Skip to Content
MDPIMDPI

Advanced

Behavioral SciencesBehavioral Sciences
PDF
  • Article
  • Open Access

26 October 2026

19 Pages

More than Class Size: Support and Autonomy Predict Pedagogical Commitment in Early Childhood Teachers

and
Department of Psychology, University of Virginia, Charlottesville, VA 22904, USA
*
Author to whom correspondence should be addressed.
Behav. Sci.2026, 16(11), 2003;https://doi.org/10.3390/bs16112003
This article belongs to the Section Educational Psychology
Version Notes
Order Reprints

Abstract

High teacher turnover in early childhood education underscores the need to understand what sustains teachers’ commitment to their work. Although policy often centers on structural factors like class size and teacher experience, organizational support and autonomy may matter more. Using survey data collected from 143 early childhood teachers in 2022–2023 for a randomized controlled trial of preschool outcomes, we compared how workplace support, teacher beliefs, and structural factors predicted teachers’ intention to continue teaching their current pedagogical approach. Roughly half of the responding teachers taught in public Montessori schools, and program type (Montessori vs. comparison) was a predictor in all models. Three-level hierarchical mixed-effects models showed that experience and class size had negligible associations with pedagogical commitment, whereas organizational support and pedagogical autonomy each predicted it (βs = 0.21–0.31). Montessori teachers reported higher pedagogical commitment than comparison teachers, a difference not accounted for by perceived support or autonomy. An exploratory analysis of teacher beliefs was limited by sample size and was not conclusive. Findings suggest that strengthening early childhood teachers’ sense of support and autonomy may support pedagogical commitment, with implications for early childhood education leadership and policy.

1. Introduction

Early childhood teachers are leaving the field at alarming rates in the United States (Bassok et al., 2021; Kwon et al., 2025), disrupting program continuity, weakening relationships with children and families, and diminishing educational quality (Bryant et al., 2023; Cassidy et al., 2011; Kini & Podolsky, 2016). However, efforts to slow the exodus by improving structural conditions (smaller classes, for example) have largely failed (Bryant et al., 2023). This raises the question of what sustains teachers in their work. One possibility is teachers’ commitment, not only to their schools or their students, but to their pedagogical method. Here we examine whether such a pedagogical commitment depends more on psychological factors like autonomy and support or structural factors like class size and years of teaching experience.
We begin by defining pedagogical commitment, then consider the psychological factors that undergird commitment, including the possible contribution of teacher beliefs. Next, we consider why it is interesting to explore teacher commitment in the Montessori education setting as well as other early childhood settings, before introducing the present study.

1.1. Pedagogical Commitment

Employees may hold multiple commitments at once (Meyer et al., 2021); for teachers, this could include commitments to various targets such as their school, principal, coworkers, the teaching profession generally, and/or the specific pedagogical method they use (Firestone & Pennell, 1993; Klein et al., 2022). Commitment research (including research on teachers) has focused almost exclusively on external referents such as the school, the profession, or students (Tyree, 1996); commitment to a specific pedagogy has received little empirical attention. This gap is notable given that teachers in specialized training programs undergo philosophical preparation that may cultivate identity-based attachment to a particular approach, above and beyond broader professional or organizational commitment (Herman et al., 2023).
Here we focus on pedagogical commitment, operationalized as a teacher’s intention to continue teaching the pedagogy they presently teach. Consistent with Meyer and Herscovitch’s (2001) general model of commitment, pedagogical commitment is conceptualized here as behavioral intention, driven by identification with and investment in one’s pedagogical approach (Herman et al., 2023; Li & Ma, 2025). Commitment to a pedagogy, while related conceptually to other forms of teacher commitment, should be understood as distinct; a teacher may leave a school while continuing to teach the same pedagogy, or remain in a school while abandoning the approach they were trained in.
Intentions of this kind have been shown to predict later behavior: teachers’ intentions to leave are among the strongest predictors of actual attrition (Madigan & Kim, 2021), and commitment has been linked to student achievement (Park, 2005), making pedagogical commitment a plausible but untested option for similar attention.

1.2. What Undergirds Commitment

Recent theory and empirical research suggest that psychological factors, such as teachers’ pedagogical autonomy and their sense of administrative and peer support at work, may play a more central role in commitment than structural conditions (Fradkin-Hayslip, 2021; Klassen & Chiu, 2010; Lovison & Mo, 2024; Moss & Wheeler, 2024; Zhao & Jeon, 2024). Self-Determination Theory (SDT; Ryan & Deci, 2000) provides a useful framework for understanding why. SDT proposes that human motivation and wellbeing stem from satisfying three fundamental psychological needs: autonomy (experiencing choice and control), competence (feeling effective), and relatedness (feeling connected to others). Applied to teachers, SDT implies that commitment depends less on working conditions and more on whether those conditions support these three fundamental needs.
Research across K-12 settings supports this prediction. Teachers who experience greater autonomy (Fradkin-Hayslip, 2021; Nguyen et al., 2026; Toropova et al., 2021) and organizational support (Boyd et al., 2011; Grissom, 2011; Shogbesan et al., 2024; Zhao & Jeon, 2024) report higher job satisfaction, lower burnout, and greater commitment. In addition, teachers who feel supported and autonomous provide higher-quality care and form more positive relationships with children (Markowitz et al., 2024; Whitaker et al., 2015). In contrast, structural factors such as class size and years of experience show inconsistent or weak relationships with commitment (Bowne et al., 2017; Fresko et al., 1997; Toropova et al., 2021). Despite this, policy and public discussions often focus on structural factors (Friedman-Krauss et al., 2026).

1.3. The Role of Teacher Beliefs

Teachers’ pedagogical beliefs may also shape motivational experiences pertinent to Self-Determination Theory. When teachers believe strongly in a particular instructional approach, they may experience greater autonomy, competence, and meaning when implementing that approach, reinforcing commitment. Every teacher holds implicit or explicit beliefs about how children learn, what constitutes effective teaching, and whether their approach aligns with children’s needs (Fives & Buehl, 2012; Pajares, 1992). Teachers with student-oriented beliefs emphasize child agency, exploration, and discovery learning, viewing the teacher as a facilitator; teachers with teacher-oriented beliefs emphasize direct instruction, structured lessons, and teacher control (De Charms, 1976; Kirschner et al., 2006; Loveless, 2001; Rosenshine, 2012; Ryan & Grolnick, 1986).
Emerging evidence suggests that such pedagogical beliefs may predict teacher commitment. Teachers who hold strong convictions about their teaching method and who believe their approach is right for children may experience greater meaning in their work, which in turn could sustain commitment (Herman et al., 2023; Klassen & Chiu, 2011; Li & Ma, 2025; Miller & Youngs, 2021). Conversely, teachers who feel their pedagogical beliefs conflict with school expectations or policy mandates may experience dissatisfaction and disengagement (Valli & Buese, 2007; Zang & Chen, 2022).

