Beyond Quality Features,Toward Teacher Needs: Unpacking How Professional Development Quality Relates to Teacher Engagement Through Self-Determination Theory

Abstract

While the features of high-quality professional development (PD) have been well-explored, some feature-based PD programs yielded inconsistent teacher engagement, prompting investigations into how these features relate to teacher engagement. Drawing on Self-Determination Theory, this study collected data from 985 German in-service teachers to examine basic psychological needs (BPN) as a mediator between perceived PD quality and teacher engagement. Using both variable-centered and person-centered approaches, we (a) tested whether BPN mediates this relationship, (b) identified teachers’ perceived PD quality profiles, and (c) explored whether this mediation varies across profiles. Latent profile analysis identified four profiles: High-Quality Balanced, Moderate-Quality Weak-Collaboration, Moderate-Quality Strong-Collaboration, and Low-Quality Strong-Collaboration. Both variable-centered and person-centered mediation analyses revealed that BPN consistently mediated the relationship between PD quality and engagement, with no significant differences in the BPN-engagement pathway across all profiles. These findings suggest a shift from merely including features toward designing them in ways that support teachers’ psychological needs.

Keywords

teacher professional development teacher engagement Self-Determination Theory basic psychological needs latent profile analysis

Introduction

The Challenge: From Design Features to Teacher Engagement

Teacher professional development (PD) is widely recognized as a key lever for educational improvement (Guskey, 2002; Kennedy, 2016). However, simply attending PD does not necessarily mean that teachers are truly engaged in the learning process (Lipowsky & Rzejak, 2015; Liu et al., 2025). Teacher engagement represents a complex interplay of behavioral, cognitive, and affective investment (Fütterer et al., 2024), which has been linked to greater uptake and enactment of PD content in classroom practice, making it an important indicator for understanding the quality and potential impact of PD (Beisiegel et al., 2018; Ji, 2021).

Decades of research have identified common design features of high-quality PD programs, such as collaboration, practical relevance, and content focus (Desimone, 2009; Richter & Richter, 2024; Sims & Fletcher-Wood, 2021; Walter & Briggs, 2012). Yet including these features alone does not always lead to high engagement. Several well-designed PD programs have struggled to elicit sustained teacher engagement (Hill et al., 2018; Karsenty & Brodie, 2023; Smylie, 2014). One potential reason is that design features establish external conditions, but teachers engage based on how they internally experience these conditions (Boström & Palm, 2020). It remains unclear how teachers experience PD programs and what supports their engagement during the learning process (Borko, 2004; Kennedy, 2016). Consequently, there is a pressing need to shift from merely identifying effective PD features toward illuminating the internal psychological processes that drive teacher learning (Opfer & Pedder, 2011).

A Needs-Based Perspective: Self-Determination Theory

To unpack this black box between teachers’ PD experiences and engagement, research has suggested drawing theories from other fields, such as psychology or human resource development literature (Kennedy, 2016; Thurlings & den Brok, 2017). Self-Determination Theory (SDT; Deci & Ryan, 2012) offers a well-established framework for understanding how perceived environments relate to human motivation and behavior. SDT posits that motivation and engagement are not automatic responses to external stimuli but are driven by the satisfaction of three basic psychological needs (BPN): autonomy, competence, and relatedness (Ryan & Deci, 2017). When the environment supports these needs, individuals universally tend to exhibit greater intrinsic motivation and engagement (Ryan & Deci, 2017). Qualitative work in PD settings has corroborated these claims, highlighting the importance of teachers’ needs satisfaction for sustained engagement (e.g., Power & Goodnough, 2019; Yang, 2021). However, quantitative evidence for BPN satisfaction as a mediator between perceived PD quality and teacher engagement remains limited (Noonan, 2024).

Integrating Variable-Centered and Person-Centered Approaches

In practice, teachers may experience varying levels of PD quality due to differences in program design (Desimone, 2009; Garet et al., 2001), facilitation and implementation quality (Borko, 2004), and teachers’ individual motivations and learning preferences (Opfer & Pedder, 2011; Richter et al., 2025). Existing PD studies often use variable-centered approaches to capture teachers’ experiences (e.g., Fütterer et al., 2024; Meyer et al., 2023). This approach estimates associations at the overall level (Bergman & Trost, 2006) and assumes that teachers’ experiences are homogeneous, neglecting individual differences in PD experiences. In addition, while valuable for comparing the relative strength of individual predictors, this approach typically treats PD dimensions as separate variables and may miss how combinations of features shape teachers’ overall experiences. A person-centered approach, such as latent profile analysis (LPA), can complement variable-centered work by identifying naturally occurring configurations of perceived PD quality (Howard & Hoffman, 2018) and testing whether the proposed mediation pathway holds across profiles.

The Present Study

Against this background, the present study integrates SDT as the theoretical foundation with a dual-perspective analytical strategy. In this study, this strategy is used to unpack the black box between perceived PD quality and teacher engagement by examining (a) an internal psychological pathway via BPN satisfaction and (b) the universality of this pathway across different levels of perceived PD quality. First, a variable-centered mediation analysis tests whether BPN satisfaction mediates the association between perceived PD quality and teacher engagement at the sample level, establishing evidence for a general psychological pathway. Second, the person-centered LPA extends this inquiry by identifying distinct profiles of teachers’ PD quality experiences and examining whether the BPN mediation pathway holds consistently across these profiles. Accordingly, we aim to address three research questions (RQ):

Research Question 1: Does teachers’ perceived PD quality relate to teacher engagement via BPN satisfaction?

Research Question 2: What distinct latent profiles of perceived PD quality can be identified among teachers?

Research Question 3: Does the mediation model specified in RQ1 hold consistently across the identified latent profiles?

Literature Review

Defining Perceived PD Quality

Evaluating the quality of PD programs has long been a topic of discussion among researchers and practitioners. Over the past decades, research has moved toward a consensus on core features of high-quality PD programs, such as collaboration and practical relevance (Darling-Hammond et al., 2017; Desimone, 2009). Although these features provide a foundation for effective design, reliably measuring them remains a challenge (Sims & Fletcher-Wood, 2021). Many existing evaluation tools are either tailored to specific PD formats (Meyer et al., 2023; Soine & Lumpe, 2014) or lack comprehensive psychometric validation (Gaumer Erickson et al., 2017). More recently, the TPD Monitor was developed to address these limitations by integrating established design features with empirical evidence, and it has demonstrated reliability and measurement invariance across diverse PD environments (Richter & Richter, 2024). Because the present study requires a validated and format-independent measure of perceived PD quality, we employed TPD Monitor to categorize perceived PD quality into four dimensions: clarity and structure, cognitive activation, collaboration, and practical relevance.

First, clarity and structure emphasize the importance of explicit learning goals and transparent expectations. Structured PD designs, such as scaffolded guidance (Kleickmann et al., 2016), have proven more effective than unstructured approaches, particularly when facilitators align activities with explicit objectives (Antoniou & Kyriakides, 2013). Second, cognitive activation involves deep intellectual engagement, encouraging teachers to connect new content to prior knowledge and tackle challenging tasks. For example, PD that prompts teachers to critically evaluate established beliefs and invest effort in transformative learning can facilitate shifts in their beliefs and practices (Ebert & Crippen, 2010). Third, collaboration fosters professional growth by enabling peer dialogue and collective refinement of teaching methods. Collaborative activities, whether through group discussions or joint projects, enable teachers to critique assumptions and integrate diverse perspectives (Vangrieken et al., 2015). Finally, practical relevance ensures PD content directly addresses classroom needs, such as alignment with discipline-specific curricula or the use of authentic teaching materials.

While each dimension is individually important, studies also suggest they function as an integrated learning environment rather than as isolated components. For example, Voogt and colleagues (2015) found that when collaboration was combined with a focus on teacher accountability, it fostered further application of new knowledge and skills in teachers’ practice. Kleickmann and colleagues (2016) demonstrated that clearly structured PD that simultaneously emphasizes practical relevance is more likely to elicit teachers’ change than PD that only provides structural support. This interdependence suggests that understanding PD quality requires attention to both individual dimensions and their combinations as an integrated learning environment.

