Beyond Teacher Shortages: Structural Turnover and Workforce Instability in New Mexico Schools

Abstract

This study examines educator staffing instability in New Mexico by analyzing certified staffing rosters from the New Mexico Public Education Department (2014–2019) alongside a statewide Teacher Working Conditions Survey (N = 4481). The goal was to identify which working conditions districts can influence and to highlight practical strategies for improving teacher retention. Headcount and vacancy analyses show that instability persisted even during periods of workforce growth: vacancies remained high despite increases in educator numbers, reflecting replacement churn and role-specific shortages rather than an overall teacher supply deficit. Vacancy patterns also fluctuated year to year, indicating a labor market responsive to shocks rather than moving toward stability. Turnover estimates further show that educator loss is structural and cumulative across districts, not episodic. The survey findings indicate that job satisfaction varies by grade band, while years of experience do not, suggesting turnover risk is driven more by organizational context than career stage. District-level regression models support this: compensation, leadership instability, student behavior and discipline conditions, and class size predict both annual and long-term turnover. Time for planning, preparation, and collaboration uniquely predicts long-term retention, while administrative discipline support is more strongly associated with annual exits. Overall, the findings highlight retention—not supply—as the central challenge.

1. Introduction

New Mexico’s educator workforce sits at the intersection of national and local pressures: persistent teacher shortages, uneven student outcomes, and the compounded effects of poverty and historical inequities (Learning Policy Institute, n.d.-c; Legislative Education Study Committee & New Mexico Legislative Finance Committee, 2022). A substantial research stream argues that school staffing problems are driven at least as much by high teacher turnover—teachers leaving schools or the profession—as by shortages in the supply pipeline (Ingersoll, 2001, 2002; Learning Policy Institute, n.d.-d). Ingersoll’s national analyses demonstrate that most “shortages” are not caused by too few licensed teachers, but by high rates of departure driven by working conditions, administrative support, and school climate. More recent national work from the Learning Policy Institute similarly finds that turnover accounts for the majority of annual demand for new teachers, far outweighing retirement or enrollment growth as drivers of hiring needs (Learning Policy Institute, n.d.-d). Research on teacher effectiveness further shows that experience strongly predicts instructional quality, strengthening the case that retention—rather than recruitment alone—is central to workforce stability and student outcomes (Learning Policy Institute, n.d.-b).

New Mexico’s context intensifies these workforce debates. The state’s geographic isolation, cultural and linguistic diversity, and high poverty rates contribute to uneven working conditions across districts (Oakes et al., n.d.). New Mexico’s Educator Equity Profile has documented substantial inequities in staffing, including much higher rates of out-of-field teaching in high-poverty schools than in low-poverty schools (New Mexico Public Education Department, 2015). Vacancy monitoring by the SOAR Center at New Mexico State University further shows persistent instability since 2015, with pronounced shortages in teaching positions and sustained unmet demand for educational assistants and special education personnel (New Mexico State University SOAR Center, n.d.). Federal oversight has recently required NMPED to develop a plan to collect and publicly report data on ineffective, inexperienced, and out-of-field teachers by school type, underscoring that monitoring assignment patterns is now a formal equity obligation (U.S. Department of Education, Office of Elementary and Secondary Education, 2024). At the same time, state policy responses have included increasing required instructional time through HB130 in 2023, and the state has also raised base teacher salaries (via Senate Bill 1 in 2022) and expanded its Teacher Loan Repayment Program, offering up to $6000 per year in student-loan forgiveness for teachers in high-need schools as complements to instructional time mandates. These measures targeted both recruitment and retention incentives (New Mexico Legislature, 2022; New Mexico Higher Education Department, n.d.). However, the effectiveness of these measures ultimately depends on the stability and preparation of the instructional workforce serving those hours (Oakes et al., n.d.; New Mexico Public Education Department, 2015; New Mexico State University SOAR Center, n.d.).

Within this setting, turnover estimates for New Mexico vary substantially by data source and time window. Legislative and external analyses have reported annual turnover near one in four teachers during some periods, placing the state among the highest nationally (Legislative Education Study Committee & New Mexico Legislative Finance Committee, 2022; New Mexico Public Education Department, 2020). At the same time, some internal state-level documents from a 2020 EARS report show a teacher turnover rate of 14.9% from 2018 to 2019; however, this figure is presented without metadata, denominator, or methodological explanation in the report. As a result, it is unclear whether the figure reflects attrition, mobility, or a partial sample and cannot be treated as a validated statewide turnover metric (New Mexico Public Education Department, 2020). The next report does not appear until 2023–24 and does not report any turnover data. This inconsistency underscores the state’s failure to produce transparent, standardized turnover reporting, complicating efforts to monitor staffing stability, assess policy impact, and evaluate equity across districts. Such variation further highlights the need for district-level measures paired with educators’ reports of working conditions.

Theoretical Background:

The field has long diverged on the causes and remedies for teacher shortages. Earlier work emphasized a “supply problem” as demographic shifts, retirement trends, and enrollment growth acted as primary drivers of staffing shortfalls (Darling-Hammond, 1984; Haggstrom et al., 1988). Later research reframed shortages as organizational and systemic problems, arguing that working conditions, leadership quality, preparation pathways, and compensation structures shape “retention” and, therefore, “staffing stability” (Ingersoll, 2001; Darling-Hammond, 2002; Learning Policy Institute, n.d.-a). These perspectives generate different hypotheses as either a supply problem resolved through recruitment or a conditions problem addressed through retention-focused reform. Both frameworks appear in current policy discourse. The costs of turnover, which include financial, instructional, and relational costs, are well documented (Guin, 2004; Watlington et al., 2010), as are equity concerns when vacancies are disproportionately filled by inexperienced or out-of-field teachers in disadvantaged schools (Lankford et al., 2002). More recent research further suggests that teacher retention is shaped less by teacher demographics or career stage than by school- and district-level conditions, including leadership stability, administrative support, workload, professional time, and school climate (Carver-Thomas & Darling-Hammond, 2017; Office of the Governor of New Mexico, 2022; Ingersoll, 2003; Ronfeldt et al., 2013; Kraft et al., 2016; Nguyen et al., 2020). Research on induction and mentoring likewise suggests that these supports may help teacher retention, but are often insufficient when broader organizational conditions remain weak (New Mexico Public Education Department, 2018; New Mexico Legislative Finance Committee, 2023; Legislative Education Study Committee & New Mexico Legislative Finance Committee, 2024).

