Hybrid, Online, or In-Person Teaching: Exploring Teacher Satisfaction Across Instructional Modalities During the COVID-19 Pandemic

Abstract

The COVID-19 pandemic introduced significant shifts in educational delivery, leading to widespread adoption of hybrid and remote instructional modalities. This study investigates the relationship between these modalities and teacher job satisfaction during the 2020–2021 academic year in the U.S., while also examining how teacher, classroom, and school characteristics predict satisfaction within each modality. Using data from the 2020–2021 National Teacher and Principal Survey (NTPS) on 31,160 full-time public school teachers, linear regression analyses were employed to assess the relationship between instructional modalities and a composite measure of teacher satisfaction, controlling for various teacher, classroom, and school characteristics. The analysis revealed that hybrid teaching was significantly associated with lower teacher satisfaction compared to in-person teaching; however, remote teaching’s initial association with lower satisfaction became non-significant after accounting for school-level factors. Specific teacher characteristics (e.g., female, experienced, younger, holding an advanced degree) and school characteristics (e.g., school level, size, student characteristics) exhibited varying associations with satisfaction across modalities. These findings suggest that hybrid teaching posed unique challenges during the pandemic and highlight the importance of understanding modality-specific risk factors to inform targeted support for teachers in future crises.

1. Introduction

The COVID-19 pandemic disrupted global education, and schools across the U.S. scrambled to implement procedures to limit its spread. Policymakers at various levels tried to manage the disease’s impact, forcing school districts to scale distance, hybrid, and modified in-person instructional modalities in response to local health mandates. This required educators to adopt unfamiliar instructional methods and technologies (Reich et al., 2020; Schwartz et al., 2020; Sparks, 2021). Research shows that during the pandemic, teachers faced reduced satisfaction (Redding & Nguyen, 2024) and difficulties transitioning to new instructional modalities (e.g., Kraft et al., 2021; Pressley et al., 2021). For instance, teachers reported higher stress and lower satisfaction due to increased workloads and mental pressures from adapting to new formats, especially hybrid models (Steiner & Woo, 2021; Raes et al., 2020).

Hybrid teaching, which required educators to engage both in-person and remote students, was particularly challenging for educators to adapt to because it created logistical and cognitive challenges (Raes et al., 2020). The digital literacy demands created by adopting these online instructional approaches further exacerbated these challenges (Truzoli et al., 2021; Suganya & Sankareshwari, 2020). However, no research has specifically examined how the different modalities (remote, hybrid, and in-person) adopted during the COVID-19 pandemic directly compare all these modalities in their relationship to teacher satisfaction. To position this study clearly, we explicitly hypothesize that teachers who taught in hybrid or remote modalities during the COVID-19 pandemic reported lower levels of job satisfaction than those teaching fully in person. Furthermore, the first research question is confirmatory, testing hypothesized differences in teacher satisfaction across instructional modalities. In contrast, the second question is exploratory, examining how teacher, classroom, and school characteristics predict satisfaction within each modality. Specifically, we examine the following research questions:

• To what extent are instructional modalities (in-person, hybrid, and remote) associated with teacher job satisfaction during the 2020–2021 school year?
• To what extent do teacher, classroom, and school characteristics predict teacher job satisfaction across different (in-person, hybrid, and remote) instructional modalities during the 2020–2021 school year?

By addressing these questions, the study provides new empirical evidence on how instructional formats impacted teachers’ well-being during the first full academic year of the pandemic, offering guidance for school leadership and policy in future crises.

We leverage the National Teachers and Principals Survey (NTPS), a nationally representative sample of teachers with detailed school, classroom, and teacher characteristics, allowing us to control for these factors when estimating the relationship between instructional modalities and job satisfaction. Furthermore, we also explore whether the relationship between teacher satisfaction and school, classroom, and teacher characteristics varies across the three different modalities. Although the 2020–2021 school year reflects an early stage of the COVID-19 pandemic, it now serves as an important transitional reference point for understanding how instructional disruptions may inform future educational planning. In the following section, we will provide an overview of the research on the impact of COVID-19, focusing on the changes to instruction and teacher job satisfaction. the descriptive methods we used. We will then describe our data and the methods used for data description. We will then provide a description of our study’s findings and conclude with a discussion of the implications of our study for future leadership and policymaking.

2. Literature Review

In this section, we begin by describing the theoretical framework on which we based this study. We then review the literature on teacher satisfaction in relation to instructional modalities, as well as school, classroom, and teacher characteristics. Following the literature review, we develop a conceptual framework that examines the relationship between instructional modalities and teacher satisfaction.

2.1. Theoretical Framework

This study draws on Robinson et al.’s (2023) framework, which builds on Bronfenbrenner’s (1977) ecological theory to explain how teachers’ experiences are shaped by multiple, interrelated environments, the. Within this perspective, teacher satisfaction is viewed as a multidimensional construct influenced simultaneously by factors operating at the individual, classroom, and school levels rather than as an isolated psychological state. including.

The ecological model is beneficial for understanding the COVID-19 period, when shifts in modality affected every layer of teachers’ work environment, from personal routines and classroom interactions to school-wide organizational practices and policy constraints. Teacher characteristics like gender, race, and experience, classroom characteristics such as class size or student needs, and school characteristics like locality, size, and grade level all shape how teachers experience their work (e.g., Robinson et al., 2023; Dogan et al., 2021; Blatchford & Russell, 2020; Ferguson et al., 2012). The COVID-19 pandemic simultaneously disrupted all these areas. With the shift to online, hybrid, and in-person teaching, accompanied by new precautions, teachers had to respond to changes across these different teaching levels. This study employs an ecological perspective to explore how changing conditions and instructional formats influenced teacher satisfaction during the pandemic.