Perceived PD Quality and Teacher Engagement

High-quality PD activities are expected to enhance teachers’ professional competence, thereby improving teaching practices and student learning outcomes (Desimone, 2009; Guskey, 2002). While the ideal pathway seems straightforward, research suggests that not all PD succeeds in promoting educational improvement (Karsenty & Brodie, 2023). Engagement in PD is widely recognized as a key condition for teacher learning and for the transfer of PD content into classroom practice (Lipowsky & Rzejak, 2015). While often used interchangeably with participation, engagement represents a distinct construct. In PD settings, participation refers to teachers’ attendance or fulfillment of formal obligations in PD (Krille, 2020), whereas engagement describes a more voluntary investment of time, energy, and psychological resources in the learning opportunities provided, representing a multidimensional construct (Klassen et al., 2013; Liu et al., 2025). Emerging evidence further suggests that teachers’ engagement in PD is associated with indicators of PD effectiveness beyond attendance alone (Ji, 2021; Liu et al., 2025).

Existing research has established a close association between PD quality and teacher engagement (Antoniou & Kyriakides, 2013; Forte & Flores, 2014). For instance, Fütterer and colleagues (2024) found that higher perceived PD quality predicted greater engagement across both online and in-person formats. Similarly, Pellas (2025) found that teachers’ overall quality ratings of massive open online courses and learning management systems were significantly associated with higher autonomous learning. Nevertheless, the link between PD quality and engagement is not automatic. Some large-scale evaluations have shown that substantial investments in PD could sometimes yield limited engagement (Garet et al., 2008; Hill et al., 2013; Ryan, 2019). Brown and Herbst (2023) observed that despite providing clear goals and collaborative opportunities, teacher discussions often stalled or remained superficial. Similarly, Vivante and Vedder-Weiss (2023) found that a science PD program incorporating many recommended design features nonetheless motivated only moderate levels of teacher engagement.

To explain this variability, recent work has identified various individual and contextual factors, such as career stage (Vivante & Vedder-Weiss, 2023) and school culture (Picard & Kutsyuruba, 2017). While informative, these studies largely catalog correlates of engagement and often focus on relatively stable characteristics. As a result, they offer limited guidance on how researchers and practitioners can activate teachers’ engagement through changeable PD design. This underscores the need to unpack the professional learning process and explore malleable mechanisms linking the perceived environment to teacher engagement.

The Mediating Role of BPN Satisfaction

To move beyond correlates and toward malleable mechanisms, we turn to SDT, which offers a robust framework for unpacking these internal processes (Ryan & Deci, 2017). Central to SDT is the proposition that all individuals have three BPN: autonomy, competence, and relatedness (Deci & Ryan, 2012). Autonomy refers to the experience of volition and self-endorsement in one’s actions. In a PD context, this may manifest when teachers feel they have choices in their learning goals (Noonan, 2024). Competence involves feeling effective and capable. Teachers experience this when PD content enhances their practical skills and knowledge (Jaramillo-Baquerizo et al., 2021). Relatedness is the need to feel connected to and cared for by others. It could be satisfied when teachers experience a sense of belonging within the learning community and perceive that their contributions matter (Bayly et al., 2022). While autonomy, competence, and relatedness are conceptually distinct, SDT also emphasizes that they are interdependent and mutually reinforcing, producing a synergistic influence on outcomes (Deci & Ryan, 2012; Ryan & Deci, 2017). Empirical research has found high intercorrelations among the three needs, suggesting they often function as a coherent motivational system (Chen et al., 2015; Howard et al., 2025). In teacher PD specifically, program features typically support multiple needs simultaneously (Chiu et al., 2026; Power & Goodnough, 2019; Vedder-Weiss et al., 2025; Yang, 2021). This theoretical and empirical literature supports conceptualizing BPN satisfaction as a composite indicator of overall need satisfaction. It also aligns with our methodology consideration, which is described in section “Measures.”

SDT proposes that the satisfaction of these three needs is associated with positive motivational and behavioral outcomes (Deci & Ryan, 2012). This general pattern has been widely supported in broader educational and organizational settings, where need satisfaction has been linked to job engagement, well-being, and performance (Deci & Ryan, 2014; Slemp et al., 2021). Applied to the PD context, this suggests that when teachers’ needs are satisfied during professional learning activities, they are more likely to experience intrinsic motivation and sustained engagement. Emerging qualitative research in PD settings provides preliminary support for the needs-based pathway. For example, Brown and Herbst (2023) found that collaborative tools only fostered deep discussion when they addressed teachers’ needs for relevant knowledge. Jaramillo-Baquerizo and colleagues (2021) reported that PD activities closely aligned with daily practice strengthened teachers’ sense of competence, whereas abstract or disconnected content undermined this need, leading to disengagement. Furthermore, a growing body of research suggests need satisfaction as a robust predictor of sustained engagement and effort in professional learning (Bayly et al., 2022; Vedder-Weiss et al., 2025). In a case study, Noonan (2024) illustrated how responsive and encouraging facilitation helped a teacher feel recognized, which in turn was linked to deeper engagement. These findings suggest that teachers’ psychological experiences may serve as an important pathway linking perceived PD quality to teacher engagement.

While existing evidence supports BPN satisfaction as a general mediating pathway, an important further question is whether this pathway operates similarly across different PD experiences. A core tenet of SDT is the universality claim, which holds that the pathway from need satisfaction to outcomes is universal across situations and populations (Deci & Ryan, 2012). Nevertheless, emerging SDT research has begun to examine potential boundary conditions of the universality claim, investigating whether contextual factors moderate the strength of the needs–outcomes association (Ryan et al., 2025; Vansteenkiste et al., 2020). Research suggests that individuals with regularly satisfied BPN develop heightened sensitivity to need-supportive environments and gain greater psychological benefits, while those experiencing chronic need deprivation may become desensitized, limiting their ability to benefit from sporadic need-supportive experiences (Moller et al., 2010). This question is particularly relevant for PD settings, where teachers’ perceptions of quality may vary (Powell & Bodur, 2019; Ryan, 2019). Examining whether the BPN satisfaction–engagement pathway remains similar across teachers with different perceptions of PD quality would not only help refine SDT in this context but also provide more targeted guidance for PD design.

Despite the compelling qualitative support and theoretical rationale, the pathway linking perceived PD quality, BPN satisfaction, and teacher engagement has rarely been examined quantitatively in PD settings. In addition, few studies have explored whether the BPN satisfaction–engagement association operates similarly across different configurations of perceived PD quality.

The Conceptual Framework

Grounded in SDT, the present study aims to unpack teachers’ professional learning process by exploring how perceived PD quality relates to teacher engagement through BPN satisfaction and whether this path is consistent across different groups of teachers. Figure 1 presents the conceptual model. We conceptualize the four dimensions of PD quality (clarity and structure, cognitive activation, collaboration, and practical relevance) as perceived environmental antecedents associated with teacher engagement through teachers’ composite BPN satisfaction. To test this model, we employ a dual-perspective analytical strategy, combining variable-centered mediation (RQ1) to establish the general pathway with person-centered LPA (RQ2) to identify configurations of quality perceptions, and person-centered mediation (RQ3) to test the consistency of the pathway across profiles.

Figure 1.

Conceptual model.

Method

Participants

To address RQ1–3, we drew on a sample of in-service teachers participating in authentic PD courses. This naturalistic setting allowed us to capture variation in teachers’ perceptions of PD quality (RQ2) and their engagement responses (RQ1, RQ3).

We drew on data from a larger research project funded by the Ministry of School and Education of the German federal state of North Rhine-Westphalia. Although the data were collected as part of a larger validation project, the research questions and measures for the present study were specified a priori. We selected a stratified random sample of PD courses offered by state-run providers during the 2021–2022 school year. These courses covered a wide range of topics, such as subject matter education, school development, and classroom management, and addressed various target groups across different school types and grade levels.

We collected data online between February and July 2022. At the end of each course, participants were invited by their course facilitators to complete a standardized online questionnaire. Participation was voluntary and in accordance with all applicable data protection and ethical regulations. We employed a cross-sectional design for three main considerations. First, all three focal constructs capture teachers’ experiences within the same PD session, and thus fall within a single conceptual time window. Second, the research questions focus on the relationships between these concurrent experiences rather than changes over time, for which cross-sectional measurement is a commonly used approach. Third, because data collection was embedded in the regular state government PD system, we limited it to a single survey to reduce additional burden and avoid shortening PD instruction time. Nevertheless, we took several procedural measures to reduce potential bias. For example, data were collected immediately after each PD session to reduce retrospective recall bias while capturing teachers’ perceptions within the context of their specific PD experience. In addition, data were collected anonymously to reduce social desirability bias.

