UTeach have found that new science teachers who completed the program were more likely to continue teaching than their peers who did not participate in the program. Moreover, UTeach educators’ students had higher academic outcomes than students of other new teachers. The benefits were about on par with academic gains seen for students of Teach for America participants.
Choose mentors and coaches carefully and support them
Several educators noted the expansion of local teacher residencies and district partnerships with higher education teaching programs to boost the number of home-grown science teachers.
However, Laura Booker, the executive director of the Tennessee Education Research Alliance at Vanderbilt University, said schools often do not put enough planning into which veteran teachers work with the program. Tennessee’s developed resources to help districts choose mentors that were most effective in improving teachers’ instructional practice.
“Rather than just asking people to raise their hands and volunteer to have a student teacher, districts and schools and educator preparation programs were being more thoughtful and strategic about who was being selected to serve as mentors,” she said. A randomized controlled evaluation of the program found districts that used the more thoughtfully prepared lists chose on average more “instructionally effective” mentors and their pre-service teachers reported feeling better prepared than those in districts that did not participate.
Tennessee districts, like many nationwide, have hired a slew of subject-specific instructional coaches in recent years to support new teachers. But most of these coaches receive only the same standard professional development given to teachers: “They weren’t really getting professional development on how to be a coach,” Booker said.
To bolster the coaches’ preparation, the Tennessee alliance recruited an array of teachers with varying levels of experience and content specialties to develop a statewide coaching-the-coaches program. Beginning with math coaches, members recorded their own training sessions with teachers and analyzed them with colleagues to develop clearer protocols to support teachers.
In a study of that effort, Booker said, “we found that coaches had deeper and more specific conversations with teachers in the coaching network, and engaged in more conversations around depth and specificity of the math learning goals and that they improved in their quality of feedback and conversations.”
The researchers also found that students of teachers who had been coached by someone in the network had higher academic achievement in math than students of teachers who had not received such coaching.