Research Questions:

• What is the relationship between district-level educator staffing levels and teacher vacancies in New Mexico public school districts?
• How can teacher turnover across New Mexico districts be described and interpreted using available state-level data?
• Are there measurable differences in teacher satisfaction across groups defined by years of experience and grade level?
• Which teaching and working condition indicators are associated with teacher turnover?
• Which workplace conditions appear most actionable at the state and district levels for improving teacher retention?

2. Materials and Methods

2.1. Research Context

New Mexico’s districts and state-authorized charters face persistent staffing strain and uneven working conditions. This study combines district-level teacher turnover (constructed from NMPED staffing rosters from Santa Fe, New Mexico) with an educator working condition survey on perceived satisfaction levels (spring 2020). The goal is to identify district-level, modifiable factors associated with retention rather than to evaluate supply-chain-only solutions.

2.2. Purpose of the Study

This study aims to identify which working conditions are most strongly associated with teacher turnover in New Mexico. By comparing district-level turnover data and educator survey responses, the analysis highlights organizational factors that separate stable districts from those with high turnover. The results are intended to guide retention-focused policy and practice by pointing to conditions that districts can address most effectively.

2.3. Data Sources

(a)

Staffing lists (turnover inputs). NMPED provided the district rosters of certified teaching staff for 2014–2019 via public record requests. State-level teacher headcounts for 2020–2023 were obtained through additional records requested to contextualize trends only.

(b)

Teacher Working Conditions Survey (TWCS). A 34-item working condition survey was administered electronically through a Survey Monkey link to certified educators statewide in spring 2020 (initial March outreach; May reminder). Each item used a 0–100 satisfaction scale, with higher values indicating greater satisfaction. Of 22,073 certified teachers in 2019, 4481 responded (~20%), spanning 85/89 districts, 341 of which were charter school teachers and were excluded from this analysis to maintain a focused examination of state-operated district systems. Charter school data are retained in the full dataset and may support future analyses.

2.4. Analytic Window

The turnover metrics reported as findings are computed for 2014–2019 only. The 2020–2023 figures referenced in the paper are descriptive context and not used in turnover computations presented as results. The survey data were collected in 2020 and analyzed in relation to district turnover patterns observed in 2014–2019.

2.5. Turnover Definitions and Formulas

2.5.1. Operational Definitions

Attrition: Attrition refers narrowly to educators leaving the teaching profession altogether. Because the available roster data do not allow profession exits to be distinguished from movement to another district, attrition cannot be directly measured in this study and is defined here only for conceptual clarity.

Joiner: An educator who appears on the year t + 1 roster but not on the year t roster.

Leaver (district exit): An educator who appears on the year t roster but has no exact first-and-last-name match on the year t + 1 roster.

Mid-year changes: Rosters are annual snapshots; mid-year hires and departures are not modeled.

Scope: Turnover rates are based on the NMPED records for certified personnel. Classifications reflect agency records and were not independently verified. Because the roster files did not include a stable unique educator identifier across years, we cannot distinguish between educators who moved to another New Mexico district and those who exited New Mexico teaching altogether; both are captured as district exits.

Stayer: An educator who appears on both rosters with identical first and last name strings.

Turnover: In this study, turnover refers to educators leaving a school district between roster years, whether by moving to another district or by leaving teaching altogether. Because the roster files do not include a stable educator identifier across years, these forms of exit cannot be separated and are therefore treated collectively as district exits.

Turnover rate (primary): Leavers ÷ headcount, using the year-t certified teacher headcount as the denominator. This denominator was used for comparability across districts, including small-N contexts.

2.5.2. Turnover Formulas

• Annual turnover (year t):

  Annual Turnover_t = #{teachers present in t but absent in t + 1}
            #{certified teachers in t}

  “Exits” include teachers who leave the district (to another district or out of the profession). Denominator is the year-t certified teacher headcount.
• Long-term turnover (t → t + k):

  Long-Term Turnover_t->t+k = #{baseline teachers in t no longer present by t + k}
                 #{baseline certified teachers in t}

2.6. Time Window and District-Size Effects

Turnover estimates are reported over a set multi-year period. Because the working conditions survey was administered in 2019, that year provided the most appropriate annual reference point for comparison with the long-term turnover measure used elsewhere in the study. No cases were excluded or adjusted as outliers. High turnover rates are most often found in smaller districts, where the departure of a few teachers can lead to large percentage changes. These results are interpreted in context rather than as anomalies. Response rates were proportional across district-size categories. Approximately 20% of certified educators responded within both small (≤100 certified staff) and large (>100 certified staff) districts, reducing the likelihood that comparisons by district size reflect sampling imbalance.

2.7. Record Linkage, Cleaning, and Inclusion Rules

To estimate annual teacher turnover, year-to-year rosters were built from publicly available district staff lists. A strict name-matching rule was used: a teacher is counted as a stayer only if their first and last name match exactly across years. Any difference, such as a middle name, suffix, or spelling change, is treated as a non-match, and the teacher is coded as a leaver. This method was chosen because the size and inconsistency of the rosters made manual review impractical, and it reduced the risk of introducing subjective bias. For example, in 2018 to 2019:

Exact match → stayer: “Martinez Mary” (2018) → “Martinez Mary” (2019). No match under strict rule → leaver: “John Seanne” (2018) → “John Seanne Jean” (2019). Although these refer to the same individual, the added middle/second name prevents an exact match, so the 2018 record is coded as a leaver.

2.8. Survey Measures

The TWCS was developed for this study by the first author as a brief statewide instrument using 34 single-item working condition measures. For subgroup analyses, working condition satisfaction ratings were compared by grade band (elementary, middle, and high), experience (1–2; 3–10; and 11+ years), and district chronic turnover tier (above vs. below states average turnover for 2014–2019). Very small subgroups (e.g., multi-grade or certain SPED categories) were excluded from subgroup tests.