After excluding invalid responses (see section “Data analysis”; n = 51), the final sample comprised 985 in-service teachers from 85 PD courses. All participants were actively teaching at the time of the survey and represented a diverse cross-section of the teacher population in the federal state. We collected data on PD format (online, in person, or hybrid), gender, and teaching experience. According to program records provided by state agencies, typical PD courses ranged from 90 to 480 min (approximately 6 hr on average). Participation in PD courses was generally voluntary, but could be mandatory in a few cases. There were no economic or career-related rewards for participating in PD courses, and participation was not required for maintaining teacher certification. Course facilitators were all in-service teachers who had completed additional certifications and received a reduction in their teaching workload. However, information such as course duration, teachers’ motivation to participate, implementation fidelity, or instructor characteristics beyond baseline qualifications was not recorded at the individual level. Table 1 provides the demographic information.

Table 1.

Demographic Information of Participants.

Teacher characteristic	Frequency	Percentage
Gender
Female	662	67.2
Male	317	32.2
Diverse	6	0.6
Teaching experience
0–5 years	253	25.7
6–10 years	435	44.2
More than 10 years	297	30.2
PD formats
In person	757	76.9
Online	158	16.0
Hybrid	70	7.1

Note. Percentages have been rounded and may slightly exceed 100%.

Measures

Teachers’ Perceived Professional Development Quality

Teachers’ perceived PD quality served as the independent variable in RQ1 and RQ3, and as the basis for identifying quality perception profiles in RQ2. We measured perceived PD quality using the Teacher PD Monitor (Richter & Richter, 2024). We selected this instrument because it is theoretically grounded and has been validated with German teacher samples. It includes 18 items across four dimensions: clarity and structure (5 items, e.g., “The goals of the event were clearly stated,” Cronbach’s alpha = .80), practical relevance (4 items, e.g., “The course addressed problems that I encounter in my everyday professional life,” Cronbach’s alpha = .83), cognitive activation (6 items, e.g., “My prior knowledge was incorporated into the course,” Cronbach’s alpha = .84), and collaboration (3 items, e.g., “During the course, I had the opportunity to discuss the course content intensively with the other participants,” Cronbach’s alpha = .75). Teachers rated each item on a 4-point scale (1 = does not apply at all; 4 = fully applies). Higher scores indicate higher perceived PD quality. The structural validity of the TPD Monitor was acceptable (comparative fit index [CFI] = 0.96, Tucker–Lewis index [TLI] = 0.95, standardized root mean square residual [SRMR] = 0.05, root mean square error of approximation [RMSEA] = 0.06). The overall reliability in the current sample was good (Cronbach’s alpha = .90).

BPN Satisfaction

BPN satisfaction was the hypothesized mediator (RQ1, RQ3) linking perceived PD quality to teacher engagement. We measured BPN satisfaction using the Basic Psychological Needs Questionnaire (BPNQ; Korthagen & Evelein, 2016). We selected this instrument because it was specifically designed and validated for teacher populations, ensuring construct validity in the professional learning context. The BPNQ comprises 11 items across three dimensions: autonomy (4 items, e.g., “I was able to raise issues that are currently important to me,” Cronbach’s alpha = .73), competence (4 items, e.g., “The event helped me to expand my competencies,” Cronbach’s alpha = .79), and relatedness (3 items, e.g., “The atmosphere in the group was friendly and relaxed,” Cronbach’s alpha = .83). Teachers rated each item on the same 4-point scale (1 = does not apply at all; 4 = fully applies). Higher scores indicate higher need satisfaction. The structural validity was acceptable: CFI = 0.92, TLI = 0.89, SRMR = 0.07, RMSEA = 0.08. The overall reliability in the current sample was good (Cronbach’s alpha = .82).

In section “The Mediating Role of BPN Satisfaction,” we present the theoretical and empirical evidence for treating BPN as a composite variable reflecting overall need satisfaction. This decision was also guided by methodological considerations. The current study uses the Bolck–Croon–Hagenaars (BCH) method (detailed in section “Data analysis”) for person-centered mediation analysis, where model parsimony is critical for estimation stability (McLarnon & O’Neill, 2018). This composite approach balances theoretical alignment with analytical feasibility and is consistent with recent SDT applications in complex models (e.g., Lin et al., 2023; Moè & Katz, 2022).

Teacher Engagement

Teacher engagement in PD served as the outcome variable in RQ1 and RQ3. To capture the nuance of teacher professional learning, studies recognized a pressing need to move beyond traditional objective evaluations and center the authentic voices of teachers (Kennedy, 2016; Opfer & Pedder, 2011). Accordingly, we utilized a self-report survey design. In addition, it is suited for the present study because the construct encompasses not only observable behaviors but also cognitive and affective dimensions, which are internal psychological states best captured through direct self-assessment. While behavioral engagement can sometimes be observed externally, observation alone risks capturing performative compliance rather than authentic psychological investment. This self-report approach aligns with established practices in teacher professional learning research (e.g., Fütterer et al., 2024; Ji, 2021) and organizational engagement research (Schaufeli et al., 2002).

We assessed engagement using the scale adapted from Chan and colleagues (2023), which captures the multidimensional nature of engagement and has demonstrated strong psychometric properties with German teacher samples (Fütterer et al., 2024). It consists of 11 items across three dimensions: behavioral engagement (4 items, e.g., “I actively participated in the event,” Cronbach’s alpha = .80), cognitive engagement (3 items, e.g., “I thought deeply about the content during the event,” Cronbach’s alpha = .75), and affective engagement (4 items, e.g., “I found what I learned in the event interesting,” Cronbach’s alpha = .81). Teachers rated each item on a 4-point scale (1 = does not apply at all; 4 = fully applies). As with BPN, we treated teacher engagement as a composite variable. Higher scores indicate greater engagement. The structural validity was acceptable (CFI = 0.96, TLI = 0.95, SRMR = 0.04, RMSEA = 0.06), and the reliability was reasonable (Cronbach’s alpha = .85).

Covariates

We included teacher and course characteristics as covariates because prior research suggests that they are associated with perceived PD quality and engagement (Fütterer et al., 2024; Kwakman, 2003; Mulaimović et al., 2025). Specifically, we controlled for gender and teaching experience (0 = 0–5 years, 1 = 6–10 years, 2 = more than 10 years) as teacher characteristics, and PD format (0 = in person, 1 = online, 2 = hybrid) as a course characteristic.

Data Analysis

We conducted the analysis in four phases. We ran descriptive analyses in SPSS 28 (IBM Corp, 2021), and we conducted mediation analyses and LPA in Mplus 8.3 (Muthén & Muthén, 1998–2017).

Phase 1: Preliminary Analyses

We cleaned the data and evaluated psychometric properties. The initial dataset included 1,036 responses. We applied the three-sigma rule to identify extreme outliers and removed cases with more than 80% missing data. This resulted in 51 exclusions and a final sample of 985 teachers. We assessed internal consistency using Cronbach’s alpha and evaluated structural validity using confirmatory factor analysis (CFA). We judged model fit using standard cutoffs (CFI ≥ 0.90, TLI ≥ 0.90, SRMR ≤ 0.08, RMSEA ≤ 0.08). To assess common method bias, we conducted Harman’s single-factor test (Podsakoff et al., 2003). A one-factor exploratory model showed that the first factor explained 36% of the variance, below the 50% threshold, suggesting that common method variance is unlikely to fully account for the observed associations. We then computed means, standard deviations, and correlations as descriptive evidence for subsequent analyses.

Phase 2: Variable-Centered Mediation Analysis

To address RQ1, we employed a variable-centered approach to test whether BPN satisfaction mediates the relationship between PD quality and engagement. We first estimated the direct associations of PD quality dimensions with engagement and then tested indirect effects via BPN satisfaction.

Phase 3: Person-Centered Latent Profile Analysis

To address RQ2, we conducted LPA using the four PD quality dimensions as indicators to identify subgroups of teachers with distinct patterns of perceived PD quality. We selected the number of profiles based on the Akaike information criterion (AIC), Bayesian information criterion (BIC), adjusted Bayesian information criterion (aBIC), entropy values, and theoretical reasonableness. Lower values of AIC, BIC, and aBIC, along with an entropy value closer to 1, indicate a better model fit (Spurk et al., 2020). We also considered the bootstrap likelihood ratio test (BLRT), which has been shown to outperform other indicators in identifying the correct number of classes in mixture modeling (Nylund et al., 2007). A significant p-value indicates a better estimated model fit compared to a model with one less profile. Because large sample sizes can influence fit statistics, we used an elbow plot to further aid in determining the optimal number of profiles (Spurk et al., 2020).