2.2. Teacher Job Satisfaction and COVID-19 Pandemic

Teacher job satisfaction has long been recognized as a critical component of teacher retention, instructional quality, and student outcomes. Prior to the pandemic, research identified consistent predictors of teacher satisfaction, such as manageable workload, classroom conditions, and opportunities for professional growth (Ingersoll & May, 2011; Johnson et al., 2012; Skaalvik & Skaalvik, 2020). During and after the COVID-19 pandemic, however, teachers faced new stressors, including technological adaptation, health concerns, and abrupt shifts in instructional delivery. As schools transitioned to remote and hybrid formats, many educators reported difficulty maintaining effective instruction and classroom management, key dimensions of satisfaction (M. Li & Yu, 2022; Pressley, 2021; Redding & Nguyen, 2024). Similar findings were reported by González et al. (2023), who showed that K-12 teachers’ satisfaction and sense of effectiveness dropped as they moved to online teaching, mainly because of higher workload and emotional strain. Also, recent cross-national evidence indicates that teachers’ well-being and job satisfaction declined sharply during this period, with emotional exhaustion and uncertainty about teaching formats emerging as significant predictors (Sacré et al., 2023).

2.2.1. Teachers’ Experiences and Satisfaction in Different Modalities

In the midst of the pandemic, schools modified or implemented new instructional modalities to balance safety and continue the education of children. Dorn et al. (2020) identified distance teaching (primarily online with some offline methods), hybrid teaching (either simultaneous management of in-person and remote students or alternating between online and face-to-face sessions), and in-person learning with added safety precautions such as physical distancing and mask requirements. Each of these modalities presented challenges for teachers, and several studies highlight the difficulties educators faced implementing these instructional modalities during the pandemic. While in-person teaching may have required the fewest changes to traditional instruction, the pandemic induced significant changes to comply with evolving health protocols. Research on in-person instruction during the pandemic highlighted how requirements such as mask-wearing, social distancing, and sanitation measures added new responsibilities for teachers, increasing their stress levels as they worked to balance teaching with maintaining a safe environment (Hargreaves & Blais, 2021).

Studies have shown that distance teaching introduced significant challenges for educators, many of whom reported feelings of isolation and heightened stress as they adapted to fully remote formats (M. Li & Yu, 2022; Suganya & Sankareshwari, 2020; Truzoli et al., 2021; Watermeyer et al., 2021). For example, Truzoli et al. (2021) found that the lack of face-to-face interaction made it difficult for teachers to gauge student engagement and progress, leaving many uncertain about their students’ active learning. Research also highlights the rapid shift to digital instruction, which required teachers to acquire new digital literacy skills and adapt to online platforms, often with minimal preparation or support (M. Li & Yu, 2022). Suganya and Sankareshwari (2020) observed that this steep learning curve was particularly challenging for educators with limited prior experience with these technologies, which in turn contributed to their cognitive load and reduced overall job satisfaction. Similarly, research by Watermeyer et al. (2021) reported that balancing the demands of teaching while learning to navigate unfamiliar technologies increased workload and stress.

Hybrid teaching presented a distinct set of challenges for educators who were required to navigate the complexities of managing both in-person and remote instruction (K. C. Li et al., 2023; Raes et al., 2020; Steiner & Woo, 2021). Studies examining different approaches to Hybrid teaching found that both simultaneous and sequential hybrid instructional models caused additional strain on teachers’ cognitive load and stress related to managing technologies, curricula, and students across in-person and online settings (K. C. Li et al., 2023; Raes et al., 2020). Research has also shown that hybrid teaching was generally less preferred by educators compared to fully in-person instruction (Pressley, 2021; Steiner & Woo, 2021). Steiner and Woo (2021) found that only 23% of teachers in their sample expressed a preference for hybrid teaching, and Pressley (2021) observed that teachers engaged in hybrid instruction found it particularly challenging to manage their work–life balance. In addition, Griffard et al. (2025) found that teachers working in hybrid and remote school environments experienced higher stress and a lower sense of connection within their schools, which negatively affected their satisfaction.

2.2.2. Teacher Characteristics and Teacher Satisfaction

A rich body of literature has explored the relationship between teacher satisfaction and teacher characteristics. Prior research has found that personal characteristics such as gender, race/ethnicity, age, experience, and educational levels are all significant predictors of teacher satisfaction both before and during the COVID-19 pandemic, but these relationships are not all stable following the onset of the pandemic (Ingersoll & May, 2011; Redding & Nguyen, 2024). Ingersoll and May (2011) and Carver-Thomas (2018) both find evidence to suggest that teachers of color experience higher levels of dissatisfaction and turnover. Generally, studies find that older educators are less satisfied, but this pattern may in part be dependent on the context in which they teach (Steiner & Woo, 2021). Both more experienced and more educated teachers tend to be more highly satisfied (Ferguson et al., 2012; Ingersoll & May, 2011), but Redding and Nguyen (2024) found some evidence that education levels and satisfaction are negatively related, which may signal a larger labor market change between education and job satisfaction.

2.2.3. Classroom Characteristics and Teacher Satisfaction

Research has shown that classroom factors, including student–teacher ratios and the proportions of students with Individualized Education Programs (IEPs) and Limited English Proficiency (LEP), influence teacher satisfaction (Billingsley & Bettini, 2019; Cooper & Travers, 1996; Pettit, 2011). Larger class sizes and higher student–teacher ratios are linked to dissatisfaction, as they increase workload, reduce individualized attention, and create classroom management challenges (Cooper & Travers, 1996; Pettit, 2011). Special education teachers working with students requiring IEPs often report additional burdens due to time management challenges and the necessity for specialized instructional plans to meet diverse student needs (Billingsley & Bettini, 2019; Blatchford & Russell, 2020). Similarly, teaching LEP students adds complexity, as educators must navigate communication barriers and modify curricula, increasing workload and stress (Pettit, 2011).