For the district-level comparative and regression analyses, 11 focal indicators were selected because they reflected the school- and district-level conditions most consistently correlated with the teacher retention and turnover literature to workplace satisfaction, organizational stability, and teachers’ decisions to remain or leave. These indicators captured four recurring themes in prior research: compensation and workload (salary and compensation, teacher workload, and planning/preparation/collaboration time), leadership and administrative conditions (administrative turnover, administrative support, and teacher decision-making power), school climate and student-facing conditions (parent support, student behavior, and administration’s role in student discipline), and staffing supports or structures (class size and induction and mentoring programs) (Simon & Johnson, 2015; Geiger & Pivovarova, 2018; Johnson et al., 2012; Boyd et al., 2011; Grissom & Bartanen, 2019; Ingersoll & Strong, 2011). This selection is also consistent with the broader review literature identifying supportive leadership, mentoring, collaboration, autonomy, workload, and compensation as recurring factors in teacher retention and attrition (Koerber et al., 2023).

Because each of the 11 focal measures is based on a single survey item rather than a multi-item scale, measures of internal consistency are not appropriate. The TWCS is therefore best understood as a brief survey using face-valid single-item indicators of discrete working conditions rather than as a validated latent-scale measure. This approach covered a broad range of concrete working conditions with low respondent burden, but it also limited the psychometric claims that could be made because the analyzed measures were single items rather than multi-item scales (Diamantopoulos et al., 2012; Dolbier et al., 2005).

2.9. Sampling and Contact Procedures (Survey)

Official educator emails were sourced from NMPED public records for research communication only. The survey was sent to certified teachers with one reminder. Participation was voluntary; no incentives were offered.

2.10. Statistical Analysis

Descriptive statistics were first used to summarize district-level turnover patterns and teachers’ survey responses across the selected Teaching, Working Conditions, and Satisfaction (TWCS) indicators. Two-way analyses of variance (ANOVA) were then used to test for differences in survey responses by grade level, years of experience, and the interaction between these two factors. When the ANOVA showed a meaningful difference, preplanned follow-up t-tests were conducted to identify which groups differed.

Associations between working condition ratings and turnover were also examined using district-level multiple linear regression. In these models, annual and long-term turnover served as the outcomes, and the 11 working condition indicators served as predictors. These analyses help identify patterns of association across districts; the results are interpreted descriptively and do not imply causality.

2.11. Software

All turnover calculations, matching routines, and summary tables were completed in Microsoft Excel. Statistical modeling and hypothesis testing were carried out in Systat 12.0 (Grafiti LLC, 2023). Replication materials include Excel workbooks containing the underlying formulas and de-identified data.

2.12. Ethics

This human-subjects protocol (survey of educators) received approval from New Mexico Institute of Technology’s IRB committee, IRB#2019-08-001 (initial approval date 8 August 2019). The data was analyzed in de-identified form.

A copy of the TWCS and teacher recruitment email can be found in Supplementary Data. De-identified survey responses and the district-level turnover series (2014–2019) will be made available upon request. Because the raw email lists and unredacted teacher rosters contain personally identifiable information, they cannot be publicly shared. However, all processed and de-identified datasets necessary for replication will be accessible.

3. Results

3.1. Teacher Numbers and Vacancies, 2018–2023

From 2019 to 2023, New Mexico districts averaged 22,154 certified teachers, reaching a high of 23,681 in 2023–24. Over this period, teacher numbers varied but resulted in a net increase of 1921 educators (Figure 1).

From 2019 to 2023, districts reported persistent teacher vacancies, even during periods of rising teacher headcounts (Figure 2).

Table 1. District teacher headcount and vacancies by year (2019–2023). The counts reflect district (non-charter) systems only. The rows report total vacancies, total certified teachers, the year-over-year net change in teacher headcount (Δ vs. prior year), and the vacancy rate (vacancies ÷ total teachers).

Starting School Year	2019	2020	2021	2022	2023
Number of Vacancies	644	571	1048	690	751
Number of Teachers in District	22,073	20,988	23,040	21,383	23,681
Net Gains and Losses in NM Educators	313	−1085	2052	−1657	2298
Percentage of Vacancies	2.92%	2.72%	4.55%	3.23%	3.17%

summarizes statewide district teacher vacancies, total teacher headcount, net annual changes in educators, and vacancy rates in New Mexico districts from 2019 to 2023. Over these five years, reported vacancies ranged from 571 to 1048, while total teacher headcount stayed relatively stable between 20,988 and 23,681. Vacancy rates varied from 2.72% to 4.55%, with the highest rates observed in 2021. Vacancy numbers shifted considerably from year to year, without a clear trend. Vacancies dropped by 73 between 2019 and 2020, rose by 477 in 2021, fell by 358 in 2022, and increased by 61 in 2023. Throughout this period, vacancy rates remained above 2.7%. Net annual changes in educator numbers ranged from a loss of 1657 in 2022 to a gain of 2298 in 2023. Importantly, increases in educator headcount may not be related to similar reductions in vacancies. For example, in 2021, the highest vacancy count and rate occurred alongside a net gain of 2052 educators. These year-to-year swings likely reflect multiple overlapping shocks and staffing pressures.

3.2. New Mexico Statewide Teacher Turnover

Teacher turnover in New Mexico was measured using exact name matching and annual district data across 89 school districts. Between 2014 and 2019, over half of district teachers left their positions, resulting in an average long-term turnover rate of 56.6%. In 2019, the average annual turnover rate was 20.1%, and turnover was concentrated in the 10–30% range, which included 72 of 89 districts (80.9%), as shown in Figure 3a. Only six districts (6.7%) reported annual turnover below 10%, while a smaller right tail of districts exceeded 30% annual turnover (11 districts), including three districts above 40%. Because the working conditions survey was administered in 2019, that year provided the most appropriate annual reference point for comparison with the long-term turnover measure used elsewhere in the study.