1.4. Montessori Education as an Ideal Context

Montessori education provides a prime case for examining these questions: it is the most common alternative pedagogy in the world, implemented in over 16,000 schools (Debs et al., 2022), including about 600 public Montessori schools in the United States (Suchman, 2026). We use “alternative” to refer to pedagogical approach rather than institutional status; Montessori differs from conventional curricula in its instructional model while also operating within public school systems.
Beginning over 100 years ago with atypically developing, and then low-income children in Rome (Montessori, 1964), Montessori today serves diverse populations including many low-income and minority children (Ansari & Winsler, 2014; Debs, 2016). Montessori classrooms are characterized by highly student-centered pedagogy, featuring mixed-age groupings (ages 3–6), large class sizes (often 25 or more children), one lead teacher and one assistant, and an emphasis on child-directed learning using hands-on materials (Lillard & McHugh, 2019a). Within this environment, teachers facilitate learning through one-on-one or small-group lessons, but they generally expect children to work independently with materials (Montessori, 1964).
Montessori teachers receive specialized training that emphasizes a coherent pedagogical philosophy (Cossentino, 2006; Lillard & McHugh, 2019b; Whitescarver & Cossentino, 2007). This training may cultivate more student-centered beliefs compared to what is cultivated in traditional programs. At the same time, public Montessori programs operate under the same state standards, accountability pressures, and resource constraints as traditional schools, making public Montessori school teachers a strong test of whether psychological factors matter independent of structural conditions (Benson et al., 2024; Culclasure et al., 2018).
Despite increasing attention to Montessori education in public contexts (Lillard et al., 2025; Randolph et al., 2023), empirical studies of public Montessori teacher commitment remain limited. Culclasure et al. (2018) evaluated South Carolina’s public Montessori programs and found high levels of teacher satisfaction, which the authors attributed to autonomy, child-centered practice, and collegial support. More recently, Moss and Wheeler (2024) examined beliefs about autonomy and control among Montessori and traditional public school teachers, and found that Montessori educators more strongly endorsed autonomy-supportive and student-centered beliefs, factors that have been linked to higher commitment (Choi et al., 2018; Löfström et al., 2026). Another study of Montessori and traditional teachers found that Montessori teachers with higher levels of teacher autonomy, a crucial aspect of SDT, experienced lower levels of burnout (Benson et al., 2024), which may predict commitment to staying in the profession. In sum, Montessori education presents a good test for examining whether psychological and structural factors better predict teacher commitment, using the motivational lens of Self-Determination Theory.

1.5. The Present Study

This study examines predictors of pedagogical commitment using two years of survey data (2022–2023) from 143 early childhood teachers in public Montessori and comparison programs. Our research questions are:
  • Do structural factors (class size, years of teaching experience) or psychological factors (organizational support, pedagogical autonomy) better predict pedagogical commitment?
  • Do teacher beliefs add predictive power?
We hypothesize that (1) psychological factors will predict commitment beyond structural factors, and (2) teacher beliefs will add predictive power for commitment beyond the psychological factors.
One feature of the study context warrants note. Data collection began in fall 2021 and continued through 2023, in the immediate aftermath of the COVID-19 pandemic’s disruption to schooling, a period that could plausibly bear on pedagogical commitment. Because our two waves fall at different points in this recovery, we include survey year as a covariate in all models to account for potential time-related trends. As reported below, commitment did not in fact differ across waves.

2. Materials and Methods

This study analyzed teacher survey data from a multi-year randomized controlled trial examining public Montessori preschool programs (Lillard et al., 2025). The teachers taught at least one of the 588 children in that study, which launched in 2021 with families who applied for a PK3 seat in one of 24 public Montessori programs in 10 US states; each child was randomly assigned via lottery to receive a seat offer (treatment) or was placed on a waitlist (control). Waitlisted families (n = 346) enrolled in comparison schools, which included both traditional public preschools and other alternative programs. Note that although the Montessori schools were oversubscribed, comparisons of third grade test scores with those of other public Montessori schools suggest they were not particularly high-performing among public Montessori programs (Lillard et al., 2025).

2.1. Participants

The participants were 143 preschool teachers (78 Montessori, 65 comparison) in 10 states across the US, teaching in 71 schools. As shown in Table 1, the full sample averaged 41.5 years of age (SD = 11.3); 90.9% self-identified as female and 4.9% as male. Most were White (56.6%) but there was a range of ethnicities; see Table 1 for full demographic breakdown by group. Comparison teachers held more advanced degrees than Montessori teachers (i.e., 63% held an MA vs. 56%). Racial composition also differed: the comparison group included more Black or African American teachers (35.4% vs. 7.7%), while the Montessori group included more Hispanic teachers (17.9% vs. 6.2%).
Table 1. Sample characteristics by program type.
All 78 Montessori teachers taught in public Montessori schools that formed the treatment condition of the parent randomized trial (Lillard et al., 2025). The 65 comparison teachers, from 41 schools, represented a range of curricula, most of which would be considered traditional. The most common were Creative Curriculum (33.8%) and district- or school-designed curricula (23.1%), followed by other approaches (15.4%), including play-based and combined programs. Smaller proportions reported Every Child Ready (9.2%), International Baccalaureate (7.7%), no formal curriculum (4.6%), Frog Street (3.1%), or Reggio Emilia (3.1%).
Classroom age composition differed markedly by program type. Montessori classrooms spanned an average of 3.4 years in the 2-to-7-year range, and 94.9% enrolled children across a full 3-to-5-year-old range; nearly all enrolled 3-year-olds (98.7%), 4-year-olds (100%), and 5-year-olds (96.2%), and 42.3% also enrolled children aged 6 or older. Comparison classrooms served narrower age-bands, averaging a 2.0-year span, with only 13.8% spanning ages 3 to 5; they concentrated on 4-year-olds (86.2%), enrolled 3- and 5-year-olds less often (67.7% and 38.5%, respectively), and never enrolled children aged 6 or older. Few classrooms in either group enrolled children under age 3 (2.6% Montessori, 6.2% comparison), confirming that teachers in both groups primarily taught preschool-aged children.

2.2. Measures

The survey, administered on Qualtrics, contained 106 items assessing teacher demographics, working conditions, pedagogical beliefs, and classroom practices. The responses analyzed here were selected based on our theoretical frameworks (see Introduction), with pedagogical commitment as the outcome variable. Predictor variables were the structural factors commonly targeted in retention policies (experience, class size) and psychological factors theorized to support teacher motivation (organizational support, autonomy, and teacher beliefs). Teachers were surveyed in 2022, 2023, and 2024. Analyses use the 2022 and 2023 waves, in which pedagogical autonomy and organizational support were assessed; these measures were not administered in 2024. Survey items related to teacher- versus student-oriented beliefs (discussed below) were only gathered in Year 2 (2023); Table 2 shows which variables appeared each year, ns for that variable, and which analyses each variable was used in.
Table 2. Sample characteristics and variable availability by year and analysis.