2.2.4. School Characteristics and Teacher Satisfaction

Research indicates that school factors significantly influence teacher job satisfaction. School location can play an important role in shaping teacher satisfaction, as different settings offer unique challenges and advantages. Ladd (2011) found that rural teachers face higher attrition rates due to limited resources and institutional support. Knoblauch and Chase (2015) found that rural and urban schools often face issues such as limited funding and teacher shortages, whereas suburban schools typically have greater access to resources. At the same time, rural schools may offer benefits such as smaller class sizes, closer teacher-student relationships, and stronger community support, which can boost satisfaction and autonomy (Knoblauch & Chase, 2015). Teachers working at schools in towns also reported higher satisfaction with organizational decisions, likely due to less administrative oversight and a more supportive work environment (Knoblauch & Chase, 2015). Regarding school level, Allen (2005) found that primary school teachers report higher satisfaction than secondary teachers. Additionally, Dogan et al. (2021) observed that elementary teachers and high school teachers find different factors more salient to their satisfaction in the job. Skaalvik and Skaalvik (2020) found that despite higher morale among high school teachers, elementary teachers experience greater self-efficacy, though managing younger students can add stress and reduce overall satisfaction.

This brief overview of the literature on the factors associated with teacher job satisfaction points to potential factors that may serve to moderate the relationship between instructional modalities and satisfaction. Some of the more recent work to emerge during and after the pandemic has suggested that personal, classroom, and school characteristics may have differentially shaped how teachers experienced teaching in these new instructional modalities. While a few studies about new instructional modalities during the pandemic have pointed to differences in teacher job satisfaction, our study hopes to extend this research by exploring whether teacher, classroom, and school characteristics’ relationships with job satisfaction varied across instructional modalities. In the following section we provide an overview of our conceptual framework that illustrates how these elements are interconnected. Taken together, these findings show that teacher satisfaction is strongly shaped by the broader conditions in which teachers work. Across the literature, satisfaction often differs depending on school resources, local policies, and wider structural conditions, which highlights the need to consider these ecological and contextual factors when examining differences across instructional modalities.

2.3. Conceptual Framework

Based on the theoretical framework and literature review, we developed a conceptual framework to guide our study on how various characteristics of the school environment influence teacher satisfaction during the COVID-19 pandemic. We draw on Bronfenbrenner’s (1977) ecological systems theory, which emphasizes the interconnected levels of the environment. Figure 1 illustrates how we conceptualize teacher, classroom, and school characteristics situated within broader contextual layers. The framework also recognizes the COVID-19 pandemic as a contextual force that affected all parts of the educational environment.

The literature suggests that the impact of the pandemic was not consistent across schools, classrooms, or teachers. Instead, it varied depending on the specific characteristics of each setting, such as instructional modality, student composition, and access to support. Additionally, the relationship between pandemic-related changes and the challenges experienced by teachers was not a simple or linear process. The pandemic did not affect schools first, followed by classrooms, and then teachers. Instead, these levels influenced one another in complex and reciprocal ways, and each level could be affected independently.

Figure 1 reflects these assumptions by presenting the dynamic interactions among teacher, classroom, and school characteristics. While the model acknowledges that the pandemic broadly influenced the entire school environment, our primary focus is on understanding how different instructional modalities, including remote, hybrid, and in-person formats, are related to teacher satisfaction. Additionally, the model explores how teacher satisfaction is associated with teacher, classroom, and school characteristics within each modality during the COVID-19 pandemic.

Based on the theoretical framework, related literature, and conceptual framework, we hypothesize that teachers in hybrid and remote modalities felt less satisfied than teachers in in-person settings, confirming what other studies have found regarding these modalities (K. C. Li et al., 2023; M. Li & Yu, 2022; Raes et al., 2020; Steiner & Woo, 2021; Suganya & Sankareshwari, 2020; Truzoli et al., 2021; Watermeyer et al., 2021). Our second research question is more exploratory, and we do not propose a hypothesis. Instead, we hope to find evidence that may suggest whether certain teacher, classroom, and school characteristics were related to lower or higher levels of satisfaction within the subsamples of teachers teaching in hybrid, distance, and in-person instructional modalities.

3. Materials and Methods

3.1. Data Sources

This research utilizes data from the 2020–2021 National Teacher and Principal Survey (NTPS), conducted by the National Center for Education Statistics (NCES). For the 2020–2021 cycle, six questionnaires were developed, including those for principals, schools, and teachers, with separate versions tailored to public and private institutions to address their unique contexts. The focus of this study is the data collected for the public school teacher questionnaire. The sample included schools providing instruction in grades 1–12. A nationally representative sample of 9900 public schools and 68,300 teachers were surveyed nationwide. Data collection took place between the summer of 2020 and the spring of 2021, utilizing a combination of mail, internet-based methods, and follow-up via telephone and email to maximize response rates. For public school teachers, the response rate reached 55%. Rigorous data preparation steps included quality control checks, imputation to handle missing responses through donor matching or statistical modeling, and weighting adjustments to correct for nonresponse and sampling probabilities. These processes enhanced the dataset’s reliability and national representativeness (National Center for Education Statistics, 2021). For this study, part-time teachers and those not teaching in regular schools were excluded, resulting in a final dataset of 31,160 full-time teachers from public schools spanning all 50 states and the District of Columbia.

3.2. Key Variables

Our dependent variable is teacher satisfaction and we operationalize it using eight Likert-scale questions that measure different aspects of teachers’ satisfaction, such as emotional exhaustion, overall job satisfaction, professional fulfillment, intent to leave the profession, perceptions of school management, sense of collegial support, motivation levels, and thoughts of absenteeism or transferring schools. These questions resemble the types of questions used in prior research to measure teacher job satisfaction (Redding & Nguyen, 2024). Appendix A presents the specific questions we used to create our measure of teacher satisfaction. We then performed an exploratory factor analysis, which suggested a single underlying factor. Appendix A presents the factor loadings for the survey items related to teacher satisfaction. These loadings were used to create a latent variable representing overall teacher satisfaction, derived from eight items measuring different aspects of the construct. The resulting composite factor, referred to as SAT, reflected a unidimensional structure of satisfaction as confirmed through exploratory factor analysis. To ensure comparability across analyses, the SAT scores were standardized to have a mean of 0 and a standard deviation of 1. After standardization, the SAT ranged from −3.19 to 1.50 and showed a slight negative skew, indicating that teachers generally reported higher levels of satisfaction than dissatisfaction.