Across the five-year period (2014–2019), most districts exhibited long-term turnover between 50% and 70%, and 83 of 89 districts (93.3%) fell between 40% and 80% long-term turnover, as shown in Figure 3b. Five districts reported long-term turnover below 40%, while 11 districts fell between 70% and 80%, and one district exceeded 80%. The statewide long-term turnover rate of 56.6% provides a benchmark for interpreting these district-level patterns.

3.3. District Size and Teacher Turnover (Annual and Long-Term)

For the purposes of this study, districts with fewer than 100 certified educators are classified as small, while those with 100 or more are considered large. Of the approximately 89 districts in the state, 56 fall into the small category and 33 are large, making small districts the majority. Over the period from 2014 to 2019, long-term turnover was higher in small districts at 58.8 percent, compared to 53.1 percent in large districts, a difference of 5.7 percentage points. Annual turnover between 2018 and 2019 showed a smaller difference, with rates of 20.7 percent in small districts and 19.1 percent in large districts, a gap of 1.6 percentage points (Figure 4).

Figure 5a–d presents the annual and long-term turnover rates for large and small school districts. The overall pattern of long-term turnover is similar for both groups, with most districts falling between 50 and 70 percent turnover. Smaller districts display greater variability. Their long-term turnover ranges from 21 to 81 percent, and annual turnover spans 0 to 47 percent. In contrast, large districts show a narrower range, with long-term turnover between 43 and 66 percent and annual turnover from 12 to 30 percent. This indicates that small districts experience more fluctuation in staff retention. Some small districts maintained stable staffing over five years, while others faced persistent losses. Large districts tend to cluster near the middle of the range. Although annual turnover rates appear similar across district sizes, small districts are marked by wider variation and more frequent high long-term turnover.

3.4. Survey Descriptives

The survey gathered responses from 4481 certified educators. Most taught at the elementary level, while middle and high school teachers were also well represented (Figure 6a). Notably, over a quarter of respondents did not specify their grade level, resulting in a significant “Unknown” category. Reported grade levels were 32.1% elementary, 16.6% middle, and 21.2% high school, with smaller shares for special education (1.4%) and multi-grade (2.1%). The high rate of nonresponse makes it more difficult to determine the exact distribution across grade levels.

In 2019, educators reported an average of 14.5 years of experience, indicating a workforce with substantial tenure. Early-career teachers with 1 to 3 years of experience represented 11.0% of respondents. Mid-career educators, defined as those with 4 to 15 years in the field, comprised 45.6%. Those nearing retirement, with 16 to 25 years of experience, made up 31.3%, and 12.1% had more than 25 years. Overall, educators with over 16 years of experience accounted for 43.4% of the sample, while new teachers were a much smaller group (11%), a pattern that may be consistent with concerns about limited replenishment into the workforce (Figure 6b).

3.5. Grade-Level and Teacher Experience Differences in Working Conditions (ANOVA and t-Tests)

An analysis of variance revealed that the ratings of working conditions differ notably by grade level, while teacher experience groups showed no statistically significant differences across the working conditions in this survey.

Based on these grade-level differences, follow-up pairwise comparisons to identify which grade bands differed show statistically significant results (

Table 2. Pairwise grade-level contrasts in working conditions satisfaction (cells show direction and significance; “A > B” = higher satisfaction; N.S = not significant; * p < 0.05, ** p < 0.01, *** p < 0.001; N.S = not significant).

Working Condition	Elementary-Middle	Elementary-High	Middle-High
Salary and Compensation	M > E **	H > E ***	N.S
Parent Support	E > M ***	E > H ***	H > M *
Teacher Workload	M > E ***	H > E ***	H > M **
Administrative Turnover	N.S	H > E ***	N.S
Student Behavior	E > M ***	H > E ***	H > M ***
Administration’s Role in Student Discipline	N.S	H > E ***	H > M *
Class Size	N.S	N.S	H > M **
Administrative Support	N.S	H > E ***	H > M *
Planning, Preparation, and Collaboration Time	M > E ***	H > E ***	H > M ***
Teacher Decision-Making Power	N.S	H > E ***	N.S

). Teachers at the elementary, middle, and high school levels reported distinct views on working conditions, as shown in Table 2. In contrast, no substantial differences emerged across experience groups, and induction or mentoring experiences were consistent regardless of grade level or years of service.

Grade-level ANOVA results showed statistically significant differences across several working condition indicators; however, the associated effect sizes were small throughout. The largest effects were observed for student behavior, planning/preparation/collaboration time, parent support, and teacher workload, suggesting that grade-level differences were present but substantively modest.

Although several working conditions differed significantly by grade level, the associated effect sizes were generally small. As prior methodological work has noted, effect sizes in education research are often modest and are best interpreted in context rather than against conventional benchmark cutoffs alone (Kraft, 2020; Evans & Yuan, 2022).

High school teachers reported greater satisfaction than elementary teachers across most working conditions. Satisfaction for planning/prep/collaboration time averaged 69.41% for high school vs. 58.75% for elementary (+10.66 pp), and teacher workload 61.84% vs. 51.80%; (+10.04 pp). High school also exceeded elementary on administrative support (72.85% vs. 67.00%; +5.85 pp) and the administration’s role in discipline (62.31% vs. 57.47%; +4.84 pp). The one area of higher satisfaction for elementary teachers was parent support (58.26% vs. 54.08%; +4.18 pp). Comparing middle and high school, the largest gaps again favored high school on student behavior (61.83% vs. 51.79%; +10.04 pp) and planning time (69.41% vs. 63.91%; +5.50 pp). Compared to elementary teachers, middle school teachers reported higher satisfaction with salary/compensation, workload, and planning time, while elementary teachers rated parent support and student behavior more favorably. The complete table of means and standard deviations for all working conditions across grade levels is available in the Supplementary Materials.

3.6. Satisfaction Scores: High- vs. Low-Long-Term-Turnover Districts (t-Tests)

Table 3. Mean workplace satisfaction scores (and standard deviations) for 11 working conditions in low- and high-turnover districts. (Significant; * p < 0.05, ** p < 0.01, *** p < 0.001; N.S = not significant).