Phase 4: Person-Centered Mediation Analysis

To address RQ3, we tested whether the mediation pathway identified in RQ1 operated consistently across the profiles. We used the manual 3-step BCH approach (Bolck et al., 2004) to account for classification uncertainty (Asparouhov & Muthén, 2014; McLarnon & O’Neill, 2018). First, we re-estimated the selected LPA model and saved the BCH weights. Second, the saved BCH weights were incorporated via the TRAINING option in Mplus to estimate the auxiliary structural model while adjusting for classification error. Third, we specified a model in which BPN satisfaction was regressed on profile membership and covariates, and teacher engagement was regressed on BPN satisfaction, profile membership, and covariates. Because the BCH procedure does not directly yield standardized estimates, BPN satisfaction and teacher engagement were standardized prior. Accordingly, the coefficient for the path from BPN satisfaction to teacher engagement can be interpreted as a standardized regression coefficient, whereas the coefficients comparing profiles represent differences in standardized BPN or teacher engagement units relative to the reference profile rather than fully standardized coefficients. Profile membership was represented by dummy variables, with one profile serving as the reference group.

In addition, we extended the mediation model to a moderated mediation model to test potential boundary conditions by examining whether profile membership moderated the BPN–engagement association. To achieve this, we established the profile-specific paths between BPN satisfaction and teacher engagement. Thus, moderation was represented by between-profile differences in the BPN-engagement slope, analogous to interaction effects in a dummy-coded regression framework (McLarnon & O’Neill, 2018). Finally, we compared model fit using a log-likelihood ratio test (LLRT; Meng & Rubin, 1992). The model constraining the BPN–engagement path to be equal across profiles served as the null model, and the model allowing profile-specific paths served as the alternative model.

Our data had a hierarchical structure, with teachers nested within PD programs. We utilized the “TYPE = COMPLEX MIXTURE” option in Mplus to adjust standard errors for clustering (Muthén & Muthén, 1998–2017). This approach allowed us to obtain cluster-robust standard errors, accounting for the nested data structure.

Missing data ranged from 0.2% to 2.1% across variables, and 9.3% of cases had missing values on at least one variable. We handled it through full information maximum likelihood (FIML) and the multiple imputation technique. FIML was initially applied to variable-centered analysis and LPA, maximizing available information to address missing data. Based on LPA results and BCH weights, we generated 50 imputed datasets incorporating key variables and covariates such as gender and teaching experience. Including BCH weights contributes to leveraging the existing correlations within the dataset, potentially enhancing imputation accuracy and the reliability of parameter estimates (Asparouhov & Muthén, 2014). We reran the variable-centered models and LPA using the imputed datasets, and the substantive pattern of results remained unchanged. Because the findings were consistent across approaches, we report the estimates based on the imputed data to reduce potential bias due to missingness.

Results

Preliminary Analysis

The preliminary analysis examined the descriptive statistics and correlations among the key variables. Table 2 presents the means, standard deviations, and correlations for all studied variables. All four dimensions of PD quality were positively associated with both BPN satisfaction and teacher engagement. Teachers’ BPN satisfaction was also strongly correlated with their engagement during the PD.

Table 2.

Descriptive Statistics and Correlations of the Variables.

No	Variable	M	SD	1	2	3	4	5	6
1	Clarity and structure	3.56	0.47	-
2	Practical relevance	3.47	0.53	0.58***	-
3	Collaboration	3.70	0.48	0.33***	0.26***	-
4	Cognitive activation	3.31	0.53	0.62***	0.63***	0.39***	-
5	Basic Psychological Needs	3.48	0.38	0.57***	0.57***	0.40***	0.68***	-
6	Teacher Engagement	3.38	0.44	0.60***	0.61***	0.35***	0.68***	0.68***	-

Note. N = 985. ***p < .001.

Variable-Centered Mediation Analysis

To examine the associations between perceived PD quality and teacher engagement (RQ1), we estimated models while controlling for gender, teaching experience, and PD format (complete results are shown in Supplemental Appendix 1). We first tested the total associations of the four PD quality dimensions with engagement in a model without BPN. The analysis revealed that cognitive activation (β = 0.366, p < .001), practical relevance (β = 0.216, p < .001), and clarity and structure (β = 0.179, p < .001) were significantly associated with teacher engagement. Collaboration showed a weaker yet statistically significant positive association (β = 0.063, p = .018).

We then examined the mediating role of BPN satisfaction, as illustrated in Figure 2. The results indicated that all PD quality dimensions were positively associated with BPN satisfaction, which in turn was positively associated with teacher engagement (β = 0.300, p < .001). Notably, significant indirect effects were observed for each PD quality dimension on teacher engagement through BPN satisfaction: clarity and structure (β = 0.047, p < .001), practical relevance (β = 0.051, p = .001), cognitive activation (β = 0.124, p < .001), and collaboration (β = 0.040, p < .001). This suggests that teachers who perceived higher-quality PD programs reported higher BPN satisfaction, which was in turn linked to higher levels of engagement. Regarding the control variables, compared with novice teachers (0–5 years), highly experienced teachers (>10 years) reported slightly lower BPN satisfaction (β = −0.075, p = .018) but higher overall engagement (β = 0.099, p < .001). Online PD formats were negatively associated with engagement compared to in-person formats (β = −0.114, p < .001).

Figure 2.

Variable-centered mediation model between PD quality dimensions, BPN, and teacher engagement.

Latent Profiles of Teachers’ Perceived PD Quality

To address RQ2, which examines the heterogeneity in teachers’ PD quality perceptions, we conducted LPA to identify distinct subgroups of teachers based on their ratings of the four PD quality dimensions.

The fit indices for the LPA models are presented in Table 3. After comparing these indices, the four-profile model was selected as the optimal solution. The three-profile model was excluded because the elbow plot (as shown in Supplemental Appendix 2) suggested that the fit only began to stabilize after the four- and five-profile models. Furthermore, the significant BLRT result of the four-profile model compared to the three-profile model suggesting four-profile model fitted better than the three-profile model. The five-, six-, and seven-profile models were also evaluated but were excluded because their smallest group comprised less than 5% of the overall sample, which is below the acceptable threshold for meaningful interpretation (Spurk et al., 2020). In addition, the BLRT results for the five-, six-, and seven-profile models showed that no replicated optimal log-likelihood values were found in the bootstrap samples. This suggests that the P-values for these models may be unreliable, which could lead to inaccurate model evaluations. Consequently, the four-profile solution was chosen as the final model as it exhibited relatively lower AIC, BIC, and aBIC values, reasonable entropy, and a significant p-value for the BLRT.

Table 3.

Fit Indices for Teachers’ Perceived PD Quality Profiles.

No.	Loglikelihood	AIC	BIC	ABIC	Entropy	p BLRT	Proportion
1	–2,857.96	5,731.91	5,771.05	5,745.64	1.00	—	—
2	–2,291.10	4,608.19	4,671.80	4,630.51	0.84	<0.001	0.29/0.71
3	–2,127.66	4,291.31	4,379.38	4,322.21	0.83	<0.001	0.09/0.34/0.57
4	–1,964.94	3,975.88	4,088.41	4,015.36	0.87	<0.001	0.07/0.09/0.28/0.56
5	–1,882.57	3,821.13	3,958.12	3,869.19	0.90	—	0.04/0.08/0.09/0.23/0.55
6	–1,801.99	3,669.97	3,831.43	3,726.62	0.91	—	0.03/0.03/0.08/0.10/0.23/0.54
7	–1,729.49	3,534.97	3,720.89	3,600.21	0.91	—	0.02/0.04/0.04/0.06/0.08/0.24/0.52

Note. N = 985. AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; ABIC = Adjusted BIC; BLRT = Bootstrap Likelihood Ratio Test.

The selected LPA model revealed four distinct profiles of teachers’ PD quality perceptions, as illustrated in Figure 3. Detailed descriptive statistics for each profile are provided in Supplemental Appendix 3.

Figure 3.

Latent profiles of teachers’ perceived PD quality.