The key independent variable is a categorical measure of instructional modalities reported by teachers during the COVID-19 pandemic. These modalities are categorized into three primary groups: distance classes (fully remote instruction), in-person classes, and hybrid classes (a combination of remote and in-person instruction). Two additional categories, “classes unaffected by COVID-19” and “classes completely cancelled due to COVID-19,” represent less than 1% of responses and are excluded from the analysis.

The teacher characteristic variables included in the analysis are gender, race (White, Black, Asian, Native Hawaiian/Pacific Islander, Hispanic), novice teacher status (3 or fewer years of experience, more than 3 years of experience), age (30 years old and younger, older than 30 years old), and highest degree (Master’s degree or not). For classroom characteristics, we include the number of students per teacher and the proportion of students with IEPs (Individualized Education Programs) and LEP (Limited English Proficiency) per teacher. Regarding school characteristics, we include school level (primary, middle, and high school), locality (suburban and city schools versus rural and town schools), the charter school flag, school size (number of students), the proportion of non-White students, and the proportion of students in the National School Lunch Program (NSLP).

3.3. Descriptive Analysis

Table 1. Distribution of Teacher, Classroom, and School Characteristics Across Instructional Modalities During the COVID-19 Pandemic.

Category	Remote	Hybrid	In-Person	% From Sample/Mean
Total Proportion from the Sample	18.68%	57.58%	23.74%	100%
Teacher Characteristics
Male (% from each modality)	25.23%	26.4%	19.22%	24.48%
Race (% from each modality)
White	72.99%	80.29%	89.14%	81.06%
Black	7.61%	5.88%	3.15%	5.56%
Asian	5.48%	2.7%	1.31%	2.89%
Native Hawaiian/Pacific Islander	0.85%	0.54%	0.3%	0.54%
American Indian/Alaska Native	1.37%	1.25%	1.06%	1.23%
Hispanic	11.69%	9.34%	5.04%	8.71%
3 or fewer Years of Experience (% form each modality)	12.49%	11.37%	12.27%	11.79%
30 Years Old and Lower (% from each modality)	16.61%	16.75%	19.05%	17.27%
Lower than Graduate Degree (% from each modality)	33.09%	38.94%	44.48%	39.16%
Classroom Characteristics
Students per Teacher Ratio	16.31	14.62	14.24	14.84
Proportion of IEP Students	15.40%	16.26%	15.18%	16%
Proportion of LEP Students	14.26%	8.97%	6.99%	9%
School characteristics
Level (% from each modality)
Primary	47.99%	37.04%	61.28%	44.84%
Middle	21.54%	25.84%	18.28%	23.24%
High	30.47%	37.12%	20.44%	31.92%
Urban (City + Suburb) (% from each modality)	78.83%	64.28%	47.92%	63.27%
School is Charter (% from each modality)	13.26%	6.38%	3.61%	7.01%
Proportion of Non-White Students	66.13%	47.23%	34.92%	47%
Student Enrolment in the School	869.13	876.64	642.06	819.44
Proportion of Students Enrolled in Lunch Program	63.65%	54.63%	51.72%	55.62%

(N = 31160, rounded to the nearest 10). Source: (U.S. Department of Education, National Center for Education Statistics, 2021) National and Teacher Principal Survey (NTPS) “Public School Teacher Data File,” 2020–2021.

provides a comparative overview of how different teacher, classroom, and school characteristics are distributed across instructional modalities (remote, hybrid, and in-person teaching) relative to their overall representation in the sample. Among the total sample, 18.68% of teachers were teaching remotely, 57.58% were teaching in hybrid formats, and 23.74% were teaching entirely in person. These distributions provide a foundation for examining how different attributes corresponded with each modality.

Regarding teacher characteristics, male teachers were proportionally more represented in remote and hybrid settings than in in-person instruction. White teachers were disproportionately present in in-person teaching, while non-White teachers, including Black, Asian, Hispanic, and others, were more concentrated in remote settings. Teachers with fewer than three years of experience were relatively evenly distributed across modalities, although they were slightly more concentrated in in-person instruction. Younger teachers (under 30) were more represented in in-person settings, while older teachers were more evenly distributed. Teachers without a graduate degree were most prominent in in-person instruction, while those holding a master’s degree or higher were more represented in remote teaching.

Classroom characteristics also varied across modalities. Student-to-teacher ratios were highest in remote settings and lowest in in-person instruction. The percentage of students with Individualized Education Programs (IEPs) was relatively consistent across modalities, though slightly higher in hybrid formats. In contrast, the share of Limited English Proficiency (LEP) students was highest in remote settings, suggesting that remote classrooms faced the added challenge of being more linguistically diverse. School characteristics revealed additional distinctions. Primary school teachers were largely concentrated in in-person instruction, while middle and high school teachers were more involved in hybrid teaching. Urban schools (located in cities and suburbs) had a greater share of teachers in remote settings, while non-urban schools (in towns and rural areas) leaned more toward in-person and hybrid instruction. Charter school teachers, although a smaller portion of the sample, were disproportionately represented in remote teaching. Schools offering remote and hybrid instruction had higher average student enrollment than those offering in-person instruction. Schools with a greater share of students enrolled in the National School Lunch Program were more likely to be remote, while in-person schools had the lowest lunch program participation.

Beyond the descriptive findings presented in Table 1, the analysis of the latent teacher satisfaction variable (SAT) revealed that satisfaction was highest among in-person teachers (SAT = 0.07), lowest among hybrid teachers (SAT = −0.028), and nearly zero among remote teachers (SAT = −0.001).