Working Condition	Mean, SD Low Turnover%	Mean, SD High Turnover%	Significance
Salary and Compensation	63.5 (SD = 26.4)	64.3 (SD = 26.2)	N.S
Parent Support	55.1 (SD = 25.7)	53.9 (SD = 27.0)	N.S
Teacher Workload	55.5 (SD = 29.1)	55.4 (SD = 29.1)	N.S
Induction and Mentoring Programs	54.9 (SD = 30.8)	54.0 (SD = 31.2)	N.S
Administrative Turnover	64.0 (SD = 30.9)	57.6 (SD = 32.6)	***
Student Behavior	56.2 (SD = 26.6)	57.3 (SD = 27.8)	N.S
Administration’s Role in Student Discipline	59.2 (SD = 30.1)	57.3 (SD = 31.5)	N.S
Class Size	62.3 (SD = 30.4)	68.6 (SD = 29.5)	***
Administrative Support	69.4 (SD = 29.6)	66.5 (SD = 31.0)	**
Planning, Preparation, and Collaboration Time	62.8 (SD = 29.4)	60.3 (SD = 30.0)	*
Teacher Decision-Making Power	56.0 (SD = 30.0)	54.4 (SD = 30.2)	N.S

summarizes mean workplace satisfaction scores for low- and high-long-term-turnover districts (2014–2019) across 11 working conditions. Low-turnover districts report higher satisfaction with most working conditions, particularly in administrative support and time for planning, preparation, and collaboration. On average, 69.4% of teachers in low-turnover districts are satisfied with administrative support, compared to 66.5% in high-turnover districts. Satisfaction with planning, preparation, and collaboration time is also higher in low-turnover districts, at 62.8% versus 60.3%. Both differences are statistically significant. Teachers in more stable districts also report greater confidence in leadership continuity, with 64.0% expressing satisfaction compared to 57.6% in high-turnover systems. However, teachers in high-turnover districts are more satisfied with class size, at 68.6% compared to 62.3%. This pattern is consistent with earlier findings that smaller, often rural districts tend to have higher turnover and can naturally have smaller class sizes due to geography and enrollment patterns (Monk, 2007; Lankford et al., 2002). Small districts may therefore offer more favorable class sizes even as they struggle with staffing stability. Overall, these results indicate that strong administrative support, stable leadership, and dedicated time for collaboration are common features of districts that retain teachers more effectively (Geiger & Pivovarova, 2018; Johnson et al., 2012; Boyd et al., 2011; Grissom & Bartanen, 2019).

Parallel comparisons using annual turnover (2018–2019) were also conducted and showed a broadly similar pattern, although the set of statistically significant working conditions was not identical.

3.7. Predictors of Annual and Long-Term Turnover (Multiple Regression)

District-level multiple regression models examined both annual and long-term teacher turnover as outcomes, using eleven working condition indicators as predictors summarized in

Table 4. Statistical significance of 11 working condition indicators as predictors of district annual and long-term teacher turnover from multiple regression models (*** p < 0.001, ** p < 0.01, * p < 0.05; N.S = not significant).

Working Conditions	p-Value Annual Turnover	p-Value Long-Term Turnover
Salary and Compensation	***	***
Parent Support	N.S	N.S
Teacher Workload	N.S	N.S
Induction and Mentoring Programs	N.S	N.S
Administrative Turnover	***	***
Student Behavior	***	***
Administration’s Role in Student Discipline	*	N.S
Class Size	***	***
Administrative Support	N.S	N.S
Planning, Preparation, and Collaboration Time	N.S	**
Teacher Decision-Making Power	N.S	N.S

. Both models were statistically significant overall, but the working condition indicators explained only a small share of variance in turnover (annual turnover: R² = 0.025, adjusted R² = 0.023, p < 0.001; long-term turnover: R² = 0.041, adjusted R² = 0.038, p < 0.001). Even so, the significant predictors identify district conditions that appear meaningfully related to turnover, even though turnover is shaped by many overlapping factors beyond those included in the model, including New Mexico’s geographic isolation, high-poverty district contexts, inequitable staffing patterns, differences in district size, and broader labor market and policy shifts (Monk, 2007; New Mexico Public Education Department, 2018).

Differences in pay, leadership stability, student conduct, and class size were significant predictors in both models. Sufficient time for planning, preparation, and collaboration predicted long-term turnover only, indicating that teachers are more likely to stay when they have time to plan and work with colleagues. In contrast, how administration manages student discipline was associated with annual turnover, indicating that discipline practices may influence decisions to leave within a given year. Other factors, such as parent support, teacher workload, induction and mentoring, administrative support, and teachers’ decision-making power, were not significant predictors. Together, the regression results indicate that leadership stability, compensation, student behavior, and class size are the most consistent predictors of teacher turnover, while supports like mentoring and decision-making and autonomy do not independently predict turnover when broader working conditions are taken into account.

4. Discussion

4.1. Teacher Headcounts and Vacancies

This section focuses on the persistent disconnect between teacher headcounts and vacancy rates in New Mexico. The analysis of headcount and vacancy data (Table 1) shows that vacancy levels have remained high, even during years when the number of educators increased. For instance, some years saw net teacher losses, such as the reduction of more than 1500 teachers in 2022, while later years reflected workforce gains. Increases in headcount did not correspond to direct reductions in vacancy rates; instead, high vacancy numbers persisted despite growth in the total number of educators. This ongoing disparity signals a structural gap between the overall supply of teachers and the fulfillment of district-specific staffing needs. The prevalence of replacement churn, where new hires primarily replace departing educators while vacant positions remain unfilled, exacerbates this issue. Supporting this interpretation, the annual Educator Vacancy Reports consistently document chronic shortages in hard-to-staff areas, such as special education and specialized subject assignments (New Mexico State University SOAR Center, n.d.). These reports further illustrate that statewide advances in workforce size can coexist with elevated vacancy counts in specific districts or roles, highlighting the complex, multifaceted nature of New Mexico’s staffing challenges. This misalignment reflects a central lesson from national research: shortages are not only about overall numbers, but also about where and in what roles educators are willing and able to work, especially in settings with lower pay, difficult working conditions, and high-need assignments (Anne Quiroz, 2004).

