Contextual Influences on Self-Assessed TPACK: A Comparison of Physics Undergraduates and In-Service Science Teachers

Abstract

The Technological Pedagogical Content Knowledge (TPACK) framework is widely used to conceptualize teacher knowledge as an interplay of content, pedagogy, and technology. Following recent research interests in examining TPACK as contextually situated knowledge, this study investigates how pre-service physics teachers (undergraduate students in a physics department) and in-service science teachers perceive the domains of TPACK and explores what these differences imply for university-based teacher education. A total of 48 pre-service physics undergraduates and 27 in-service teachers completed an adapted 21-item self-assessment questionnaire, which combined validated items with context-specific modifications. Data analysis included internal consistency reliability tests, independent samples t-tests, and correlation analysis. Results revealed that pre-service teachers reported higher self-assessed competencies, especially in integrative domains, although their knowledge structures appeared less coherent. In contrast, in-service teachers exhibited more coherent and integrated knowledge frameworks, possibly reflecting their accumulated professional experience, despite reporting lower self-confidence. These findings confirm the contextual and situated nature of TPACK, highlighting the divergence between perceived competence and structural coherence. The study contributes by proposing that university science education programs should not only promote theoretical understanding of TPACK but also deliberately embed technology-rich, practice-oriented experiences.

1. Introduction

The Technological Pedagogical Content Knowledge (TPACK) framework has significantly influenced research and practice in teacher education and professional development, generating a substantial body of research [1]. It extends Shulman’s [2] concept of pedagogical content knowledge (PCK) to include educational technology, characterizing teacher knowledge as a complex interaction among the three core components of content knowledge (CK), pedagogical knowledge (PK), and technological knowledge (TK) [3]. These domains intersect to generate distinct but interrelated forms of knowledge, including pedagogical content knowledge (PCK), technological content knowledge (TCK), and technological pedagogical knowledge (TPK), which together constitute the integrated TPACK construct. Subsequent refinements to the TPACK framework have integrated an explicit dimension concerning educational environments, namely contextual knowledge [4]. Its acquisition and application are linked to specific teaching and technological environments, suggesting that TPACK should be understood as a situated practice [5].

The TPACK framework has been widely adopted both as an analytical tool and as a pedagogical framework for guiding teacher education and professional development [6,7,8,9]. Investigating how pre-service teachers (undergraduate students) and in-service teachers develop and perceive their TPACK is therefore essential both for assessing professional readiness and for guiding the design of teacher education curricula in higher education.

For pre-service teachers, engagement with the TPACK framework offers opportunities to build confidence in applying technology-rich pedagogical approaches as they prepare for professional practice. Developing TPACK during teacher education programs is crucial for preparing them to enter schools as effective, digitally competent educators [10,11,12]. Exposure to TPACK in university courses has been shown to foster flexible, inquiry-oriented practices and to strengthen readiness to integrate digital tools meaningfully into subject teaching [13,14]. Engaging in TPACK-based training equips pre-service teachers with the ability to transform abstract content knowledge into accessible, technology-enhanced learning experiences for diverse learners. Research indicates that undergraduate students often report high levels of self-efficacy when engaging with integrative TPACK tasks, even though their knowledge structures remain relatively fragmented and disconnected [15,16].

For in-service teachers, cultivating TPACK supports continuous professional growth by enabling them to select and adapt digital tools critically to evolving curricular goals and classroom needs [17]. It fosters confidence in using technology not as an add-on, but as an integral part of pedagogy and content delivery. Research indicates that teachers who develop strong TPACK are better able to design innovative learning experiences, adapt technology to diverse classroom contexts, and enhance student engagement and achievement [18,19,20]. Evidence further suggests that their professional learning is most effective when training environments are tailored to subject matter and teaching contexts, rather than being generic [21]. Professional learning initiatives grounded in the TPACK framework have been shown to positively influence classroom innovation, teacher adaptability, and the integration of technology to support disciplinary learning goals [7,22]. Crucially, the development of in-service teachers’ TPACK is shaped by the situated demands of authentic classroom contexts, underscoring the gap that often exists between theoretical preparation and practical enactment.