Outcome: Pedagogical Commitment

A single item assessed commitment: “I plan to continue teaching the pedagogy I teach” (1 = Strongly Disagree to 5 = Strongly Agree). This item captures teachers’ intent to persist with their current pedagogical approach. Single-item measures of workplace attitudes can demonstrate adequate validity and reliability for assessing global affective responses (Matthews et al., 2022; Nagy, 2002), but the item was included in a survey designed for the parent trial rather than for the present analysis, and it cannot capture additional facets of commitment that a multi-item scale would distinguish.

2.3. Structural Predictors

The three structural predictors were the teacher’s years of experience teaching ages 3–6, the total number of children that teachers reported enrolled in their classrooms (class size), and whether they taught in Montessori or comparison schools (program type).

2.4. Psychological Predictors

Pedagogical autonomy was measured with the item: “I am able to implement good pedagogy while incorporating state standards” (1 = Strongly Disagree to 5 = Strongly Agree). This item captures teachers’ sense of autonomy to enact their teaching approach while also meeting externally imposed standards. This is one facet of autonomy as described in Self-Determination Theory rather than a general measure of the construct, and we refer to it as pedagogical autonomy to reflect that narrower scope.
Organizational support was measured with: “I feel supported, not isolated, by administrators and other teachers” (1 = Strongly Disagree to 5 = Strongly Agree). This item assesses perceived support from colleagues and leadership. We use organizational support to refer to perceived support within the workplace; the item combines administrative and collegial sources, which we cannot separate.
Teacher beliefs, namely student-oriented and teacher-oriented learning orientations, were assessed using the Teachers’ Learning Orientation Scale, a validated two-factor instrument (TLOS; Carroll et al., 2026). The TLOS distinguishes student-oriented beliefs (6 items; e.g., “Students who get to choose and generate their own work tend to learn more,” “Students learn to manage their behavior best when free to make choices about their learning”) from teacher-oriented beliefs (6 items; e.g., “Whole-class instruction is the most effective form of teaching,” “Teachers impart knowledge and skills to students”). All items were rated on a 7-point scale from 1 (Strongly Disagree) to 7 (Strongly Agree). We computed factor scores using CFA on the present sample (see Supplementary Materials). The student-oriented and teacher-oriented factors were moderately negatively correlated (model-implied r = −0.50), indicating related but distinct dimensions. The measurement model was estimated on the 66 teachers who completed all 12 belief items. Two Montessori teachers in the 2023 sample left one and two items unanswered, respectively; their belief scores were multiply imputed in the exploratory regression.

2.5. Control Variable

Survey year. Year of survey completion (2022, 2023) was included as a covariate in all models to account for potential time-related trends, given the proximity of our initial data collection to the COVID-19 disruption in schools.

2.6. Procedure

Preschool teachers of study children were invited to complete annual online surveys administered via Qualtrics in spring 2022 (n = 219) and 2023 (n = 215). Teachers were re-contacted up to three times to request they fill out the survey and were offered $50 compensation. Response rates were 51% in 2022 (111 of 219) and 36% in 2023 (77 of 215).

2.7. Analytic Sample

We applied a sequence of exclusions to the 188 returns (111 in 2022, 77 in 2023) before analysis. Five returns were abandoned surveys (progress ≤ 25%, median completion time approximately 35 s, no usable items). Two returns came from a single respondent whose data were unreliable: both responses were complete straight lines and were internally inconsistent with that teacher’s reported commitment and experience across waves. Three further returns could not be assigned to a program type or linked to a study school or were duplicate teacher-year records resolved to a single row per teacher per year.
These steps retained 178 of 188 returns (94.7%): 110 responses in 2022 and 68 in 2023, from 143 unique teachers across 71 schools. Seventy-five teachers responded in 2022 only, 33 in 2023 only, and 35 (24%) responded in both years and contributed repeated observations. All eligible responses were retained, on the reasoning that each year represents a distinct configuration of working conditions and commitments; results did not differ substantively when only each teacher’s first response was analyzed. Montessori teachers contributed 62% of observations (n = 111) and comparison teachers 38% (n = 67). Belief measures were administered in 2023 only. The analyses involving teacher beliefs therefore use the 2023 cross-section (n = 68 teachers across 40 schools).

2.8. Analytic Strategy

We employed a three-level hierarchical analytic approach to test predictors sequentially, using multiple imputation to handle missing data and mixed-effects models to account for repeated measures. Three-level models were preferred over two-level models for analyses with repeated measures to avoid treating repeat observations from the same teacher as independent, which would artificially inflate precision. For the secondary exploratory analysis, which used all available data in 2023, OLS regression with clustered standard errors was used (this is explained further below). Because 60% of schools in the 2023 sample contributed a single teacher, we estimated standard errors using the bias-reduced CR2 cluster-robust variance estimator with Satterthwaite degrees of freedom. To account for multiple comparisons within each predictor family, Holm–Bonferroni corrections were applied. Continuous predictors were standardized prior to analysis.

2.9. Missing Data

Missing data were minimal (no variable exceeded 2.9% missingness) and were handled with multiple imputation by chained equations (van Buuren & Groothuis-Oudshoorn, 2011) implemented via the mice package in R. We generated 20 imputed datasets for each analysis using predictive mean matching for continuous variables. All available variables were included in the imputation models to maximize information and satisfy the missing-at-random assumption. Each imputation model also included school identifier, so that school membership contributed to the imputation; teacher identifier was not included as a predictor, as the number of teachers approached the number of observations. Pedagogical commitment was itself imputed in the four cases where it was missing; results were substantively unchanged in complete-case models (n = 172). The results were pooled across imputations using Rubin’s rules.

2.10. Primary Analysis: Hierarchical Models (2022–2023, N = 178)

Our first analysis tested the impact of structural and psychological predictors on commitment. We used three-level hierarchical linear mixed-effects models with random intercepts for teachers (Level 2) and schools (Level 3) to account for clustering. The 2022–2023 sample included 178 observations from 143 unique teachers across 71 schools (with 40 singleton schools); 35 teachers contributed observations in both years. Within each step, p-values for the focal predictor family were adjusted using the Holm–Bonferroni method; both unadjusted and adjusted values are reported.

2.10.1. Step 1: Structural Factors

Initial models included only years of teaching experience, class size, program type (Montessori vs. comparison), and survey year as a covariate.

2.10.2. Step 2: Adding Psychological Factors

We simultaneously added pedagogical autonomy and organizational support to test whether psychological factors predicted outcomes beyond structural factors.