3.4. Analytical Strategies

The statistical analysis for the first research question was conducted using four weighted linear regression models with teacher satisfaction (Υ_i) as the dependent variable and state fixed effects (π_s) in all regressions. In the first regression, only instructional modalities are included as predictors. The second regression adds teacher characteristics, such as gender, age, experience, and education level. The third regression incorporates classroom factors, including student–teacher ratios and the proportion of students with IEPs and LEPs. Finally, the fourth regression is expanded to include additional school characteristics, such as school size, locality, and student characteristics.

This sequential approach allows us to explore how the relationship between instructional modalities and teacher job satisfaction are sensitive to the inclusion of various factors shaping teacher satisfaction during the pandemic. Also, weights are applied in all regression analyses to account for the survey’s sampling design. The regressions are as follows:

Υ_i = β₀ + β₁ (Instructional Modalities)i + πs + εi

(1)

Υ_i = β₀ + β₁ (Instructional Modalities)i + β₂ (Teacher Characteristics)i + πs + εi

(2)

Υ_i = β₀ + β₁ (Instructional Modalities)i + β₂ (Teacher Characteristics)i + β₃ (Class Characteristics)I + πs + εi

(3)

Υ_i = β₀ + β₁ (Instructional Modalities)i + β₂ (Teacher Characteristics)i + β₃ (Class Characteristics) + β₄ (School Characteristics)I + πs + εi

(4)

To address the second research question on the relationship between teacher, classroom, and school characteristics and teacher satisfaction in each modality, we run three regression models. Each includes the complete set of teacher, class, and school characteristics. The only difference is that each model only includes the subsample of teachers who reported teaching in a particular modality.

Υ_i = β₀ + β₁ (Teacher Characteristics) di + β₂ (Class Characteristics) di + β₃ (School Characteristics) di + πs + εi

(5)

3.5. Limitations

This study utilized teacher data from the National Teacher and Principal Survey (NTPS) for the 2020–2021 academic year. Because the NTPS captures data from a single time point, the findings reflect teachers’ experiences only at the moment of survey completion rather than across the full school year. The survey question on instructional modality referred solely to the teacher’s situation at that specific time, which limits our ability to assess the duration or proportion of time spent in each modality. Furthermore, the dataset does not indicate whether hybrid instruction occurred in a simultaneous (teaching remote and in-person students at the same time) or sequential (alternating between remote and in-person groups) format. These constraints restrict the temporal and structural specificity of our conclusions regarding hybrid and remote teaching experiences. Nevertheless, the large, nationally representative NTPS sample provides valuable insight into broad patterns of teacher satisfaction during the pandemic period.

4. Results

4.1. Teacher Satisfaction and Instructional Modalities

The results of our regression analyses, addressing our first research question, are presented in

Table 2. Regression Analysis of Teaching Modalities and Teacher Satisfaction.

	Teacher Satisfaction (Model 1)	Teacher Satisfaction (Model 2)	Teacher Satisfaction (Model 3)	Teacher Satisfaction (Model 4)
Remote	−0.136 ***	−0.132 ***	−0.122 ***	−0.039
	(0.032)	(0.031)	(0.030)	(0.030)
Hybrid	−0.117 ***	−0.115 ***	−0.113 ***	−0.066 **
	(0.026)	(0.024)	(0.023)	(0.021)
Female		−0.031 *	−0.031 *	−0.063 ***
		(0.015)	(0.014)	(0.017)
Black		−0.148 ***	−0.142 ***	−0.062
		(0.036)	(0.035)	(0.033)
Asian		−0.117 **	−0.099 **	−0.057
		(0.036)	(0.035)	(0.036)
Native Hawaiian/Pacific islander		−0.055	−0.051	−0.012
		(0.079)	(0.078)	(0.072)
American Indian/Alaska Native		−0.221 **	−0.217 **	−0.174 *
		(0.067)	(0.065)	(0.067)
Hispanic		−0.011	0.018	0.060 *
		(0.028)	(0.028)	(0.026)
More than 3 Years of Experience		−0.147 ***	−0.146 ***	−0.160 ***
		(0.033)	(0.033)	(0.032)
More than 30 Years Old		0.155 ***	0.149 ***	0.140 ***
		(0.022)	(0.021)	(0.021)
Master’s Degree or Higher		−0.047 *	−0.048 *	−0.044 *
		(0.018)	(0.018)	(0.017)
Annual Salary (Thousands		0.003 ***	0.003 ***	0.003 ***
		(0.001)	(0.001)	(0.001)
Estimated Number of Students per FTE Teacher			−0.003	−0.000
			(0.002)	(0.002)
Proportion of Students with an IEP			0.054 *	0.063 *
			(0.027)	(0.026)
Proportion of Students Who Are LEP			−0.241 ***	−0.118
			(0.056)	(0.066)
Middle School				−0.065 ***
				(0.017)
High School				−0.071 *
				(0.027)
City or Suburb				0.003
				(0.032)
School Is Not a Public Charter School.				0.027
				(0.038)
Log of School Size				−0.046 **
				(0.014)
Proportion of Students Who Are Non-Whit				−0.251 ***
				(0.060)
Proportion of Students Approved for the NSLP				−0.135 ***
				(0.031)
Constant	0.270 ***	0.122 *	0.207 ***	0.623 ***
	(0.025)	(0.052)	(0.053)	(0.110)
R-Squared	0.01	0.02	0.02	0.03
State Fixed Effects Included	Yes	Yes	Yes	Yes
Observations	31,160	31,160	31,160	31,160

NOTE: All values rounded to the nearest tens. Nationally representative weights are employed. Robust Standard Errors are in parentheses. The in-person is the reference group. The weight variable used in the analysis is TFNLWGT. (N = 31160, rounded to the nearest 10). * p < 0.05, ** p < 0.01, *** p < 0.001. Source: (U.S. Department of Education, National Center for Education Statistics, 2021) National and Teacher Principal Survey (NTPS) “Public School Teacher Data File,” 2020–2021.