Profile 1: High-Quality Balanced Perceivers (n = 555, 56%). Teachers in this profile reported the highest overall perceived PD quality (M = 3.78), and consistently high scores on all four dimensions. This suggests they experienced well-designed PD courses that simultaneously provided clear structure, practical relevance, cognitive challenge, and collaborative opportunities.

Profile 2: Moderate-Quality Weak-Collaboration Perceivers (n = 67, 7%). This small group of teachers reported moderate overall perceived PD quality (M = 3.28), with higher scores in practical relevance (M = 3.62) and clarity and structure (M = 3.68), but notably lower scores in collaboration (M = 2.58). These teachers appeared to have attended PD courses that were well-organized and relevant but lacked opportunities for meaningful peer interaction.

Profile 3: Moderate-Quality Strong-Collaboration Perceivers (n = 273, 28%). Teachers in this profile also reported moderate overall perceived quality (M = 3.28) but showed relatively strong perceptions of collaborative opportunities (M = 3.78). In contrast to Moderate-Quality Weak-Collaboration Perceivers, these teachers experienced PD sessions with rich collaborative activities but relatively lower clarity, relevance, and cognitive activation.

Profile 4: Low-Quality Strong-Collaboration Perceivers (n = 90, 9%). This profile was characterized by consistently low scores in clarity and structure (M = 2.67), practical relevance (M = 2.67), and cognitive activation (M = 2.51), yet a relatively higher perception of collaboration quality (M = 2.99). These teachers attended PD courses that, while fostering a sense of community, fell short in providing clear goals, relevant content, or intellectual challenge.

Together, these four profiles reveal that teachers differ meaningfully in how they perceive PD quality. While 56% of teachers experienced consistently high quality, the remaining 44% exhibited mixed or uniformly low perceptions, with collaboration emerging as a dimension that varied independently of the other three quality features. These findings suggest that teachers’ quality perceptions are not uniformly distributed but cluster into distinct experiential patterns.

Person-Centered Mediation Analysis

After identifying the latent profiles, we examined the mediating role of BPN satisfaction in the relationship between teachers’ perceived PD quality profiles and teacher engagement (RQ3). We used the High-Quality Balanced Perceivers as the reference group to facilitate interpretation of the model. Supplemental Appendix 4 presents the full results of the person-centered mediation analysis with covariates included. Teaching experience and PD format remained significantly associated with BPN satisfaction and engagement, while gender was not significant.

As shown in Figure 4, all three remaining profiles were negatively associated with both BPN satisfaction and engagement relative to the High-Quality Balanced Perceivers, indicating that the reference profile was associated with the highest levels of both outcomes. The magnitude of these negative associations followed a consistent gradient: the Moderate-Quality Weak-Collaboration Perceivers showed the smallest differences, followed by Moderate-Quality Strong-Collaboration Perceivers, with Low-Quality Strong-Collaboration Perceivers exhibiting the largest differences. BPN satisfaction was positively associated with teacher engagement across profiles, supporting the BPN–engagement pathway. Regarding indirect effects, relative to the High-Quality Balanced Perceivers, Moderate-Quality Weak-Collaboration Perceivers (B = −0.260, p < .001), Moderate-Quality Strong-Collaboration Perceivers (B = −0.397, p < .001), and Low-Quality Strong-Collaboration Perceivers (B = −0.686, p < .001) were all negatively and significantly related to teacher engagement through BPN satisfaction. To directly compare the two moderate-quality profiles, we conducted an additional analysis using Moderate-Quality Strong-Collaboration Perceivers as the reference group. This comparison showed that the Moderate-Quality Weak-Collaboration profile reported significantly higher BPN satisfaction (B = 0.347, p = .008) and teacher engagement (B = 0.387, p = .006) than the Moderate-Quality Strong-Collaboration profile.

Figure 4.

Person-centered mediation model between profiles, BPN, and teacher engagement.

We further tested whether PD quality profiles moderated the association between BPN satisfaction and teacher engagement by estimating profile-specific BPN–engagement paths. This analysis allowed us to examine a potential boundary condition of the SDT universality claim in this context. The results showed no significant differences in the BPN–engagement paths between the reference group and Moderate-Quality Weak-Collaboration Perceivers (B = −0.065, p = .639), Moderate-Quality Strong-Collaboration Perceivers (B = −0.039, p = .742), and Low-Quality Strong-Collaboration Perceivers (B = −0.133, p = .175). Furthermore, the LLRT revealed that the unconstrained model did not fit significantly better than the constrained model (p = .640). These results indicate that the association between BPN satisfaction and teacher engagement remained consistent across profiles, providing no evidence of a moderating effect.

Discussion

The present study aimed to explore and unpack the association between teachers’ perceived PD quality and their engagement through the lens of SDT. The findings were consistent with the hypothesized psychological pathway and helped clarify whether this pattern remained similar across teachers with different perceived PD quality profiles, offering theoretical insights and practical guidance for PD design.

Mediating Pathway Between Perceived PD Quality Dimensions and Teacher Engagement (RQ1)

To address RQ1, we used a variable-centered approach to examine the associations between PD quality dimensions, BPN satisfaction, and teacher engagement. Consistent with prior research, the four PD quality dimensions, clarity and structure, cognitive activation, practical relevance, and collaboration, were positively associated with engagement (Fütterer et al., 2024; Mulaimović et al., 2025). More importantly, BPN satisfaction significantly mediated the association between each PD quality dimension and engagement. This mediation pattern provides quantitative evidence for the needs-based pathway that has frequently been suggested in qualitative work (e.g., Noonan, 2024; Power & Goodnough, 2019; Yang, 2021).

The results also point to meaningful differences in the relative strength of PD quality dimensions. Cognitive activation showed the strongest association with engagement. One possible interpretation relates to the distinctive nature of teachers as adult professional learners. Unlike novice learners who may prioritize clear guidance and structured support, experienced teachers possess a wealth of prior knowledge and practical expertise (Borko, 2004). They may therefore find greater psychological reward in PD that challenges them intellectually, inviting them to critique existing beliefs and construct new understandings (Ebert & Crippen, 2010). Conversely, although collaboration was positively associated with engagement, its association was the weakest among the four dimensions. This appears to diverge from the central role collaboration plays in many influential PD frameworks (Darling-Hammond et al., 2017; Desimone, 2009). However, this does not imply that collaboration is unimportant, but rather that its contributions vary depending on the circumstances (Chong & Kong, 2012; Picard & Kutsyuruba, 2017), suggesting the need to further examine the contexts in which collaboration occurs.

Beyond the primary findings, a secondary finding emerged regarding teaching experience. Compared with novice teachers, teachers with more than 10 years of experience reported slightly lower BPN satisfaction but higher engagement. This pattern is broadly consistent with prior work showing that novice and experienced teachers often differ in how they perceive and engage with PD (Mahmoudi & Özkan, 2015). Novice teachers may be more dependent on immediate contextual supports to build confidence and navigate learning opportunities, whereas experienced teachers may remain engaged by drawing on accumulated professional experience and stronger self-efficacy beliefs (Tschannen-Moran & Hoy, 2007). Although this finding should be interpreted cautiously as a secondary exploration rather than a primary hypothesis, it suggests that career stage may shape how teachers experience and engage in the same PD environment.

The variable-centered findings identified a general needs-based pathway linking perceived PD quality to engagement, while also suggesting that the configuration of PD features may matter. We therefore turned to a person-centered approach to examine how these dimensions clustered in teachers’ perceptions of PD and whether the needs-based pathway remained evident across subgroups.

Distinct Profiles of Teachers’ Perceived PD Quality (RQ2)

Addressing RQ2, the person-centered analysis identified four distinct profiles: High-Quality Balanced Perceivers, Moderate-Quality Weak-Collaboration Perceivers, Moderate-Quality Strong-Collaboration Perceivers, and Low-Quality Strong-Collaboration Perceivers. These profiles demonstrate the heterogeneity in teachers’ perceptions of PD quality and reveal two notable patterns.

First, a pattern of synergy emerged in the largest subgroup, the High-Quality Balanced Perceivers. Teachers in this profile reported uniformly high perceptions across all four quality dimensions. This suggests that when teachers experience PD as high quality, they tend to perceive multiple dimensions positively at the same time. Such a pattern aligns with theoretical arguments that effective PD depends on the coordinated integration of multiple design features rather than the isolated presence of any single feature (Darling-Hammond et al., 2017; Garet et al., 2001; Walter & Briggs, 2012).

