Everyone in K-12 knows there's a staffing problem. But "teacher shortage" is actually a sloppy term. It makes it sound like there's one faucet and it stopped running.
The reality is more like a garden hose with holes punched in it every two feet. People leak out at every stage, from the moment someone considers becoming a teacher, all the way through their fifth year in a classroom. Researchers call this the teacher pipeline, and it has six distinct stages where potential and current teachers are lost.
We dug into the peer-reviewed empirical research (meta-analyses, randomized experiments, quasi-experimental policy evaluations) to understand what's actually proven to work at each stage. Not opinions. Not vendor pitches. Evidence.
Here's what we found.
- The six stages of the teacher pipeline
- Retention: where the strongest evidence lives
- Recruitment: two different problems
- Preparation: who mentors you matters
- Licensure: screens have side effects
- Hiring: the process itself creates inequity
- Induction: helpful, but not a cure-all
- What districts should do right now
Most districts focus almost all their energy on one or two of these stages, usually recruitment ("we need more applicants!") or retention ("we keep losing people!"). But the research is clear: you have to think in systems. A great recruitment campaign means nothing if your hiring process screens out diverse candidates, your induction program is a formality, and your working conditions push people out within three years.
Retention Is Where the Evidence Is Strongest, and Most Actionable
Let's start at the end of the pipeline, because that's where the research is deepest and the implications are most immediate.
A large-scale meta-analysis synthesizing over 100 U.S. studies on teacher attrition1 found that the strongest factors associated with teachers staying or leaving aren't salary, prestige, or passion. They're organizational: the day-to-day conditions of the job.
The four biggest retention levers, ranked by effect size:
Mentoring programs: 45% lower odds of leaving.
Collaboration and professional networks: 37% lower odds of leaving.
Competitive salary (mid-to-late career): 34% lower odds of attrition for teachers with roughly 6-30 years of experience.
Administrative support: 21% lower odds of leaving.
These aren't small differences. A mentoring program cutting attrition odds nearly in half is a massive effect by any standard in social science research.
And the research on targeted compensation is equally clear, even if more nuanced. North Carolina implemented an $1,800 annual bonus for certified math, science, and special education teachers in high-poverty or low-performing secondary schools. A quasi-experimental evaluation2 found it reduced turnover for those targeted teachers by about 17%, with experienced teachers responding most strongly.
A separate regression-discontinuity study in Tennessee3 found that retention bonuses tied to teacher effectiveness ratings in priority schools improved retention for tested-subject teachers, though results were imprecise and didn't generalize evenly across all teacher groups.
The takeaway: money matters, but it matters most when it's targeted. Aimed at specific shortage fields, in specific high-need schools, for specific teacher profiles. Across-the-board raises are expensive and diffuse. Surgical incentives produce results.
Recruitment: It's Two Different Problems
The research reveals that "recruitment" is really two distinct challenges that require different solutions.
Problem 1: Interest and Identity
Qualitative studies, like a case study of a dual-enrollment pathway for prospective teachers of color,4 show that early exposure, identity-affirming experiences, and perceptions of teaching as a viable career path shape whether people even consider becoming teachers. These studies provide credible design insights for "grow your own" programs, but they haven't tracked participants all the way through licensure and employment. We know these programs can spark interest; we don't yet have hard numbers on conversion rates through the full pipeline.
Problem 2: Financial Friction
This one has causal evidence behind it, and the findings are striking. A randomized field experiment with over 6,000 prospective teachers admitted to a large placement program5 found that small amounts of upfront financial support (essentially liquidity, not scholarships) meaningfully increased entry into teaching among financially constrained candidates.
Key finding: Among the highest-need candidates, the researchers estimated that every additional $100 in upfront support increased the probability of starting teaching by roughly 1.8 percentage points and increased the probability of completing a two-year commitment by about 1.5 percentage points.
This matters because it reframes recruitment. Many potential teachers aren't disinterested. They're broke. They can't float the gap between accepting a position and getting paid, or they can't cover relocation costs. Solving that friction with bridging funds, relocation stipends, or redesigned financial aid timing is a concrete, evidence-backed lever that districts can actually pull.
Preparation: Who Mentors You Shapes Who You Become
Research on teacher preparation highlights one finding that has direct implications for how districts think about cooperating teachers and student teaching placements.
A study of roughly 2,900 preservice teachers and 3,200 cooperating teachers6 found that the cooperating teacher's observation-based instructional ratings predicted the preservice teacher's later observation ratings once they became teachers of record. The effect was on the order of a 0.07-0.10 increase for each 1-point increase in the mentor's rating.