2.11. Exploratory Analysis: Full Model (2023 Only, N = 68)

To test the full mixed-effects model in an exploratory analysis, we analyzed 2023 data (N = 68 observations from 68 unique teachers across 40 schools; 24 schools contributing a single teacher) where all predictors were available. Because all observations came from unique teachers (no repeated measures), and because a two-level mixed-effects model was singular (school-level variance was near zero; ICC ≈ 4%), we used ordinary least squares regression with standard errors clustered at the school level, pooled across 20 imputations. p-values for the two belief terms were adjusted using the Holm–Bonferroni method.
All analyses were conducted in R version 4.5.1 (R Core Team, 2025). Multiple imputation was performed using the mice package (van Buuren & Groothuis-Oudshoorn, 2011). Mixed-effects models were estimated using lme4 (Bates et al., 2015) with lmerTest (Kuznetsova et al., 2017) for p-values. The results were pooled using mice::pool(). Effect sizes and confidence intervals were calculated using broom.mixed.

3. Results

We first provide descriptive statistics and correlations for key variables for the full sample, then present Montessori and comparison teacher differences, then proceed through the two analyses just described.

3.1. Descriptive Statistics and Group Comparisons

Descriptive statistics and intercorrelations among study variables, based on the 2023 cross-section (N = 68), are presented in Table 3. Teachers reported high levels of pedagogical commitment (M = 4.46 out of 5), pedagogical autonomy (M = 4.28), and organizational support (M = 3.94). Pedagogical commitment was positively associated with pedagogical autonomy (r = 0.40, p < 0.001) and student-oriented beliefs (r = 0.34, p < 0.01), and negatively associated with teacher-oriented beliefs (r = −0.35, p < 0.01). Organizational support, teaching experience, and class size were not significantly correlated with commitment in this cross-section.
Table 3. Descriptive statistics and intercorrelations among study variables.
Table 4 shows means and differences between Montessori and comparison teachers on key study variables, using each teacher’s first survey response (N = 143). Montessori teachers reported significantly higher intent to continue teaching their pedagogical approach (t(106.10) = 3.50, p < 0.001, d = 0.61), a moderate to large effect. Commitment scores were concentrated at the top of the scale: 67.2% of responses were at the maximum and 86.2% at 4 or 5. This concentration was more pronounced among Montessori teachers (78.7% at the maximum) than comparison teachers (48.5%).
Table 4. Group comparisons between Montessori and comparison teachers.
Montessori teachers worked with significantly larger classes (t(119.50) = 5.66, p < 0.001, d = 0.97) and had significantly more teaching experience (t(137.30) = 2.59, p = 0.011, d = 0.44). There were no group differences in pedagogical autonomy or organizational support (ps > 0.48). Montessori teachers endorsed significantly more student-oriented beliefs (t(50.10) = 2.06, p = 0.045, d = 0.52), while comparison teachers endorsed significantly more teacher-oriented beliefs (t(51.60) = −2.37, p = 0.021, d = −0.59), a moderate effect.

3.2. Primary Analysis: Predictors of Pedagogical Commitment

We examined predictors using three-level hierarchical mixed-effects models with random intercepts for teachers (Level 2) and schools (Level 3) in the 2022 and 2023 samples (N = 178 observations from 143 unique teachers across 71 schools, with 35 contributing data in both years). Predictors were standardized (M = 0, SD = 1) and added sequentially to assess their relative contributions. The results are shown in Table 5.
Table 5. Hierarchical mixed-effects models predicting pedagogical commitment (primary analysis).

3.3. Step 1: Structural Factors Only

Initial models included only structural predictors (years of experience, class size) along with program type and survey year. Experience and class size were non-significant (although program type, included as a covariate, was significant at p = 0.047). These Step 1 results confirm that structural factors alone provide limited explanatory power for commitment. This motivated the examination of psychological predictors in Step 2.

3.4. Step 2 Results: Adding Psychological Factors

We simultaneously added two psychological predictors: pedagogical autonomy (ability to implement pedagogy within state standards) and organizational support (feeling supported by administrators and peers). Adding these two predictors reduced between-teacher variance by 58% (0.226 to 0.095) and between-school variance by 66% (0.069 to 0.024), while residual within-teacher variance was largely unchanged (0.535 to 0.499).
Fixed effects accounted for 5.5% of variance in the structural-only model and 28.8% once autonomy and organizational support were added (Δ marginal R2 = 0.233), and the joint contribution of the two psychological predictors was significant, F(2, 152.69) = 25.06, p < 0.001. The reverse comparison was not significant: adding experience and class size to a model containing the psychological predictors increased marginal R2 by 0.008, F(2, 165.71) = 0.84, p = 0.434.
Both psychological predictors were significant, even after Holm correction, with organizational support showing the strongest association. The Montessori association strengthened in Step 2 (β = 0.398, p = 0.007), indicating that Montessori teachers reported higher commitment even after accounting for their psychological work environment. Structural factors (experience and class size) remained non-significant.

3.5. Sensitivity Analyses

Because the outcome is a five-point item with a substantial ceiling, we refit the primary model as a cumulative-link mixed model with a school random intercept, on complete cases (n = 172 of 178). Conclusions were unchanged: organizational support (OR = 2.24, 95% CI [1.47, 3.42]), pedagogical autonomy (OR = 1.70, 95% CI [1.14, 2.52]), and Montessori program type (OR = 4.41, 95% CI [1.71, 11.36]) predicted higher commitment, while experience (OR = 1.07, 95% CI [0.72, 1.58]) and class size (OR = 1.24, 95% CI [0.77, 2.02]) did not. Both psychological predictors survived Holm correction.
Because some teachers contributed observations in both years and a model with teacher and school random intercepts was over-parameterized (209 effects for 172 observations), we also refit the ordinal model using each teacher’s first response only (n = 137 of 143 teachers, 69 schools; six teachers were missing a covariate on their first response). Organizational support (OR = 3.28, 95% CI [1.91, 5.63]) and Montessori program type (OR = 7.34, 95% CI [2.28, 23.64]) remained significant. The estimate for pedagogical autonomy was similar in magnitude (OR = 1.61) but less precise (95% CI [0.96, 2.69], p = 0.069) and did not survive Holm correction.

3.6. Exploratory Analysis: Full Model

Teaching beliefs (TLOS) were first collected in 2023. To test whether beliefs added explanatory power beyond structural and psychological factors, we analyzed 2023 data where all predictors (pedagogical autonomy, organizational support, and beliefs) were available (N = 68 observations from 68 unique teachers across 40 schools). Because all observations came from unique teachers, we used OLS regression with clustered standard errors at the school level.
Only autonomy remained significant in the full model. Teacher-oriented beliefs were negatively associated with commitment (β = −0.235), but the association did not reach significance (p = 0.057), and the exploratory analysis was not powered to detect effects of this magnitude. Given the reduced sample and the effective degrees of freedom produced by the clustered estimator, this analysis could detect standardized coefficients of approximately 0.44 or larger with 80% power.
Because both belief terms were non-significant despite being associated with commitment at the bivariate level (Table 3), we examined the specification. Collinearity with the psychological predictors was not implicated: variance inflation factors were 1.54 for student-oriented and 1.53 for teacher-oriented beliefs, and all others fell below 1.7. The two belief dimensions were themselves moderately inversely related (r = −0.50 in the measurement model; −0.56 among factor scores), and this held within both Montessori (r = −0.57) and comparison teachers (r = −0.44). The results of the full model are shown in Table 6.
Table 6. Full model predicting pedagogical commitment (exploratory analysis, 2023 survey wave)