Year-to-year volatility in vacancies during the studied period indicates that New Mexico’s teacher labor market was responsive to changing pressures, rather than showing movement toward stability. This pattern is visible in the data: vacancies rose sharply between 2020 and 2021 and then partially receded (New Mexico State University SOAR Center, n.d.). Even though the available data do not permit causal factors to be determined, the shifts may be related both to changing staffing conditions during and after the COVID-19 period and to major policy changes in educator compensation, including New Mexico’s 2022 increases to minimum teacher salaries and broader school personnel pay raises (New Mexico Legislative Finance Committee, 2022).

Volatility matters because it is more disruptive than a consistently high vacancy rate: when staffing conditions swing unpredictably, districts cannot plan hiring timelines, staffing assignments, or course schedules with confidence, and vacancies are more likely to be filled late (or not at all), increasing reliance on short-term fixes. Research on teacher hiring shows that late, rushed hiring processes are common and can produce weaker matches and measurable productivity costs, with downstream effects for instruction and student outcomes (Liu & Johnson, 2006; Papay & Kraft, 2016). These conditions disproportionately affect rural and high-poverty districts because they typically operate with smaller applicant pools and fewer qualified candidates for specialized roles, making them less able to absorb staffing disruptions and more likely to experience prolonged vacancies during shock years (Ingersoll & Tran, 2023). As a result, when vacancies spike statewide, these districts are more likely to experience longer vacancy durations, more out-of-field coverage, larger combined classes, and reduced course offerings, even in years when statewide teacher counts rebound (Ingersoll & Tran, 2023; Carver-Thomas & Darling-Hammond, 2017).

It is important to note that the state has reported year-to-year declines in teacher vacancies following the 2021 peak, including a 34% reduction reported in 2022 and an additional 18% decline in more recent vacancy reports (New Mexico State University SOAR Center, n.d.; Office of the Governor of New Mexico, 2022); however, vacancy counts capture open positions at a point in time and do not directly reflect educator movement, replacement churn, or cumulative workforce instability.

Overall, the data suggest that staffing challenges in New Mexico are not just a matter of the pipeline; they reflect deeper structural issues related to retention, role-specific shortages, and the sustainability of working conditions.

4.2. Teacher Turnover as a Structural Feature of New Mexico’s Education System

This study estimated an annual district-level turnover rate of 20.1% for 2019 and a cumulative turnover rate of 56.5% from 2014 to 2019, capturing both exits from the profession and movement between districts. Although statewide averages offer a useful reference point, aggregate numbers do not fully capture staffing stability in New Mexico. For example, turnover across districts is substantial: from 2014 to 2019, district turnover rates ranged from 20% to over 80%, as depicted in Figure 3b. Distribution graphs of small and large school districts reveal that rural districts experienced long-term turnover rates of 20–80%, while some larger districts maintained rates between 40% and 70%. Examining these distributions helps identify whether staffing challenges are isolated or indicate wider, ongoing instability. This distinction is important for understanding why instructional disruption can continue even when overall teacher headcounts temporarily rebound over the studied period.

Annual and long-term turnover rates differ significantly across districts, suggesting that frequent teacher loss and replacement is a consistent feature of the system rather than a temporary labor market fluctuation. Long-term turnover exceeding 50% of certified educators over five-year periods, as seen across New Mexico, weakens institutional memory, collegial relationships, and instructional continuity (Guin, 2004). Ongoing instability also creates substantial organizational and fiscal costs as districts must continually recruit, hire, and retrain educators (Watlington et al., 2010).

Turnover rates by district size (Figure 4) show that statewide averages conceal important differences between small and large systems. Large districts exhibit relatively tight clustering in both annual and long-term turnover, while small districts show much greater variation, ranging from stable staffing to high levels of turnover. Although average differences by size are modest, the higher volatility in small systems suggests that turnover is experienced differently across rural school districts. Small and rural districts can display both high resilience and high instability depending on local conditions (Lankford et al., 2002). Evidence from rural areas suggests that educators with local ties may stay longer in district positions (Ingersoll, 2003) which may explain why some small districts had the lowest turnover in the state. In New Mexico, some small districts may benefit from community attachment and continuity, while others remain vulnerable to high turnover.

In rural districts, the departure of just a few teachers can shift turnover rates by many percentage points. This statistical property does not mean the impact is only a measurement issue. In small districts, each educator often represents a significant share of instructional capacity and may hold unique certifications, course assignments, and relationships that are difficult to replace. As a result, even a small number of departures can create instability, leading to out-of-field teaching, course changes, and late hiring. Research shows these conditions are associated with measurable declines in student achievement and disruption beyond changes in average teacher quality (Papay & Kraft, 2016; Ronfeldt et al., 2013; Kraft et al., 2016). Meta-analyses confirm that turnover is associated with negative effects on student learning and school functioning and is especially consequential in schools already operating with constrained staffing flexibility (Nguyen et al., 2020). These challenges are particularly acute in rural and high-poverty areas, where staffing pipelines are thin and out-of-field coverage is more common, making the loss of experienced educators especially destabilizing for instruction and continuity (Monk, 2007; Ingersoll, 2002).

State legislative materials further reinforce the scale and churn-like nature of New Mexico’s staffing problem, while also illustrating inconsistencies in how teacher turnover has been defined and reported over time. The NMPED Educator Accountability and Reporting System (EARS) has not consistently published standardized turnover metrics; for example, the 2017–2018 report showed turnover changing marginally from 14.8% to 14.9%. Subsequent reports continued to present teacher and administrator headcounts without publishing corresponding turnover rates or methodological detail (New Mexico Public Education Department, 2018). External analyses by the Learning Policy Institute in FY12 and FY13 cited in a 2022 legislative brief placed annual turnover at 24%, among the highest nationally (Watlington et al., 2010). Legislative hearing materials offer further context: a 2023 LFC brief estimated that approximately 3000 teachers leave their school or district each year, and a 2024 LESC- LFC brief noted that roughly 1500 Level 1 teachers exit annually, highlighting particularly high turnover among early-career educators and increased reliance on continuous replacement hiring (New Mexico Legislative Finance Committee, 2023; Legislative Education Study Committee & New Mexico Legislative Finance Committee, 2024). References to “over 15%” annual turnover in a 2024 legislative slide deck lacked technical definition or citation, and this, combined with the 24% turnover for FY12–13, suggests reliance on outdated figures rather than current analytic estimates (Learning Policy Institute, n.d.-c; Legislative Education Study Committee & New Mexico Legislative Finance Committee, 2022; New Mexico Legislative Finance Committee, 2024).