. These analyses explore the relationships between teaching modalities during the COVID-19 pandemic and teacher satisfaction. The results in Model 1 show that teachers engaged in remote learning had −0.136 standard deviation units lower teacher satisfaction compared to those who taught in person. Similarly, hybrid teaching is associated with a 0.117 of a standard deviation lower in satisfaction compared to teachers engaged in in-person instruction. In Model 2, after adding teacher characteristic variables, the coefficients for remote teaching and hybrid teaching remain negative and significant with only slight changes in magnitude for the point estimates, suggesting lower job satisfaction for teachers in both remote and hybrid modalities compared to in-person teaching.

In Model 3, classroom characteristic variables are added, including the number of students per teacher, the proportion of IEP students, and the proportion of LEP students. Remote and hybrid teaching remain significantly associated with lower levels of satisfaction. In Model 4, school characteristic variables are added, and the results from the fully saturated model are reported. Remote teaching, compared to in-person, no longer has a significant relationship with teacher satisfaction. Hybrid teaching, however, remains significantly associated with lower satisfaction, but the point estimate has shrunk noticeably to 0.066 of a standard deviation. This consistency across models suggests that even after accounting for teacher, classroom, and school characteristics, modality, particularly for hybrid teaching, continues to have a meaningful and statistically significant relationship to teacher satisfaction.

Also in Model 4, we find that female teachers consistently report lower satisfaction than male teachers, American Indian/Alaska Native and Hispanic teachers also report lower satisfaction compared to White teachers, and teachers with more than three years of experience report less satisfaction than their less experienced peers. Additionally, teachers older than 30 years tend to show higher satisfaction levels compared to younger teachers, and those with a master’s degree or higher tend to be slightly less satisfied than teachers with a bachelor’s degree or lower. Higher teacher salaries are associated with slightly higher levels of satisfaction. For classroom characteristics, a higher proportion of students with IEPs was associated with slightly higher levels of satisfaction, consistent with prior findings that teachers working closely with students who receive specialized support may experience a greater sense of professional efficacy. In contrast, neither class size nor the proportion of LEP students showed a significant relationship with satisfaction. Regarding school characteristics, both middle and high school teachers report lower satisfaction than primary school teachers. Larger school size is associated with lower levels of satisfaction, as are schools with higher proportions of non-White students and those with more students approved for the National School Lunch Program. Satisfaction levels do not differ significantly across urban and non-urban schools.

4.2. Teacher Job Satisfaction Across Modalities

The regression analyses presented in

Table 3. Regression Analysis of Teacher Satisfaction Across Instructional Modalities.

	Remote	Hybrid	In-Person
Female	−0.079 **	−0.055 *	−0.072
	(0.025)	(0.024)	(0.041)
Black	−0.057	−0.067	0.005
	(0.082)	(0.034)	(0.075)
Asian	−0.003	−0.054	−0.215 *
	(0.059)	(0.063)	(0.092)
Native Hawaiian/Pacific islander	0.176	−0.043	−0.305
	(0.141)	(0.090)	(0.270)
American Indian/Alaska Native	−0.127	−0.231 *	−0.069
	(0.161)	(0.090)	(0.093)
Hispanic	0.073	0.038	0.120
	(0.053)	(0.041)	(0.067)
More Than 3 Years of Experience	−0.268 ***	−0.125 **	−0.155
	(0.048)	(0.039)	(0.080)
More than 30 Years Old	0.191 ***	0.095 **	0.203 ***
	(0.042)	(0.032)	(0.045)
Master’s Degree or Higher	−0.091 *	−0.031	−0.033
	(0.035)	(0.019)	(0.036)
Annual Salary (Thousands)	0.003 ***	0.004 ***	0.002
	(0.001)	(0.001)	(0.001)
Estimated Number of Students per FTE Teacher	−0.002	0.001	−0.003
	(0.004)	(0.002)	(0.003)
Proportion of Students with an IEPteachers’	−0.008	0.051	0.163 *
	(0.054)	(0.029)	(0.068)
Proportion of Students Who Are LEP	0.023	−0.151 *	−0.102
	(0.103)	(0.064)	(0.224)
Middle School	−0.031	−0.063 *	−0.114 *
	(0.055)	(0.025)	(0.043)
High School	−0.037	−0.078 *	−0.100 *
	(0.061)	(0.038)	(0.049)
City or Suburb	0.095 *	−0.015	0.018
	(0.043)	(0.042)	(0.035)
School Is Not a Public Charter School	0.101	−0.003	0.038
	(0.063)	(0.041)	(0.055)
Log of School Size	−0.042	−0.038 *	−0.062
	(0.031)	(0.018)	(0.031)
Proportion of Students Who Are Non-White	−0.258 *	−0.235 ***	−0.326
	(0.112)	(0.052)	(0.196)
Proportion of Students Approved for the NSLP	−0.089	−0.132 ***	−0.186 **
	(0.057)	(0.034)	(0.070)
Constant	0.498 *	0.448 ***	0.679 ***
	(0.222)	(0.123)	(0.181)
R-Squared	0.05	0.03	0.04
State Fixed Effects Included	Yes	Yes	Yes
Observations	5820	17,940	7400

NOTE: Details may not sum to totals because of rounding. All values are rounded to the nearest tens. Nationally representative weights are employed. Robust Standard Errors are in parentheses. The weight variable used in the analysis is TFNLWGT. (N = 31160, rounded to the nearest 10). * p < 0.05, ** p < 0.01, *** p < 0.001. Source: (U.S. Department of Education, National Center for Education Statistics, 2021) National and Teacher Principal Survey (NTPS) “Public School Teacher Data File”.

examines the relationship between teacher satisfaction and teacher, classroom, and school characteristics across the three instructional modalities. These models allow us to explore which factors may be more or less salient to teacher satisfaction when teachers are implementing these different modalities. This analysis may point to potential issues to target for support in the event of future crises that may require us to once again utilize these newer instructional approaches. While not causal, these exploratory analyses help to extend the conversation beyond the fact that teacher satisfaction was differential across modalities and towards identifying the specific issues that may have contributed to lower levels of satisfaction during the pandemic.