4. Discussion

Teacher retention remains a critical challenge in early childhood education in the United States, yet interventions continue to emphasize structural factors, such as class size reduction, salary increases, and recruitment of experienced teachers, with limited success (Bryant et al., 2023; Kwon et al., 2025). Because commitment to one’s pedagogical approach may be one factor influencing teachers’ decisions to stay or leave, we investigated predictors of pedagogical commitment using two years of survey data (2022–2023) from 143 early childhood educators across 178 observations in public Montessori and comparison programs.
Three key findings emerged. First, in the primary analysis, feeling supported by administrators and colleagues and having autonomy to implement pedagogy within state standards were the strongest predictors of pedagogical commitment, with effect sizes substantially larger than any structural factor. Years of experience and class size were not significantly associated with commitment in any model.
Second, teacher beliefs did not add detectable explanatory power for commitment in this exploratory analysis, contrary to our hypothesis, although we had limited power to detect moderate effects. Third, Montessori teachers showed higher pedagogical commitment (d = 0.61) and more student-oriented beliefs (d = 0.52) than comparison teachers. Notably, this commitment difference persisted after accounting for organizational support and autonomy. We discuss each finding in turn, then address practical implications, limitations, and future directions.

4.1. Psychological and Structural Predictors of Commitment

In the primary analysis, organizational support and autonomy were each associated with commitment, and both associations held in complete-case estimation and an ordinal model using all observations. Beyond those analyses, the two predictors differed. Organizational support remained significantly associated with commitment in the one-observation-per-teacher ordinal model, while the estimate for autonomy was similar in magnitude, but not significant. In the 2023 exploratory sample, the pattern reversed: autonomy was significantly associated with commitment and organizational support was not, even at the bivariate level in that sample (r = 0.18; see Table 3). Due to the small sample size and exploratory nature of the 2023 sample, we place more weight on the primary analysis but acknowledge that the relative importance of these two predictors in this study cannot be definitively understood.
These findings align with Self-Determination Theory’s emphasis on autonomy and relatedness as fundamental psychological needs (Ryan & Deci, 2000). Years of teaching experience showed no relationship with commitment in any model. In this study, experienced teachers were no more committed than novices, holding other factors constant. Class size showed little association with commitment; although it was not associated here, class size may affect instructional quality, child outcomes, or teacher stress (Bowne et al., 2017).
Why did structure not matter more in this study? We propose that teachers in large classes may thrive if supported and trusted, while even in small classes, teachers struggle without that support. Small classes and extensive resources may not guarantee need satisfaction, and large classes with limited resources may not preclude it.

4.2. Teacher Beliefs

Contrary to our hypothesis, teacher beliefs did not add detectable explanatory power beyond organizational support and pedagogical autonomy in this study. Two features of the analysis caution against reading this as evidence that beliefs are unrelated to pedagogical commitment. First, the analysis was underpowered for moderate effects; the design could reliably detect standardized coefficients of roughly 0.44 or larger and the teacher-oriented belief coefficient was −0.24, well below that threshold. Second, the two belief dimensions were moderately inversely related (r = −0.50) overall. Entering them simultaneously estimates each net of the other, which may leave little independent variance for either in a sample of this size. We did not test whether beliefs mediate the association between program type and pedagogical commitment, due to the small sample size. Whether student-oriented beliefs carry part of the Montessori difference in commitment is a question this design cannot answer, and a worthy direction for future research in a larger sample.

4.3. Pedagogical Commitment Among Montessori Teachers

Montessori teachers demonstrated substantially more student-oriented beliefs and higher pedagogical commitment than comparison teachers; the commitment difference remained even after accounting for organizational support and pedagogical autonomy. Several explanations for this are possible and point to directions for further research. Montessori teacher training is intensive, philosophy-focused, and identity-building (Cossentino, 2006; Whitescarver & Cossentino, 2007), and teacher identity has been linked to commitment and effectiveness in other contexts (Beijaard et al., 2004; Day et al., 2005; Flores & Day, 2006). Teachers do not just learn techniques in Montessori training; they may internalize a coherent belief system about child development, learning, and the teacher’s role.
In addition, the Montessori and comparison groups differed in other ways that could bear on commitment. Class size and years of teaching experience, while different between the two groups, were included in the models. Teacher age and racial composition were not: Montessori teachers were on average seven years older, and the two groups also differed in their racial make-up.
Finally, teachers who self-select into a specialized pedagogy training program may differ in other unmeasured ways that bear on commitment. We did not measure person–organization fit, training experiences, or teacher identity. The present study shows that the commitment difference is not explained by the working conditions we measured; what undergirds the commitment difference remains an open question.

4.4. Practical Implications

These findings suggest possible strategies for school leaders and policymakers. For school leaders, these findings point to the value of autonomy-supportive leadership. Teachers reported higher commitment when they felt supported and autonomous, suggesting that attention to these conditions, such as scaffolded support for teachers who need it and autonomy for teachers to enact their pedagogy within state standards, may be worthwhile.
For policymakers, these findings suggest that the psychological conditions of teaching deserve attention alongside the structural conditions that retention policy typically targets. Although this study did not show a relationship between class size and commitment, class size may well matter for outcomes we did not measure, including teacher stress, instructional quality, and children’s learning. What our findings do suggest is that structural investment alone may not secure teachers’ commitment to the approaches they practice, and that support and autonomy are worth considering when programs are evaluated.