Here's the twist: the cooperating teacher's value-added scores (student test score gains) did not consistently predict the preservice teacher's later value-added. So the metric you use to identify good mentors matters, and the intuitive choice (student test gains) isn't necessarily the right one.
On the residency model, an evaluation of the Boston Teacher Residency7 found that residency graduates were retained at significantly higher rates through year five (roughly 75% versus 51% for other novice hires) and eventually outperformed comparison teachers in math by about 0.07 standard deviations by years four and five, though early-career math performance was actually lower, suggesting a ramp-up period.
Licensure: A Gate That May Not Filter What You Think
Licensure testing is one of the most politically contested parts of the pipeline, and the evidence reflects that complexity.
A study of edTPA (a performance-based licensure assessment) in Washington State8 found that passing was associated with a 15.2 percentage point higher probability of entering public school teaching the following year. So the test clearly functions as a gateway. It shapes who enters the profession.
But whether it shapes who enters for the better is less clear. Links between licensure test performance and classroom effectiveness were imprecise, partly because of selection effects (you can only measure effectiveness for people who actually get hired into tested subjects).
At the policy level, cross-state evidence9 suggests that basic-skills testing requirements are associated with about 3.4% higher teacher wages in some analyses, consistent with supply restriction raising the price of labor. But the same research found limited evidence that testing improved teacher quality using available measures, and flagged a troubling pattern: testing requirements were associated with roughly a 2-percentage-point reduction in the share of new Hispanic teachers entering the profession.
The implication: licensure screens need to be paired with support. Test prep, coaching, funded clinical experiences. And districts and states need to track demographic impacts, not just pass rates.
Hiring: The Process Itself Creates Inequity
Two rigorous studies reveal that hiring and placement decisions often have less to do with teaching ability than we'd like to believe.
A randomized resume audit experiment10 (6,000 fictitious applications sent to real schools) found that public schools responded more favorably to applicants from more selective colleges, while private and charter schools showed different patterns. In other words, the name on the diploma was doing meaningful screening work, independent of whether it predicts teaching effectiveness.
A structural analysis of teacher-school matching using administrative data11 found that employers tend to prefer candidates with stronger academic credentials and geographic proximity, while teachers prefer schools closer to home, in suburban settings, with lower-poverty student bodies. Critically, these preference patterns differ by teacher race, meaning that the "natural" equilibrium of the labor market tends to reproduce inequitable distribution of teachers across schools.
For districts, this is a call to audit your hiring signals. Are you screening on things that predict performance, or on things that feel credible? Are your processes systematically advantaging candidates who are already proximate to less-needy schools?
Induction and Mentoring: Positive, but Don't Overpromise
Induction and mentoring programs for new teachers are widely recommended and intuitively appealing. The evidence supports them, but with important caveats.
A meta-analysis covering studies from 2010-201912 found small but statistically significant positive effects of formalized induction and mentoring programs on teacher outcomes like retention and efficacy. But "program comprehensiveness" (whether the program included multiple components like mentoring, reduced load, and collaboration time) didn't reliably predict larger effects. More isn't automatically better.
Even more sobering: a large randomized trial,13 one of the strongest study designs in the evidence base, found that a comprehensive induction program improved student achievement but did not improve teacher retention. Teachers got better at their jobs but left at the same rate.
Translation for district leaders: Induction programs are worth investing in. But if your goal is specifically retention, induction alone isn't enough. You need to pair it with the organizational conditions (admin support, collaboration, competitive pay) that the retention research points to.
What Districts Should Do Right Now
Based on the cumulative evidence, here are the highest-confidence actions for K-12 leaders who want to stop the leaking:
Mentoring programs, time for collaboration, and responsive administration aren't "soft" perks. They're the strongest empirical predictors of whether teachers stay.1 Assign every new teacher a trained mentor. Build recurring collaboration time into the weekly schedule. Survey teachers on admin responsiveness and act on the results.
Don't spread bonus dollars thin. Direct incentives to specific shortage fields (math, science, special education) in specific high-need schools. North Carolina's $1,800/year bonus for the right teachers in the right schools reduced turnover 17%.2 Identify your hardest-to-fill positions and design retention bonuses for those roles specifically.
If candidates can't afford the gap between accepting a position and getting paid, they won't show up. A randomized experiment showed that even modest upfront financial support ($100 increments) meaningfully increased whether candidates started and stayed in teaching.5 Offer bridging funds, relocation stipends, or accelerated first paychecks for new hires in shortage areas.