Second, collaboration emerged as the dimension showing the greatest heterogeneity. On one hand, collaboration was frequently perceived as a prominent feature by most teachers (Voogt et al., 2015), as evidenced by its relative strength in the High-Quality Balanced, Moderate-Quality Strong-Collaboration, and Low-Quality Strong-Collaboration profiles (encompassing nearly 93% of the sample). On the other hand, the existence of the Moderate-Quality Weak-Collaboration profile shows that collaboration is not uniformly experienced even when other aspects of PD are perceived as relatively adequate. This suggests that teachers’ perceptions of collaboration may vary more independently than the other dimensions and should not be assumed to co-occur automatically with otherwise positive PD experiences (Little, 1982). The profile solution thus highlights the value of examining PD quality as a configuration of features rather than as a set of isolated dimensions. Given this heterogeneity, it is important to examine whether the needs-based pathway identified in RQ1 remains consistent across different profiles.

Consistency of the Pathway Between Teachers’ Perceived PD Quality Profiles and Engagement (RQ3)

RQ3 extended the variable-centered findings by examining whether the needs-based pathway operated consistently across the identified PD quality profiles. The profile-based mediation analysis showed that the pathway from perceived PD quality to engagement through BPN satisfaction was evident across all four profiles. Importantly, we found no significant differences in the path strength of BPN-engagement across different profiles. This finding aligns with the universality claim of SDT and recent evidence, which holds that need satisfaction benefits are broadly applicable across environments (Howard et al., 2025; Ryan et al., 2025). One possible explanation for the absence of significant moderation effects is that previous significant moderation effects were found in family environments (Bülow et al., 2022), daily social interactions (Moller et al., 2010), or a deeply ingrained cultural background (Ryan et al., 2025), which represents exposure in long-term environments (Moller et al., 2010; Ryan et al., 2025). In contrast, most PD activities in our sample were short-term experiences, which may affect teachers’ levels of need satisfaction without substantially altering the strength of the association between BPN satisfaction and engagement. This interpretation, however, remains preliminary and requires validation in future longitudinal studies.

Although the needs–engagement link remained consistent, the profiles differed clearly in their levels of BPN satisfaction and engagement, following a consistent ordering: High-Quality Balanced > Moderate-Quality (both the Weak-Collaboration and Strong-Collaboration Perceivers) > Low-Quality Strong-Collaboration Perceivers. This pattern indicates that differences in engagement across profiles do not stem from differences in how strongly BPN satisfaction was associated with engagement, but rather from both the direct perceived affordances of the PD environment and the corresponding levels of need satisfaction. A balanced and high-quality environment, where all dimensions work in synergy, offers the most supportive conditions for need satisfaction and, in turn, for sustained engagement (Noonan, 2024; Vedder-Weiss et al., 2025).

These profile-based findings also help clarify the relatively weaker role of collaboration observed in the variable-centered analysis. Despite sharing similar overall quality ratings, the Moderate-Quality Weak-Collaboration Perceivers reported more favorable needs satisfaction and engagement outcomes than the Moderate-Quality Strong-Collaboration Perceivers. This suggests that relatively low perceived collaboration may still co-occur with favorable outcomes when other aspects of the overall environment are well-designed. In contrast, strong perceived collaboration alone may not be sufficient when other high-quality design features are perceived as weaker (Little, 1982). Because variable-centered models estimate average, linear relationships across the entire sample (Bergman & Trost, 2006), such compensatory patterns may weaken the overall statistical effect of collaboration. These results collectively suggest that collaboration is not an automatic catalyst for favorable outcomes; it requires intentional structuring and integration with other quality dimensions (Chong & Kong, 2012; Forte & Flores, 2014).

Contributions, Implications, and Limitations

The results of the present study contribute to both the PD literature and SDT-informed research. For the PD field, our findings respond to calls to move beyond defining good PD merely as a set of design features and toward explaining the internal psychological processes of teachers’ professional learning (Kennedy, 2016; Thurlings & den Brok, 2017). By identifying BPN satisfaction as a mediator, we provide a psychologically grounded explanation for the association between teachers’ perceived PD quality and their engagement. For SDT research, this study extends needs-based inquiry into the context of teacher PD and provides initial evidence that the needs–engagement link remained similar across diverse configurations of perceived PD quality.

The findings also suggest two main implications for PD practice. First, these findings enrich the understanding of the psychological pathway to teachers’ engagement in PD through the lens of SDT. We found that this pathway remained consistent across all identified teacher profiles, regardless of how teachers perceived the quality dimensions of PD. This highlights the value of designing PD as needs-supportive learning environments. While we focused on the general pathway, previous research has provided specific guidance on how PD can support teachers’ needs (e.g., Chiu et al., 2026; Jaramillo-Baquerizo et al., 2021; Noonan, 2024; Yang, 2021). For example, when PD provides clear and transparent goals and allows teachers to make choices about these goals, it helps enhance their perception of PD clarity and structure while designing it in a way to fulfill teachers’ needs for autonomy. (Noonan, 2024). Future research could examine which specific design elements within each PD quality dimension most effectively support each psychological need.

Second, our findings highlight the importance of considering PD programs as integrated configurations of quality features rather than as isolated design elements. A key insight is that strong collaboration embedded in otherwise low-quality contexts was associated with less favorable outcomes than more moderately collaborative contexts supported by stronger structure and relevance. This implies that collaboration should not be treated as a standalone design goal, but rather purposefully constructed and aligned with other quality dimensions (Vangrieken et al., 2015; Voogt et al., 2015). For policymakers and PD designers, these results highlight the importance of building balanced PD environments, where collaboration is integrated into coherent learning designs that provide clear and practice-relevant goals.

Despite these contributions, this study has several limitations. First, the cross-sectional design limits our ability to establish causal relationships. Although the mediation model was theoretically grounded in SDT, the present data cannot determine whether perceived PD quality precedes BPN satisfaction and subsequent engagement over time. Future studies could use longitudinal or experimental designs to trace temporal ordering and test causal assumptions. Second, while relying on self-reported questionnaire data was conceptually appropriate for capturing teachers’ internal psychological states and centering their voices, this single-method approach introduces potential biases. Future research could incorporate diverse data collection methods such as teacher interviews, field observations, and learning journals. Third, we did not collect data on broader contextual factors that may shape teachers’ PD experiences, such as course duration, participation motivation, and instructor characteristics. These variables may serve as antecedents to teachers’ quality perceptions or as moderators of the identified associations. Future research could collect and examine these contextual factors to clarify the relationship between objective design and teachers’ subjective perceptions. Finally, while our German sample provides valuable insights, the generalizability of our findings requires careful consideration. Future research could examine whether similar profiles emerge in educational systems with different cultural values and institutional structures.

Supplemental Material

sj-docx-1-jte-10.1177_00224871261458811 – Supplemental material for Beyond Quality Features, Toward Teacher Needs: Unpacking How Professional Development Quality Relates to Teacher Engagement Through Self-Determination Theory

Supplemental material, sj-docx-1-jte-10.1177_00224871261458811 for Beyond Quality Features, Toward Teacher Needs: Unpacking How Professional Development Quality Relates to Teacher Engagement Through Self-Determination Theory by Kexin Li, Eric Richter, Dirk Richter, Yizhen Huang and Gaowei Chen in Journal of Teacher Education

Footnotes

ORCID iDs

Kexin Li

Eric Richter

Dirk Richter

Yizhen Huang

Gaowei Chen

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research received funding from the Ministry for School and Education of the State of North Rhine Westphalia (Ministerium für Schule und Bildung des Landes Nordrhein Westfalen).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Kexin Li is a PhD student in the Faculty of Education at The University of Hong Kong. Grounded in the learning sciences, her research focuses on teacher professional development, learning motivation, and classroom discourse. Specifically, she explores the cognitive and psychological mechanisms underlying teacher learning and how artificial intelligence (AI) can be leveraged to enhance this learning process. Her current work involves investigating teachers’ attitudes and interactions with AI, as well as designing AI-driven platforms to support teachers’ professional learning.

Eric Richter is professor of School Pedagogy at the Catholic University of Eichstätt Ingolstadt, Germany. His research focuses on teacher education and professional development across career stages. He also examines teacher shortages and alternative pathways into the profession.