4.5. Limitations and Future Directions

While we tracked teachers across two years, our key predictors and outcome were measured concurrently within each year. This precludes causal inference: We cannot determine whether (1) support causes commitment or (2) committed teachers perceive more support. Experimental or longitudinal designs with lagged predictors are needed to establish causality.
Another limitation is that we measured pedagogical commitment (intent to continue teaching one’s pedagogical method) rather than actual departure from teaching or schools. While intentions to quit are a well-studied predictor of attrition (Madigan & Kim, 2021), many teachers with high commitment may still leave due to external factors (Gundlach et al., 2024). Longitudinal studies tracking actual turnover and career trajectories would strengthen conclusions about retention.
Another limitation is that our sample came from teachers affiliated with a specific RCT in public preschools, potentially limiting generalizability. Within that sample, teachers already had a considerable amount of teaching experience, suggesting that teachers who were least committed may have left before the study window. The findings therefore may underestimate the role of psychological factors among early-career teachers.
The comparison group was pedagogically heterogeneous, including Creative Curriculum, district-designed curricula, play-based and combined programs, International Baccalaureate, and Reggio Emilia. The contrast is therefore between Montessori and a mixed set of non-Montessori programs rather than between Montessori and a single traditional model, and any differences among comparison approaches are absorbed into that group.
The Montessori–comparison contrast also introduces specificity: findings may not generalize to other alternative pedagogies (e.g., Waldorf, Reggio Emilia) or to comparisons between different traditional approaches. Because the specific pedagogical orientation of programs was not measured (only their curricular model), we cannot directly assess person–organization fit within comparison schools. Our findings establish that the commitment difference is not accounted for by perceived support or autonomy; they do not establish that it reflects greater alignment between teachers and their programs. Future research might examine person–organization fit in various pedagogical models.
Another potential limitation is that we used a single-item construct to measure pedagogical commitment, support, and autonomy in this sample. Although single-item measures have been used in prior studies assessing teacher intentions and demonstrate adequate validity and reliability (Dolbier et al., 2005; Nagy, 2002), these items could not capture the multidimensional aspects of the constructs. All measures also came from teacher self-report, and concurrent self-report bias may have contributed to the reported associations. Future research should incorporate multiple methods, including interviews or observations, to examine these findings.
Pedagogical commitment scores were high on average (M = 4.46, SD = 0.80), suggesting possible ceiling effects. The ceiling was differential: Montessori teachers were at the scale maximum in 78.7% of the responses compared with 48.5% of the comparison responses, leaving less room for variation in the group showing higher commitment. Findings for organizational support and program type were robust to ordinal models in sensitivity analyses, including one restricted to a single observation per teacher; the association for pedagogical autonomy was similar in magnitude but did not reach significance in the single-observation model. Future research could benefit from samples with greater variation in commitment, namely by including more early-career teachers.
We examined only two structural conditions, experience and class size; class size was measured as total enrollment reported by the teacher rather than the number of children present on a given day. We did not assess teacher salary, benefits, commute time, family demands, or other demographic and contextual factors that might influence retention. While our focus on psychological factors was intentional and theoretically motivated, future research should test whether these effects hold when controlling for additional predictors.

5. Conclusions

Among the early childhood teachers in this study, pedagogical commitment was more closely associated with the psychological conditions of teachers’ work than with the structural conditions of their classrooms. Years of experience and class size showed minimal relationship with commitment; instead, teachers in this sample reported greater pedagogical commitment when they worked in autonomy-supportive contexts where they felt supported by their administration and colleagues. Montessori teachers reported higher pedagogical commitment than comparison teachers, and this difference was not explained by support or autonomy; notably, Montessori teachers reported this higher commitment while working with larger class sizes. Although support and autonomy, not class size or years of experience, were associated with pedagogical commitment in this sample, whether these factors bear on teacher retention remains an open question.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/bs16112003/s1, Table S1: TLOS short-form items by subscale; Table S2: Standardized Factor Loadings for the TLOS Short Form in the Present Sample.

Author Contributions

Conceptualization, L.M.G.; methodology, L.M.G.; formal analysis, L.M.G.; investigation, L.M.G.; data curation, A.S.L.; writing—original draft preparation, L.M.G.; writing—review and editing, A.S.L.; funding acquisition, A.S.L.; project administration, A.S.L.; supervision, A.S.L. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by grant 23-10339 from Arnold Ventures to A.S.L., and grant R305A180181 from the Institute of Education Sciences, U.S. Department of Education, to the American Institutes for Research. The opinions expressed are those of the authors and do not represent views of the Institute or the U.S. Department of Education.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of the University of Virginia (protocol number IRB-SBS 6452) and American Institutes for Research BS-IRB (protocol number 00000436; accepted on 22 January 2024).

Data Availability Statement

The data that support the findings of this study are available from the Open Science Framework at https://osf.io/vuc42/ (accessed on 20 October 2026).