A resume audit experiment found that public schools favored applicants from selective colleges, independent of teaching ability.10 Check whether your screening criteria (college name, GPA cutoffs, geographic proximity) actually predict classroom performance. Replace credential-based filters with structured interviews and teaching demonstrations.
Most districts measure vacancies (hiring) or turnover (retention) but don't connect the dots from recruitment through five-year retention. Build a dashboard that tracks candidate-to-hire conversion, first-year attrition, and five-year retention by subject area and school. You can't fix a system you're measuring in fragments.
Licensure testing requirements were associated with a 2-percentage-point reduction in new Hispanic teachers entering the profession.9 Hiring preferences and placement patterns also differ by teacher race.11 Disaggregate your pipeline data by race and ethnicity at every stage: application, licensure, hiring, placement, and retention.
The Bottom Line
The teacher pipeline isn't mysterious. It's a system with measurable inputs, identifiable leakage points, and evidence-backed interventions. The research tells us that the biggest, most actionable levers are working conditions, early-career supports, and targeted compensation. Not just "recruit harder."
Districts that treat staffing as a systems problem, measuring each stage, targeting interventions where the evidence is strongest, and tracking equity impacts across the pipeline, will fill classrooms more reliably than those chasing one-off fixes.
That's the kind of problem we work on every day at Spur.
References
- Nguyen, T. D., Pham, L. D., Crouch, M., & Springer, M. G. (2020). The correlates of teacher turnover: An updated and expanded meta-analysis of the literature. Educational Research Review, 31, 100355.
- Clotfelter, C. T., Glennie, E. J., Ladd, H. F., & Vigdor, J. L. (2008). Would higher salaries keep teachers in high-poverty schools? Evidence from a policy intervention in North Carolina. Journal of Public Economics, 92(5-6), 1352-1370.
- Springer, M. G., Swain, W. A., & Rodriguez, L. A. (2016). Effective teacher retention bonuses: Evidence from Tennessee. Educational Evaluation and Policy Analysis, 38(2), 199-221.
- Curci, J. D., Johnson, J. M., Terrero Gabbadon, A., & Wetzel-Ulrich, E. (2022). Expanding the pipeline to teach: Recruiting future urban teachers of color through a dual enrollment program. The Urban Review, 55(2), 224-243.
- Coffman, L. C., Conlon, J. J., Featherstone, C. R., & Kessler, J. B. (2019). Liquidity affects job choice: Evidence from Teach For America. The Quarterly Journal of Economics, 134(4), 2203-2236.
- Ronfeldt, M., Brockman, S. L., & Campbell, S. L. (2018). Does cooperating teachers' instructional effectiveness improve preservice teachers' future performance? Educational Researcher, 47(7), 405-418.
- Papay, J. P., West, M. R., Fullerton, J. B., & Kane, T. J. (2012). Does an urban teacher residency increase student achievement? Early evidence from Boston. Educational Evaluation and Policy Analysis, 34(4), 413-434.
- Goldhaber, D., Cowan, J., & Theobald, R. (2017). Evaluating prospective teachers: Testing the predictive validity of the edTPA. Journal of Teacher Education, 68(4), 377-393.
- Angrist, J. D., & Guryan, J. (2008). Does teacher testing raise teacher quality? Evidence from state certification requirements. Economics of Education Review, 27(5), 483-503.
- Hinrichs, P. (2021). What kind of teachers are schools looking for? Evidence from a randomized field experiment. Journal of Economic Behavior & Organization, 186, 395-411.
- Boyd, D., Lankford, H., Loeb, S., & Wyckoff, J. (2013). Analyzing the determinants of the matching of public school teachers to jobs: Disentangling the preferences of teachers and employers. Journal of Labor Economics, 31(1), 83-117.
- Keese, J., Thompson, C. G., Waxman, H. C., McIntush, K., & Svajda-Hardy, M. (2023). A worthwhile endeavor? A meta-analysis of research on formalized novice teacher induction programs. Educational Research Review, 38, 100508.
- Glazerman, S., Isenberg, E., Dolfin, S., Bleeker, M., Johnson, A., Grider, M., & Jacobus, M. (2010). Impacts of comprehensive teacher induction: Final results from a randomized controlled study (NCEE 2010-4027). Washington, DC: Institute of Education Sciences, U.S. Department of Education.
Struggling with Staffing Gaps?
Spur Education partners with 5,000+ schools to fill classrooms with qualified educators, with a 92%+ fill rate. Let's talk about your pipeline.
Get in Touch →