Dirk Richter is a professor of educational research in the educational sciences at the University of Potsdam in Germany. His research focuses on teacher professional development and teacher training with a special focus on the use of digital media for learning processes. He is involved in national research programs that aim to foster schools with disadvantaged children and that develop inservice training programs for teachers. He serves on the advisory board of the Ministry of Education, Youth and Sport in the state of Brandenburg and he is editor in chief of the German Journal of Educational Research.

Yizhen Huang is an assistant professor in education sciences and psychology at the Hector Research Institute of Education Sciences and Psychology (University of Tübingen) in Germany. She pursues research that deepens understanding of the cognitive, motivational, and affective dynamics of classroom learning and teaching. Her methodology integrates multimodal data from randomized controlled trials conducted in both extended reality (XR) simulations and real-world classrooms with machine learning (ML) approaches. Through this work, she develops simulation-based models of professional learning and effective applications of artificial intelligence (AI) in classrooms.

Gaowei Chen is Tin Ka Ping Foundation Outstanding Young Professor at the Faculty of Education, University of Hong Kong, and Head of the Academic Unit of Mathematics, Science, and Technology. His research interests include dialogic teaching and learning, teacher professional development, learning sciences, computer-supported collaborative learning, and AI in education. He received Research Output Prize and Faculty Outstanding Young Researcher Award from the University of Hong Kong. He currently serves as an executive editor of Teaching and Teacher Education and an associate editor of the Journal of the Learning Sciences. He is also an international associate of the Cambridge Educational Dialogue Research Group (CEDiR) at the University of Cambridge.

References

Antoniou

Kyriakides

(2013). A dynamic integrated approach to teacher professional development: Impact and sustainability of the effects on improving teacher behaviour and student outcomes. Teaching and Teacher Education, 29, 1–12.

Asparouhov

Muthén

(2014). Auxiliary variables in mixture modeling: Using the BCH method in Mplus to estimate a distal outcome model and an arbitrary secondary model. Mplus Web Notes, 21(2), 1–22.

Bayly

B. L.

Dizon

Shrestha

Smith

C. L.

Tekle

Rhoades Cooper

(2022). Leveraging self-determination theory to understand which preschool teachers benefit most from a professional development intervention. Journal of Early Childhood Teacher Education, 43(2), 194–212.

Beisiegel

Mitchell

Hill

H. C

. (2018). The design of video-based professional development: An exploratory experiment intended to identify effective features. Journal of Teacher Education, 69(1), 69–89.

Bergman

L. R.

Trost

(2006). The person-oriented versus the variable-oriented approach: Are they complementary, opposites, or exploring different worlds? Merrill-Palmer Quarterly, 52(3), 601–632.

Bolck

Croon

Hagenaars

(2004). Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political Analysis, 12(1), 3–27.

Borko

(2004). Professional development and teacher learning: Mapping the terrain. Educational Researcher, 33(8), 3–15.

Boström

Palm

(2020). Expectancy-value theory as an explanatory theory for the effect of professional development programmes in formative assessment on teacher practice. Teacher Development, 24(4), 539–558.

Brown

A. M.

Herbst

P. G.

(2023). On designing better structures for feedback in practice-based professional development: Using “failure” to innovate. Journal of Mathematics Teacher Education, 26(5), 581–605.

10.

Bülow

Neubauer

A. B.

Soenens

Boele

Denissen

J. J.

Keijsers

(2022). Universal ingredients to parenting teens: Parental warmth and autonomy support promote adolescent well-being in most families. Scientific Reports, 12(1), 16836.

11.

Chan

Maneewan

Koul

(2023). Teacher educators’ teaching styles: Relation with learning motivation and academic engagement in pre-service teachers. Teaching in Higher Education, 28(8), 2044–2065.

12.

Chen

Vansteenkiste

Beyers

Boone

Deci

E. L.

Van der Kaap-Deeder

Duriez

Lens

Matos

Mouratidis

Ryan

R. M.

Sheldon

K. M.

Soenens

Van Petegem

Verstuyf

. (2015). Basic psychological need satisfaction, need frustration, and need strength across four cultures. Motivation and Emotion, 39(2), 216–236. https://doi.org/10.1007/s11031-014-9450-1

13.

Chiu

T. K. F.

Bali

Tondeur

Howard

Chan

K. K. H

. (2026). Teacher education for artificial intelligence literacy through a self-determination theory perspective. European Journal of Teacher Education, 1–21. https://doi.org/10.1080/02619768.2026.2621848

14.

Chong

W. H.

Kong

C. A.

(2012). Teacher collaborative learning and teacher self-efficacy: The case of lesson study. The Journal of Experimental Education, 80(3), 263–283.

15.

Darling-Hammond

Hyler

M. E.

Gardner

(2017). Effective teacher professional development. Learning Policy Institute. https://learningpolicyinstitute.org/product/effective-teacher-professional-development-report

16.

Deci

E. L.

Ryan

R. M.

(2012). Self-determination theory. In Van Lange

P. A. M.

Kruglanski

A. W.

Higgins

E. T.

(Eds.), Handbook of theories of social psychology (Vol. 1, pp. 416–436). Sage.

17.

Deci

E. L.

Ryan

R. M.

(2014). The importance of universal psychological needs for understanding motivation in the workplace. In Gagné

(Ed.), The Oxford handbook of work engagement, motivation, and self-determination theory (pp. 13–32). Oxford University Press.

18.

Desimone

L. M.

(2009). Improving impact studies of teachers’ professional development: Toward better conceptualizations and measures. Educational Researcher, 38(3), 181–199.

19.

Ebert

E. K.

Crippen

K. J.

(2010). Applying a cognitive-affective model of conceptual change to professional development. Journal of Science Teacher Education, 21, 371–388.

20.

Forte

A. M.

Flores

M. A.

(2014). Teacher collaboration and professional development in the workplace: A study of Portuguese teachers. European Journal of Teacher Education, 37(1), 91–105.

21.

Fütterer

Richter

(2024). Teachers’ engagement in online professional development—The interplay of online professional development quality and teacher motivation. Zeitschrift für Erziehungswissenschaft, 27(3), 739–768.

22.

Garet

M. S.

Cronen

Eaton

Kurki

Ludwig

Jones

Uekawa

Falk

Bloom

H. S.

Doolittle

Zhu

Sztejnberg

. (2008). The impact of two professional development interventions on early reading instruction and achievement (NCEE 2008-4030). National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education. https://ies.ed.gov/ncee/pdf/20084030.pdf

23.

Garet

M. S.

Porter

A. C.

Desimone

Birman

B. F.

Yoon

K. S.

(2001). What makes professional development effective? Results from a national sample of teachers. American Educational Research Journal, 38(4), 915–945.

24.

Gaumer Erickson

A. S.

Noonan

P. M.

Brussow

Supon Carter

. (2017). Measuring the quality of professional development training. Professional Development in Education, 43(4), 685–688.

25.

Guskey

T. R.

(2002). Professional development and teacher change. Teachers and Teaching, 8(3), 381–391.

26.

Hill

H. C.

Beisiegel

Jacob

(2013). Professional development research: Consensus, crossroads, and challenges. Educational Researcher, 42(9), 476–487.

27.

Hill

H. C.

Corey

D. L.

Jacob

R. T.

(2018). Dividing by zero: Exploring null results in a mathematics professional development program. Teachers College Record, 120(6), 1–42.

28.

Howard

J. L.

Slemp

G. R.

Wang

(2025). Need support and need thwarting: A meta-analysis of autonomy, competence, and relatedness supportive and thwarting behaviors in student populations. Personality and Social Psychology Bulletin, 51, 1552–1573.

29.

Howard

M. C.

Hoffman

M. E.

(2018). Variable-centered, person-centered, and person-specific approaches: Where theory meets the method. Organizational Research Methods, 21(4), 846–876.

30.

IBM Corp. (2021). IBM SPSS statistics for windows (Version 28.0) [Computer software].

31.

Jaramillo-Baquerizo

Valcke

Vanderlinde

Aelterman

(2021). Exploring the consideration of university teachers’ basic psychological needs in the design of professional development initiatives. Journal of Higher Education Policy and Management, 43(3), 315–329.

32.

(2021). Does teacher engagement matter? Exploring relationship between teachers’ engagement in professional development and teaching practice. International Journal of TESOL Studies, 3(4), 42–60.

33.