Acknowledgments

We thank Virginia Vitiello for help with writing and analyses, the American Institutes for Research for their role in the original RCT study, and Lee LeBoeuf and Abigail Krissinger for help with data collection and cleaning.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Ansari, A., & Winsler, A. (2014). Montessori public school pre-K programs and the school readiness of low-income Black and Latino children. Journal of Educational Psychology, 106(4), 1066–1079. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Bassok, D., Markowitz, A. J., Bellows, L., & Sadowski, K. (2021). New evidence on teacher turnover in early childhood. Educational Evaluation and Policy Analysis, 43(1), 172–180. [Google Scholar] [CrossRef] [Scilit]
  3. Bates, D., Mächler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software, 67, 1–48. [Google Scholar] [CrossRef] [Scilit]
  4. Beijaard, D., Meijer, P. C., & Verloop, N. (2004). Reconsidering research on teachers’ professional identity. Teaching and Teacher Education, 20(2), 107–128. [Google Scholar] [CrossRef] [Scilit]
  5. Benson, J. R., Perry, A. R., & Napoli, A. R. (2024). Autonomy, transformational leadership, and burnout in Montessori and traditional public elementary teachers. North American Journal of Psychology, 26(3), 419–437. [Google Scholar]
  6. Bowne, J. B., Magnuson, K. A., Schindler, H. S., Duncan, G. J., & Yoshikawa, H. (2017). A meta-analysis of class sizes and ratios in early childhood education programs: Are thresholds of quality associated with greater impacts on cognitive, achievement, and socioemotional outcomes? Educational Evaluation and Policy Analysis, 39(3), 407–428. [Google Scholar] [CrossRef] [Scilit]
  7. Boyd, D., Grossman, P., Ing, M., Lankford, H., Loeb, S., & Wyckoff, J. (2011). The influence of school administrators on teacher retention decisions. American Educational Research Journal, 48(2), 303–333. [Google Scholar] [CrossRef] [Scilit]
  8. Bryant, D., Yazejian, N., Jang, W., Kuhn, L., Hirschstein, M., Soliday Hong, S. L., Stein, A., Bingham, G., Carpenter, K., Cobo-Lewis, A., Encinger, A., Fender, J., Green, S., Greenfield, D., Jones Harden, B., Horm, D., Jackson, B., Jackson, T., Raikes, H., … Wilcox, J. (2023). Retention and turnover of teaching staff in a high-quality early childhood network. Early Childhood Research Quarterly, 65, 159–169. [Google Scholar] [CrossRef] [Scilit]
  9. Carroll, C. M., Vitiello, V. E., Garrett, L. M., Daggett, E. D., & Lillard, A. S. (2026). Development and validation of the TLOS: A measure of child-centered and adult-centered learning orientations and its revelations about Montessori versus conventional pedagogy. Journal of Psychoeducational Assessment. Advance online publication. [Google Scholar] [CrossRef] [Scilit]
  10. Cassidy, D. J., Lower, J. K., Kintner-Duffy, V. L., Hegde, A. V., & Shim, J. (2011). The day-to-day reality of teacher turnover in preschool classrooms: An analysis of classroom context and teacher, director, and parent perspectives. Journal of Research in Childhood Education, 25(1), 1–23. [Google Scholar] [CrossRef] [Scilit]
  11. Choi, J., Lee, J.-H., & Kim, B. (2018). How learner-centered teaching is associated with teacher self-efficacy and job satisfaction: Analysis of data from 31 countries. Journal of International Cooperation in Education, 20(2), 67–83. [Google Scholar]
  12. Cossentino, J. M. (2006). Big work: Goodness, vocation, and engagement in the Montessori method. Curriculum Inquiry, 36(1), 63–92. [Google Scholar] [CrossRef] [Scilit]
  13. Culclasure, B. T., Fleming, D. J., Riga, G., & Sprogis, A. (2018). An evaluation of Montessori education in South Carolina’s public schools. The Riley Institute at Furman University. Available online: https://www.furman.edu/wp-content/uploads/sites/195/rileypdfFiles/MontessoriOverallResultsFINAL.pdf (accessed on 10 July 2026).
  14. Day, C., Elliot, B., & Kington, A. (2005). Reform, standards and teacher identity: Challenges of sustaining commitment. Teaching & Teacher Education, 21(5), 563–577. [Google Scholar] [CrossRef] [Scilit]
  15. Debs, M. C. (2016). Racial and economic diversity in US public Montessori schools. Journal of Montessori Research, 2(2), 15–34. [Google Scholar] [CrossRef] [Scilit]
  16. Debs, M. C., de Brouwer, J., Murray, A. K., Lawrence, L., Tyne, M., & von der Wehl, C. (2022). Global diffusion of Montessori schools: A report from the 2022 global Montessori census. Journal of Montessori Research, 8(2), 1–15. [Google Scholar] [CrossRef] [Scilit]
  17. De Charms, R. (1976). Enhancing motivation: A change in the classroom. Irvington. [Google Scholar]
  18. Dolbier, C. L., Webster, J. A., McCalister, K. T., Mallon, M. W., & Steinhardt, M. A. (2005). Reliability and validity of a single-item measure of job satisfaction. American Journal of Health Promotion, 19(3), 194–198. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  19. Firestone, W. A., & Pennell, J. R. (1993). Teacher commitment, working conditions, and differential incentive policies. Review of Educational Research, 63(4), 489–525. [Google Scholar] [CrossRef]
  20. Fives, H., & Buehl, M. M. (2012). Spring cleaning for the “messy” construct of teachers’ beliefs: What are they? Which have been examined? What can they tell us? In APA educational psychology handbook, Vol 2: Individual differences and cultural and contextual factors (pp. 471–499). American Psychological Association. [Google Scholar] [CrossRef] [Scilit]
  21. Flores, M. A., & Day, C. (2006). Contexts which shape and reshape new teachers’ identities: A multi-perspective study. Teaching and Teacher Education, 22(2), 219–232. [Google Scholar] [CrossRef] [Scilit]
  22. Fradkin-Hayslip, A. (2021). Teacher autonomy, motivation, and job satisfaction: Perceptions of elementary school teachers according to self-determination theory. Elementary Education Online, 20(2), 198–205. Available online: https://ilkogretim-online.org/index.php/pub/article/view/2253 (accessed on 20 October 2026). [CrossRef] [Scilit]
  23. Fresko, B., Kfir, D., & Nasser, F. (1997). Predicting teacher commitment. Teaching and Teacher Education, 13(4), 429–438. [Google Scholar] [CrossRef] [Scilit]
  24. Friedman-Krauss, A. H., Barnett, W. S., Hodges, K. S., Garver, K. A., Weisenfeld, G. G., Duer, J. K., & Siegel, J. A. (2026). The state of preschool 2025: State preschool yearbook. National Institute for Early Education Research (NIEER). [Google Scholar]
  25. Grissom, J. A. (2011). Can good principals keep teachers in disadvantaged schools? Linking principal effectiveness to teacher satisfaction and turnover in hard-to-staff environments. Teachers College Record, 113(11), 2552–2585. [Google Scholar] [CrossRef] [Scilit]
  26. Gundlach, H. A. D., Slemp, G. R., & Hattie, J. (2024). A meta-analysis of the antecedents of teacher turnover and retention. Educational Research Review, 44, 100606. [Google Scholar] [CrossRef] [Scilit]
  27. Herman, A. N., Dearth-Wesley, T., & Whitaker, R. C. (2023). The association between early childhood education professionals’ perception of workplace spirituality and their intention to stay in their current program. Early Childhood Education Journal, 53, 299–312. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  28. Kini, T., & Podolsky, A. (2016). Does teaching experience increase teacher effectiveness? A review of the research. Learning Policy Institute. [Google Scholar] [CrossRef] [Scilit]
  29. Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. Educational Psychologist, 41(2), 75–86. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  30. Klassen, R. M., & Chiu, M. M. (2010). Effects on teachers’ self-efficacy and job satisfaction: Teacher gender, years of experience, and job stress. Journal of Educational Psychology, 102(3), 741–756. [Google Scholar] [CrossRef] [Scilit]
  31. Klassen, R. M., & Chiu, M. M. (2011). The occupational commitment and intention to quit of practicing and pre-service teachers: Influence of self-efficacy, job stress, and teaching context. Contemporary Educational Psychology, 36(2), 114–129. [Google Scholar] [CrossRef] [Scilit]
  32. Klein, H. J., Solinger, O. N., & Duflot, V. (2022). Commitment system theory: The evolving structure of commitments to multiple targets. Academy of Management Review, 47(1), 116–138. [Google Scholar] [CrossRef] [Scilit]