In remote teaching, several significant predictors of satisfaction emerged. Female teachers reported lower satisfaction compared to male teachers. Teachers with more than three years of experience were less satisfied, while those over the age of 30 reported significantly higher satisfaction. Teachers with a graduate degree reported lower satisfaction levels compared to those without one. Higher teacher salary was positively associated with satisfaction. Teachers in city and suburban schools reported greater satisfaction than their peers in rural or town schools. Finally, a higher proportion of non-White students in the school was linked to lower teacher satisfaction.

In hybrid teaching, both overlapping and distinct patterns were observed. Female teachers again reported lower satisfaction than male teachers. American Indian/Alaska Native teachers reported significantly lower satisfaction compared to White teachers. As in the remote model, more experienced teachers were less satisfied, and older teachers were more satisfied. Higher teacher salary remained positively associated with satisfaction. A higher proportion of students with limited English proficiency and a larger school size were both associated with reduced satisfaction. Additionally, teachers in schools with higher proportions of non-White students and greater student participation in the National School Lunch Program reported significantly lower satisfaction.

In in-person teaching, the predictors of satisfaction differed in some respects. Gender and most racial categories were not significantly associated with satisfaction, although Asian teachers reported lower satisfaction. Teachers with more than three years of experience were less satisfied, and older teachers (over 30) were more satisfied, consistent with the other two models. Salary was positively related to satisfaction but was not statistically significant. Teachers with a higher proportion of students with IEPs in their classrooms reported greater satisfaction. Teachers in middle and high schools expressed lower satisfaction compared to those in primary schools. Similarly to other modalities, a higher proportion of students enrolled in the National School Lunch Program was associated with lower satisfaction. Schools with more non-White students also showed a negative, though not statistically significant, association with teacher satisfaction.

When comparing across modalities, several consistent findings emerge. Female teachers reported lower satisfaction in remote and hybrid models, but not in in-person settings. Teachers with more than 3 years of experience were consistently less satisfied than more experienced teachers across all modalities, while teachers who were more than 30 years old showed higher satisfaction in each setting. Graduate-level education was negatively associated with satisfaction only in remote instruction. Teacher salary was a positive predictor of satisfaction in remote and hybrid teaching, with a weaker effect in in-person contexts. Among classroom-level variables, working with more students with IEPs was associated with higher satisfaction only in in-person settings. In contrast, a higher proportion of LEP students reduced satisfaction in hybrid settings. At the school level, teachers in urban areas were more satisfied in remote instruction, while larger schools and those serving more racially diverse and economically disadvantaged students were associated with lower satisfaction, particularly in hybrid settings.

5. Discussion

The study’s first research question aimed to identify differences in teacher satisfaction based on the instructional modality used during the COVID-19 pandemic in the 2020–2021 academic year. Our findings indicate that hybrid teaching was consistently associated with significantly lower levels of teacher satisfaction compared to in-person instruction. Although remote teaching also showed a lower average satisfaction level than in-person teaching, this association was not statistically significant in our fully saturated model. This result aligns with existing literature highlighting the challenges and dissatisfaction teachers report in hybrid settings (e.g., K. C. Li et al., 2023; Pressley, 2021). However, when school-level characteristics were included in the regression model, the association between remote teaching and dissatisfaction diminished and was no longer statistically significant. This suggests that the initial dissatisfaction associated in the first 3 models with remote teaching may have been influenced more by contextual school characteristics than by the instructional modality itself.

The regression analysis also identified specific teacher, classroom, and school characteristics significantly related to teacher satisfaction during the pandemic. Female teachers, American Indian teachers, teachers under 30 years old, and those with a master’s degree or higher reported lower levels of satisfaction compared to their peers. At the classroom level, teachers with a lower proportion of students with Individualized Education Programs (IEPs) were less satisfied. At the school level, middle school and high school teachers were less satisfied than elementary school teachers. Teachers in larger schools, as well as those in schools with a higher proportion of non-White students or students approved for the National School Lunch Program, reported lower satisfaction during the 2020–2021 academic year.

To address the second research question, which examined how the relationship between teacher satisfaction and characteristics varied across instructional modalities, we analyzed each setting separately. In remote teaching, lower satisfaction was reported by female teachers, those under 30 years of age, those with more than 30 years of experience, teachers holding a master’s degree or higher, and those teaching in schools with a higher proportion of non-White students. In hybrid teaching, dissatisfaction was found among teachers with less than 30 years and more than 30 years of experience, American Indian or Native Hawaiian teachers, teachers with more students with IEPs, those teaching in middle or high schools, and those in larger schools or schools with a higher proportion of non-White or NSLP-eligible students. In in-person teaching environments, Asian teachers reported lower satisfaction. Additionally, teachers with fewer than 30 years of experience and those teaching classes with a lower proportion of English Language Learners reported less satisfaction. Teachers working in middle and high schools and in schools with a higher percentage of students enrolled in NSLP also reported reduced satisfaction.

These results suggest associations that identify certain teacher, classroom, and school characteristics as potential risk factors for dissatisfaction when teaching under specific modalities during a crisis like the COVID-19 pandemic. However, these findings should not be interpreted as causal relationships, given the cross-sectional nature of the data. When comparing predictors of teacher satisfaction across instructional modalities, some patterns were consistent while others varied. Female teachers reported significantly lower satisfaction in remote and hybrid teaching, but gender was not a significant factor in in-person teaching. Teachers with more than three years of experience felt less satisfied in remote and hybrid modalities but not in in-person teaching environments. Conversely, teachers older than 30 years consistently reported higher satisfaction across all modalities. Salary was positively associated with satisfaction in both remote and hybrid settings, but not in in-person teaching. Academic degree showed a negative association with satisfaction only in the remote setting.