Taken together, the distributional patterns observed in Figure 3 and Figure 4, combined with legislative evidence, indicate that teacher turnover in New Mexico is structural and cumulative rather than episodic. Persistent turnover across most districts, especially among early-career educators, limits the effectiveness of recruitment-only strategies and stresses the need to address the organizational and working conditions that shape whether teachers remain in the profession. This instability predates the COVID-19 pandemic and reflects a long-standing feature of the state’s education system.

4.3. Working Conditions, Job Satisfaction, and Predictors of Teacher Turnover

The analysis of educator survey responses highlights that teacher turnover in New Mexico is closely tied to working conditions. Although the sample reflects a workforce with substantial tenure skewed toward mid-career and late-career educators, years of experience did not significantly differentiate perceptions of working conditions. This finding suggests that dissatisfaction is not concentrated among novice teachers or those nearing retirement, but is shaped primarily by the organizational contexts in which educators work.

Job satisfaction differed by grade level, reflecting structural differences in how teaching is organized across school settings. High school teachers reported higher satisfaction than elementary teachers in planning, preparation, collaboration time, workload, administrative support, and the administration’s role in student discipline. Elementary teachers reported stronger parent support but lower satisfaction with time and workload and middle school teachers occupied an intermediate position, with pronounced dissatisfaction related to student behavior. Together, these patterns imply that turnover risk is influenced less by who teachers are than by how their roles are structured and supported within schools.

Comparisons between high- and low-turnover districts further highlight the importance of organizational conditions. Districts with lower turnover reported higher satisfaction with administrative support, leadership stability, and time for planning and collaboration, all features affiliated with professional continuity. High-turnover districts, by contrast, showed significantly lower satisfaction with leadership continuity, even when class sizes were smaller.

Regression analyses reinforce these descriptive patterns. Salary, administrative turnover, student behavior, and class size were significant predictors of both annual and long-term turnover. Time for planning, preparation, and collaboration predicted long-term turnover specifically, suggesting that teachers’ decisions to remain are shaped by sustained access to collaborative and preparatory time rather than short-term support. In contrast, the administration’s role in student discipline was associated with annual turnover, indicating that immediate decisions to leave may be triggered by day-to-day disciplinary conditions. Induction and mentoring programs were not significant predictors once broader working conditions were taken into account.

Overall, this data shows that teacher turnover in New Mexico is mainly a response to ongoing organizational conditions, not merely individual dissatisfaction or career stage. Working environments distinguished by unstable leadership, constrained planning time, weak disciplinary support, and heavy workloads are systematically associated with higher turnover, regardless of teacher experience. These results correspond with the distributional evidence presented earlier, reinforcing the conclusion that turnover in New Mexico reflects structural features of the education system rather than isolated or short-term labor market disruptions.

4.4. Structural Implications for Educational Sufficiency: Teacher Turnover, Legal Mandates, and Policy Action in New Mexico

The findings of this study have considerable implications for New Mexico’s constitutional obligation to provide a sufficient education, as defined in Yazzie/Martinez vs. State of New Mexico (2019) ruling.

These consolidated cases arose from lawsuits filed in 2014 on behalf of students including low-income, English learner, Native American, and students with disabilities, who argued that systemic deficiencies in staffing, resources, and instructional access were denying them a constitutionally sufficient education. In this sense, the ruling is not only about access to qualified teachers, but about the structural conditions necessary to recruit, support, and retain them over time.

These decisions emphasize that educational sufficiency encompasses more than access to schools or compliance with minimum standards; it requires a stable and qualified educator workforce capable of delivering consistent instruction across grade levels, subject areas, and communities. Persistent teacher turnover, particularly when it is structural, cumulative, and unevenly distributed, undermines this obligation by disrupting instructional continuity, restricting access to experienced educators, and reducing course offerings, especially in rural and high-poverty districts (Ronfeldt et al., 2013; Guin, 2004).

The evidence in this study suggests that teacher turnover in New Mexico reflects persistent organizational conditions rather than short-term disruption alone. These conditions include leadership stability, time for planning and collaboration, workload design, and the consistency and supportiveness of discipline systems. While the rulings emphasize equity, cultural responsiveness, and access to qualified teachers, they also implicitly require attention to the conditions that determine whether educators remain in classrooms long enough to achieve these goals (Ingersoll, 2001; Kraft et al., 2016).

In response to these mandates, New Mexico has introduced policies designed to strengthen the educator workforce, including statutory minimum salary increases, expanded loan repayment programs, alternative licensure pathways, mentoring and induction supports, micro-credentialing and licensure mobility mechanisms, and investments in leadership development (New Mexico Legislature, 2022; New Mexico Higher Education Department, n.d.; New Mexico Public Education Department, 2025; Office of the Governor of New Mexico, 2022). However, the results of this study suggest that pipeline-oriented and individual-support strategies are unlikely to resolve turnover at scale unless they are paired with reforms that directly address the organizational conditions associated with retention. The present findings suggest that efforts such as mentoring, induction, and leadership development are most likely to matter when they operate alongside more stable and supportive day-to-day working conditions.

To date, publicly available federal and state reporting has largely emphasized staffing conditions such as vacancies and hiring difficulty, rather than providing rigorous, statewide evidence on whether these initiatives have reduced teacher turnover or improved staffing equity across districts (Nguyen et al., 2020; U.S. Department of Education Institute of Education Sciences, 2025; Carver-Thomas & Darling-Hammond, 2017). This accountability gap suggests the need for stronger statewide evaluation—such as longitudinal tracking of teacher retention rates and comparative analysis of turnover before and after policy implementation. Additionally, collecting disaggregated data by district and conducting regular program audits could possibly support more transparent and comprehensive reporting.