  33. Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. (2017). lmerTest package: Tests in linear mixed effects models. Journal of Statistical Software, 82, 1–26. [Google Scholar] [CrossRef] [Scilit]
  34. Kwon, K.-A., Jang, W., Ford, T. G., Horm, D., Yazejian, N., & Bryant, D. (2025). A longitudinal study of Head Start teacher turnover trends and factors. Teaching and Teacher Education, 156, 104916. [Google Scholar] [CrossRef] [Scilit]
  35. Li, G., & Ma, Y. (2025). Exploring the influencing factors of teacher beliefs and their impact on teacher behaviors. BMC Psychology, 13, 993. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  36. Lillard, A. S., Loeb, D., Berg, J., Escueta, M., Manship, K., Hauser, A., & Daggett, E. D. (2025). A national randomized controlled trial of the impact of public Montessori preschool at the end of kindergarten. Proceedings of the National Academy of Sciences, 122(43), e2506130122. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  37. Lillard, A. S., & McHugh, V. (2019a). Authentic Montessori: The Dottoressa’s view at the end of her life Part I: The environment. Journal of Montessori Research, 5(1), 1–18. [Google Scholar] [CrossRef] [Scilit]
  38. Lillard, A. S., & McHugh, V. (2019b). Authentic Montessori: The Dottoressa’s view at the end of her life Part II: The teacher and the child. Journal of Montessori Research, 5(1), 19–34. [Google Scholar] [CrossRef] [Scilit]
  39. Loveless, T. (2001). The great curriculum debate. Brookings. [Google Scholar]
  40. Lovison, V. S., & Mo, C. H. (2024). Investing in the teacher workforce: Experimental evidence on teachers’ preferences. American Educational Research Journal, 61(1), 108–144. [Google Scholar] [CrossRef] [Scilit]
  41. Löfström, E., Poom-Valickis, K., & Rumma, K. (2026). Teachers’ metaphors and beliefs about teaching and their relationship with job satisfaction. Center for Educational Policy Studies Journal, 16(1), 227–251. [Google Scholar] [CrossRef] [Scilit]
  42. Madigan, D. J., & Kim, L. E. (2021). Towards an understanding of teacher attrition: A meta-analysis of burnout, job satisfaction, and teachers’ intentions to quit. Teaching and Teacher Education, 105, 103425. [Google Scholar] [CrossRef] [Scilit]
  43. Markowitz, A. J., Mateus, D. M. C., & Weisner, K. (2024). Linking early educator wellbeing to classroom interactions and teacher turnover. Early Childhood Research Quarterly, 67, 283–294. [Google Scholar] [CrossRef] [Scilit]
  44. Matthews, R. A., Pineault, L., & Hong, Y.-H. (2022). Normalizing the use of single-item measures: Validation of the single-item compendium for organizational psychology. Journal of Business and Psychology, 37(4), 639–673. [Google Scholar] [CrossRef] [Scilit]
  45. Meyer, J. P., & Herscovitch, L. (2001). Commitment in the workplace: Toward a general model. Human Resource Management Review, 11(3), 299–326. [Google Scholar] [CrossRef] [Scilit]
  46. Meyer, J. P., Morin, A. J. S., Rousseau, V., Boudrias, J.-S., & Brunelle, E. (2021). Profiles of global and target-specific work commitments: Why compatibility is better and how to achieve it. Journal of Vocational Behavior, 128, 103588. [Google Scholar] [CrossRef] [Scilit]
  47. Miller, J. M., & Youngs, P. (2021). Person-organization fit and first-year teacher retention in the United States. Teaching and Teacher Education, 97, 103226. [Google Scholar] [CrossRef] [Scilit]
  48. Montessori, M. (1964). The Montessori method (A. George, Trans.). Schocken. [Google Scholar]
  49. Moss, J., & Wheeler, T. (2024). Beliefs about autonomy support and control in the classroom: An examination of Montessori and traditional U.S. public school teachers. Journal of Montessori Research, 10(2), 96–121. [Google Scholar] [CrossRef] [Scilit]
  50. Nagy, M. S. (2002). Using a single-item approach to measure facet job satisfaction. Journal of Occupational and Organizational Psychology, 75(1), 77–86. [Google Scholar] [CrossRef] [Scilit]
  51. Nguyen, D., See, B. H., Brown, C., & Kokotsaki, D. (2026). Leadership for teacher retention: Exploring the evidence base on why and how to support teacher autonomy, development, and voice. Oxford Review of Education, 52(1), 96–116. [Google Scholar] [CrossRef] [Scilit]
  52. Pajares, M. F. (1992). Teachers’ beliefs and educational research: Cleaning up a messy construct. Review of Educational Research, 62(3), 307–332. [Google Scholar] [CrossRef]
  53. Park, I. (2005). Teacher commitment and its effects on student achievement in American high schools. Educational Research and Evaluation, 11(5), 461–485. [Google Scholar] [CrossRef] [Scilit]
  54. Randolph, R. J., Bryson, A., Menon, L., Henderson, D., Manuel, A. K., Michaels, S., Walls Rosenstein, D. L., McPherson, W., O’Grady, R., & Lillard, A. S. (2023). Montessori education’s impact on academic and nonacademic outcomes: A systematic review. Campbell Systematic Reviews, 19(e1330), 1–82. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  55. R Core Team. (2025). R: A language and environment for statistical computing [Computer software]. Available online: https://www.R-project.org/ (accessed on 10 July 2026).
  56. Rosenshine, B. (2012). Principles of instruction: Research-based strategies that all teachers should know. American Educator, 36(1), 12–19. [Google Scholar]
  57. Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. The American Psychologist, 55(1), 68–78. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  58. Ryan, R. M., & Grolnick, W. S. (1986). Origins and pawns in the classroom: Self-report and projective assessments of individual differences in children’s perceptions. Journal of Personality & Social Psychology, 50(3), 550–558. [Google Scholar] [CrossRef] [Scilit]
  59. Shogbesan, Y. O., Adeoye, M. A., Osaro-Martins, B. E., & Shabi, Z. (2024). Perceived influence of administrative support on secondary school teachers’ effectiveness. Evaluation Studies in Social Sciences, 5(2), 102–123. Available online: https://ejournal.upsi.edu.my/index.php/ESSS/index (accessed on 20 October 2026).
  60. Suchman, S. (2026). Building public Montessori together. MontessoriPublic, 10(2), 3. [Google Scholar]
  61. Toropova, A., Myrberg, E., & Johansson, S. (2021). Teacher job satisfaction: The importance of school working conditions and teacher characteristics. Educational Review, 73(1), 71–97. [Google Scholar] [CrossRef] [Scilit]
  62. Tyree, A. K. (1996). Conceptualizing and measuring commitment to high school teaching. Journal of Educational Research, 89(5), 295–304. [Google Scholar] [CrossRef] [Scilit]
  63. Valli, L., & Buese, D. (2007). The changing roles of teachers in an era of high-stakes accountability. American Educational Research Journal, 44(3), 519–558. [Google Scholar] [CrossRef] [Scilit]
  64. van Buuren, S., & Groothuis-Oudshoorn, K. (2011). mice: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45, 1–67. [Google Scholar] [CrossRef] [Scilit]
  65. Whitaker, R. C., Dearth-Wesley, T., & Gooze, R. A. (2015). Workplace stress and the quality of teacher–children relationships in Head Start. Early Childhood Research Quarterly, 30, 57–69. [Google Scholar] [CrossRef] [Scilit]
  66. Whitescarver, K., & Cossentino, J. (2007). Lessons from the Periphery: The role of dispositions in montessori teacher training. Journal of Educational Controversy, 2(2), 1–12. [Google Scholar]
  67. Zang, L., & Chen, Y. (2022). Relationship between person-organization fit and teacher burnout in kindergarten: The mediating role of job satisfaction. Frontiers in Psychiatry, 13, 948934. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  68. Zhao, X., & Jeon, L. (2024). Examining the associations between teacher job satisfaction, workplace climate, and well-being resources within Head Start programs. Early Education and Development, 35(5), 933–949. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Quantifying Maladaptive Cognitive Offloading: A Behavioral Framework for AI-Supported Decision Making Under Complexity
Previous
Digital Game Addiction Among University Students: A Cross-Sectional Study of Sociodemographic, Behavioral, and Psychosocial Correlates
Next