Karsenty

Brodie

(2023). Researching “what went wrong” in professional development (PD) for mathematics teachers: What makes it so important, and so difficult? Journal of Mathematics Teacher Education, 26(5), 573–580.

34.

Kennedy

M. M.

(2016). How does professional development improve teaching? Review of Educational Research, 86(4), 945–980.

35.

Klassen

R. M.

Yerdelen

Durksen

T. L.

(2013). Measuring teacher engagement: Development of the engaged teachers scale (ETS). Frontline Learning Research, 1(2), 33–52.

36.

Kleickmann

Tröbst

Jonen

Vehmeyer

Möller

(2016). The effects of expert scaffolding in elementary science professional development on teachers’ beliefs and motivations, instructional practices, and student achievement. Journal of Educational Psychology, 108(1), 21.

37.

Korthagen

F. A.

Evelein

F. G.

(2016). Relations between student teachers’ basic needs fulfillment and their teaching behavior. Teaching and Teacher Education, 60, 234–244.

38.

Krille

(2020). Teachers’ participation in professional development: A systematic review. Springer. https://doi.org/10.1007/978-3-030-38844-7

39.

Kwakman

(2003). Factors affecting teachers’ participation in professional learning activities. Teaching and Teacher Education, 19(2), 149–170.

40.

Lin

Mastrokoukou

Longobardi

(2023). Social relationships and social media addiction among adolescents: Variable-centered and person-centered approaches. Computers in Human Behavior, 147, 107840.

41.

Lipowsky

Rzejak

(2015). Key features of effective professional development programmes for teachers. RicercAzione, 7(2), 27–51.

42.

Little

J. W.

(1982). Norms of collegiality and experimentation: Workplace conditions of school success. American Educational Research Journal, 19(3), 325–340.

43.

Liu

Huang

Song

(2025). Categories of teacher engagement in professional development and their attributions in the Chinese context: Implications for future PDs. Teaching and Teacher Education, 164, 105090.

44.

Mahmoudi

Özkan

(2015). Exploring experienced and novice teachers’ perceptions about professional development activities. Procedia-Social and Behavioral Sciences, 199, 57–64.

45.

McLarnon

M. J.

O’Neill

T. A.

(2018). Extensions of auxiliary variable approaches for the investigation of mediation, moderation, and conditional effects in mixture models. Organizational Research Methods, 21(4), 955–982.

46.

Meng

X. L.

Rubin

D. B.

(1992). Performing likelihood ratio tests with multiply-imputed data sets. Biometrika, 79(1), 103–111.

47.

Meyer

Kleinknecht

Richter

(2023). What makes online professional development effective? The effect of quality characteristics on teachers’ satisfaction and changes in their professional practices. Computers & Education, 200, 104805.

48.

Moè

Katz

(2022). Need satisfied teachers adopt a motivating style: The mediation of teacher enthusiasm. Learning and Individual Differences, 99, 102203.

49.

Moller

A. C.

Deci

E. L.

Elliot

A. J.

(2010). Person-level relatedness and the incremental value of relating. Personality and Social Psychology Bulletin, 36(6), 754–767.

50.

Mulaimović

Richter

Lazarides

Richter

(2025). Comparing quality and engagement in face-to-face and online teacher professional development. British Journal of Educational Technology, 56(1), 61–79.

51.

Muthén

L. K.

Muthén

B. O.

(1998–2017). Mplus user’s guide (8th ed.). Muthén & Muthén.

52.

Noonan

(2024). ‘Regard me’: A case study of learner engagement and the satisfaction of basic needs in continuing professional development. Professional Development in Education, 50(5), 894–908.

53.

Nylund

K. L.

Asparouhov

Muthén

B. O.

(2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal, 14(4), 535–569.

54.

Opfer

V. D.

Pedder

(2011). Conceptualizing teacher professional learning. Review of Educational Research, 81(3), 376–407.

55.

Pellas

(2025). Comparing quality and autonomous learning of teacher professional development programs in MOOCs and LMS. Education and Information Technologies, 30(7), 9005–9039.

56.

Picard

Kutsyuruba

(2017). Teachers’ engagement in professional development: A collective case study. Educational Policies and Current Practices, 2(2), 89–100.

57.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879.

58.

Powell

C. G.

Bodur

(2019). Teachers’ perceptions of an online professional development experience: Implications for a design and implementation framework. Teaching and Teacher Education, 77, 19–30.

59.

Power

Goodnough

(2019). Fostering teachers’ autonomous motivation during professional learning: A self-determination theory perspective. Teaching Education, 30(3), 278–298.

60.

Richter

Fütterer

Eisenkraft

Fischer

(2025). Profiling teachers’ motivation for professional development: A nationwide study. Journal of Teacher Education, 76(1), 90–103.

61.

Richter

(2024). Measuring the quality of teacher professional development: A large-scale validation study of an 18-item instrument for daily use. Studies in Educational Evaluation, 81, 101357.

62.

Ryan

R. M.

Deci

E. L.

(2017). Self-determination theory: Basic psychological needs in motivation, development, and wellness. Guilford Press.

63.

Ryan

R. M.

Jang

Wang

J. C. K.

Matos

Gordeeva

Kaplan

Behzadnia

Kantas

Ö.

Ferber

K. A.

Soenens

Vansteenkiste

. (2025). Variations in need supports in education as a function of cultural and economic factors: Perspectives from self-determination theory. Educational Psychology Review, 37(4), Article 118. https://doi.org/10.1007/s10648-025-10088-2

64.

Ryan

T. G.

(2019). Naturalistic Observation of Engagement and Disengagement within Professional Development in Education. International Online Journal of Education and Teaching, 6(1), 37–54.

65.

Schaufeli

W. B.

Salanova

González-Romá

Bakker

A. B.

(2002). The measurement of engagement and burnout: A two sample confirmatory factor analytic approach. Journal of Happiness Studies, 3(1), 71–92.

66.

Sims

Fletcher-Wood

(2021). Identifying the characteristics of effective teacher professional development: A critical review. School Effectiveness and School Improvement, 32(1), 47–63.

67.

Slemp

G. R.

Lee

M. A.

Mossman

L. H.

(2021). Interventions to support autonomy, competence, and relatedness needs in organizations: A systematic review with recommendations for research and practice. Journal of Occupational and Organizational Psychology, 94(2), 427–457.

68.

Smylie

M. A.

(2014). Teacher evaluation and the problem of professional development. Mid-Western Educational Researcher, 26(2), 7.

69.

Soine

K. M.

Lumpe

(2014). Measuring characteristics of teacher professional development. Teacher Development, 18(3), 303–333.

70.

Spurk

Hirschi

Wang

Valero

Kauffeld

(2020). Latent profile analysis: A review and “how to” guide of its application within vocational behavior research. Journal of Vocational Behavior, 120, 103445.

71.

Thurlings

den Brok

(2017). Learning outcomes of teacher professional development activities: A meta-study. Educational Review, 69(5), 554–576.

72.

Tschannen-Moran

Hoy

A. W.

(2007). The differential antecedents of self-efficacy beliefs of novice and experienced teachers. Teaching and Teacher Education, 23(6), 944–956.

73.

Vangrieken

Dochy

Raes

Kyndt

(2015). Teacher collaboration: A systematic review. Educational Research Review, 15, 17–40.

74.

Vansteenkiste

Ryan

R. M.

Soenens

(2020). Basic psychological need theory: Advancements, critical themes, and future directions. Motivation and Emotion, 44(1), 1–31.

75.

Vedder-Weiss

Roth

Mishaeli

(2025). Supporting teacher reflection and motivation through psychological needs satisfaction in collaborative reflection-based PD. The Journal of Experimental Education, 93(2), 320–339.

76.

Vivante

Vedder-Weiss

(2023). Examining science teachers’ engagement in professional development: A multimodal situated perspective. Journal of Research in Science Teaching, 60(7), 1401–1430.

77.

Voogt

Laferriere

Breuleux

Itow

R. C.

Hickey

D. T.

McKenney

(2015). Collaborative design as a form of professional development. Instructional Science, 43, 259–282.

78.

Walter

Briggs

(2012). What professional development makes the most difference to teachers? Oxford University Press.

79.

Yang

(2021). Understanding and enhancing Chinese TEFL teachers’ motivation for continuing professional development through the lens of self-determination theory. Frontiers in Psychology, 12, Article 768320.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.07 MB