To summarize, this analysis suggests that achieving educational sufficiency in New Mexico will require aligning current reforms with more sustained attention to the conditions under which teachers work and stay. In this study, leadership stability, class size, student behavior, and salary and compensation were consistent correlates of turnover across both annual and long-term models, while planning, preparation, and collaboration time was especially relevant to long-term turnover. Comparative analyses also showed that lower-turnover districts tended to report stronger administrative support, greater leadership continuity, and more protected professional time.

These findings do not test particular reforms directly, but they do suggest several areas that may merit closer attention in state and district planning, including (1) strengthening leadership stability through clearer district-level expectations and support for principal and superintendent continuity; (2) providing, protecting and prioritizing time for planning, preparation, and collaboration as a retention strategy, particularly in high-turnover systems; (3) strengthening the coherence and fairness of student discipline systems, potentially including restorative or other supportive approaches, to provide teachers more consistent support with student discipline; and (4) developing evaluations that distinguish exits from inter-district moves and measures whether reforms reduce turnover and improve staffing equity over time. Measures in these areas may help advance policy beyond replacement hiring alone and support the educator stability needed for educational sufficiency (New Mexico Poverty Law, n.d.). Because this study is associational, these implications should be interpreted cautiously. Even so, they support a more retention-centered view of educational sufficiency, one that treats workforce stability as part of ensuring consistent access to qualified experienced educators across schools and communities.

Failure to address extreme and persistent turnover may pose not only a workforce concern but also a risk to compliance with the state’s constitutional obligation to provide an adequate education. When severe or persistent turnover remains unaddressed, particularly in districts serving historically underserved students, the state risks failing to meet the constitutional standard articulated in Martinez/Yazzie. Teacher turnover therefore should not be treated as a secondary workforce issue or left entirely to local discretion. Instead, it should be understood as an important part of educational sufficiency, with accountability efforts that extend beyond recruitment alone to include retention, working conditions, and governance.

4.5. Limitations

This study relied on educational staffing rosters provided by the New Mexico Public Education Department through a public records request. Although these records represent official state-reported data for certified educators across districts, the datasets may contain coding inconsistencies, reporting errors, or delays in updates that are outside the researchers’ control. Turnover estimates, particularly long-term estimates, treat any change in the certified staff roster as a turnover event, which may capture both exits from the profession and inter-district movement. While this approach reflects system-level instability, it may overstate turnover.

Exact name matching procedures were used to track educator continuity across years. Although this method provides high confidence in longitudinal matching, it may misclassify some educators due to name changes, formatting inconsistencies, or common names, introducing the possibility of both false exits and false continuities.

The working conditions survey, while statewide in scope, may be subject to response bias. Certain educator groups may be underrepresented, and nonresponse on some items (including grade band) limits precision in subgroup comparisons. Additionally, the cross-sectional design of the survey and the district-level aggregation of working condition indices definitive causal inference. While regression analyses identify significant associations between organizational conditions and turnover, they cannot establish that working conditions caused departures. Future research should employ longitudinal teacher-level tracking, school-level analysis, mixed-methods case studies, and post-policy implementation evaluations to more precisely identify the mechanisms driving teacher retention and turnover in New Mexico.

5. Future Research

The findings of this study point to several priorities for future research and policy evaluation, but certain limitations should be acknowledged. First, the reliance on aggregate data may obscure important within-district variation and individual-level factors influencing turnover, underscoring the need for district-level case studies of systems experiencing chronic turnover—defined here as cumulative losses surpassing 70 percent—to clarify how turnover operates in different organizational contexts. Second, future studies should address the challenge of distinguishing between healthy mobility and harmful churn, particularly in rural and high-poverty districts (Guin, 2004), since the present analysis cannot definitively separate voluntary transfers from destabilizing attrition. Relatedly, incorporating reasons for leaving into district exit procedures or forms could support qualitative analyses of chronic turnover and provide deeper insight into the organizational and personal factors contributing to persistent staffing instability. Third, the absence of fully standardized and transparent turnover metrics limits comparability across districts and over time, suggesting a need for measurement tools that distinguish exits from the profession from inter-district movement. Finally, while this study references recent policy reforms, it cannot directly evaluate their causal impacts, highlighting the need for the rigorous, longitudinal evaluation of policy initiatives—including salary reforms and leadership development investments—to determine whether these strategies meaningfully reduce turnover and improve staffing equity across New Mexico districts.

6. Conclusions

This study suggests that educator staffing instability in New Mexico cannot be understood by vacancy and headcounts alone. Vacancy data show persistent misalignment between total supply and specific staffing needs, even during periods of workforce growth. District-level turnover estimates reveal that educator loss is structural and cumulative, with more than half of certified positions turning over within five years. Distributional analyses further indicate that instability is not confined to a small number of outlier districts but is widespread across systems, with particularly high volatility in smaller and rural contexts. Together, these findings suggest that New Mexico’s teacher workforce challenges reflect ongoing structural churn rather than temporary fluctuations in labor supply alone. Survey and regression analyses reinforce this interpretation.

Turnover is closely associated with organizational conditions, particularly leadership stability, time for planning and collaboration, student behavior, workload pressures, and class size, rather than compensation or years of teaching experience, and retention is associated with school and district-level conditions including salary and compensation, leadership stability, student behavior, class size, and time for planning and collaboration. While some other conditions differed between high- and low-turnover districts, they did not remain significant once the predictors were considered together in the full regression models. Administrative support may appear important by itself, but its relationship to turnover may overlap with other factors, such as administrative turnover or student behavior.

These patterns align with national research and underscore that recruitment-focused strategies, while necessary, are insufficient to ensure sustained workforce stability. In light of the Martinez/Yazzie ruling, stabilizing the educator workforce is not solely a personnel objective but an important part of educational sufficiency. Ensuring consistent access to qualified educators across grade levels, subject areas, and communities will require sustained attention to the structural conditions that shape whether teachers remain in New Mexico’s classrooms.