At the classroom level, only in-person teachers working with a higher proportion of IEP students reported greater satisfaction. The presence of English Language Learner students negatively affected satisfaction in hybrid teaching but was not significant in other modalities. At the school level, teachers in cities and suburbs had higher satisfaction in remote teaching. In contrast, larger school size and a greater proportion of NSLP-eligible students were associated with lower satisfaction in both hybrid and in-person settings. A higher percentage of non-White students was a significant negative predictor in remote and hybrid modalities but not in in-person teaching.

Although this study identifies general characteristics that predict dissatisfaction for teachers during the COVID-19 pandemic, each modality presents distinct combinations of risk and protective factors that shape teacher experience. Associations between teacher satisfaction and school composition variables likely reflect broader contextual and structural conditions rather than individual-level differences, so these patterns should be interpreted within the constraints of the NTPS data and the sociocultural contexts in which instructional modalities were adopted. These findings highlight the need for a nuanced, context-specific understanding of teacher experience across instructional formats. Moreover, it points to particular teachers that may be candidates for additional support when implementing these instructional modalities. Given the limited resources available to public schools, it may be most efficient to try and understand why certain types of teachers experienced lower levels of satisfaction in particular modalities. For instance, an important avenue for further exploration may be understanding why women were less satisfied with hybrid and distance learning and what particular support may have helped to alleviate that dissatisfaction with their work.

6. Conclusions

Based on the findings of this study, we conclude that the hybrid modality, despite being designed to preserve instructional quality during times of crisis (Dorn et al., 2020), is associated with significantly lower teacher satisfaction compared to in-person instruction. While the dataset does not specify whether teachers taught in simultaneous or sequential hybrid formats, existing literature suggests that both formats place additional burdens on teachers. In simultaneous hybrid teaching, managing two groups of students at once creates significant complexity and workload (Raes et al., 2020). In sequential hybrid formats, teachers must adapt to two separate instructional methods, increasing stress and preparation demands (Pressley, 2021). Additionally, hybrid settings presented challenges to maintaining work–life balance, as noted by Cañal et al. (2024) and Kraft et al. (2021), who found that hybrid teaching blurred personal and professional boundaries. Earlier research has also associated hybrid teaching with multitasking, burnout, and reduced job satisfaction (Pressley, 2021; Raes et al., 2020; Steiner & Woo, 2021), all of which may help explain the lower satisfaction reported by teachers in our study.

In contrast, although we hypothesized that remote teaching would also predict lower satisfaction due to feelings of isolation (Truzoli et al., 2021) and lack of digital literacy among some teachers (M. Li & Yu, 2022), no significant dissatisfaction was observed in our fully saturated model. This may be because remote teaching does not impose the same multitasking burden as hybrid formats or because in-person teaching presented its own challenges during the pandemic, such as enforcing safety protocols that may have had a comparable impact on job satisfaction (Hargreaves & Blais, 2021).

The findings also show that teachers who were female, American Indian, had more than three years of experience, were under 30 years old, and held a master’s degree or higher were more likely to report dissatisfaction during the COVID-19 pandemic. These results are largely consistent with prior studies conducted before and during the pandemic, stating that gender (Klassen & Chiu, 2010; Pyhältö et al., 2011), race (Carver-Thomas, 2018; Ingersoll & May, 2011), and holding a master’s degree or higher (Redding & Nguyen, 2024) predict lower satisfaction for teachers in general. In contrast, teachers who are in the early stages of their careers have been shown to be more satisfied than others (Ingersoll & May, 2011).

Furthermore, teachers who taught in schools with more IEP students reported slightly higher satisfaction, whereas those in secondary levels, larger schools, and schools with greater proportions of non-White or NSLP-eligible students reported lower satisfaction. The literature supports the theory that teachers with higher IEP student populations (Billingsley & Bettini, 2019) and middle school and high school teachers (Dogan et al., 2021) feel less satisfied than their peers (Ingersoll & May, 2011; Klassen & Chiu, 2010; Klusmann et al., 2008). Our findings seem to point to the potential that teachers in these contexts may be at greater risk of turnover and burnout. Therefore, school leaders and policymakers should pay particular attention to supporting these teachers during crises.

Looking more closely at each instructional modality, the hybrid setting emerges as the most problematic. Teachers with at-risk characteristics in hybrid teaching, such as being female, having more than three years of experience, being younger than 30 years of age, an American Indian or Native Hawaiian background, higher proportions of IEP students, teaching in secondary school, and working in larger or more disadvantaged schools were consistently less satisfied. It would behoove school districts and policymakers to prioritize support for these groups, especially when implementing hybrid models. Extra attention is essential when teachers fall into more than one of these categories, potentially increasing their vulnerability.

Remote and in-person modalities may not have been perceived as unfavorably by teachers to the same extent as hybrid teaching; however, the dissatisfaction reported by female teachers, those under 30 years of age, those with more than 30 years of experience, teachers holding a master’s degree or higher, and those teaching in schools with a higher proportion of non-White students should be noted in remote settings. For in-person teaching, Asian teachers, teachers with fewer than 30 years of experience, teachers working in middle and high schools, and those working in schools with a higher percentage of students enrolled in the NSLP also reported reduced satisfaction. These groups should be considered at-risk when teaching in person during crisis conditions.

Regarding possible future crises, schools should be prepared to respond effectively to the frustrations of teachers. It is important that school leaders understand how different instructional modalities during emergencies may impact teachers’ job satisfaction. Knowledge of teacher characteristics that are associated with dissatisfaction in specific modalities can further help schools provide targeted support to potentially at-risk teachers during such crises. Further research is needed to better understand how these characteristics influence teacher satisfaction in each modality, both during crises and under normal conditions.