Over the past two decades, the TPACK framework has generated a substantial body of research, much of which has relied on self-assessment questionnaires to investigate teachers’ perceived knowledge across its seven domains. One of the earliest validated instruments was developed by Schmidt et al. [10] for pre-service teachers and has since been adapted for use across multiple contexts and populations [23,24,25]. Later instruments sought to refine this approach by offering shorter scales with improved psychometric properties [15]. Reviews of the field have further emphasized the importance of contextualizing questionnaire items with subject-specific pedagogical practices [16,26,27]. In TPACK research, questionnaire-based approaches are frequently employed to facilitate systematic comparisons across teacher groups and educational settings [7,22].

Findings from questionnaire-based studies consistently point to systematic differences between pre-service and in-service teachers. Recent research shows that pre-service teachers often report higher levels of confidence, particularly in technological competencies and TPACK-related readiness [28], yet these self-assessments tend to reveal weaker internal consistency and less coherent knowledge structures [5,14,23,29]. In contrast, in-service teachers generally provide more modest self-ratings but display stronger internal consistency and more interconnected domain relationships, reflecting the influence of classroom experience [7,15,24,30,31]. Research in science education further illustrates this distinction, showing that pre-service teachers often conceptualize technology as an add-on, whereas in-service teachers tend to integrate it more meaningfully with pedagogy and content [32,33,34].

Cross-cultural investigations extend these findings by emphasizing that TPACK is contextually situated knowledge, shaped by teacher education programs, disciplinary practices, and local environments [4,29,35,36]. Such work highlights the importance of examining not only reported levels of TPACK but also the reliability, coherence, and contextual variability of teachers’ self-assessments. By addressing both the measurement challenges and the contextual nature of TPACK, recent studies underscore the need for instruments and analyses that account for differences in professional stage, discipline, and teaching context [4,5].

Although several comparative studies have examined differences between pre-service and in-service teachers’ TPACK perceptions, their findings show notable variability. For example, some research has shown higher TPACK self-confidence among in-service teachers [37], whereas others report stronger technological and integrative self-perceptions among pre-service teachers [28]. These discrepancies reflect the strong contextuality of TPACK: results vary across national settings, institutional programs, disciplinary backgrounds, and the specific technological tools emphasized during training. Additionally, only a limited number of studies have focused on subject-specific comparisons within science domains, and empirical evidence from the context of Greek physics education remains scarce. This indicates a clear need for studies that examine TPACK perceptions within a common disciplinary framework while also considering the professional stage.

Despite this extensive body of work, relatively few studies directly compare how pre-service and in-service teachers perceive and structure their TPACK competencies, especially within subject-specific contexts such as physics education. Understanding these differences may support the refinement of university science education programs and the design of ongoing professional development, ensuring that support is aligned with teachers’ evolving needs across career stages. In this context, the present exploratory study examines how pre-service (undergraduate Physics students) and in-service science teachers perceive their TPACK competencies, investigates differences in self-assessed confidence across domains, and explores how contextual experience may be associated with the internal structure and interrelations of TPACK. Specifically, the research questions are:

RQ1. 

To what extent do pre-service and in-service science teachers differ in their self-assessed competencies across the TPACK domains?

RQ2. 

How are the different domains of the TPACK framework interrelated within each teacher group, and what similarities or differences emerge in the patterns of these relationships?

2. Materials and Methods

2.1. Participants and Context

This study employed a cross-sectional comparative survey design. The participants were 48 pre-service teachers, students in the Department of Physics at Aristotle University of Thessaloniki, and 27 in-service science teachers (Physics, Chemistry, Biology, Geology, and Mathematics). In the Greek educational system, all science specialties are formally qualified to teach Physics at the secondary level, minimizing potential disciplinary differences relevant to TPACK. The pre-service teachers attended an introductory semester course in Didactics of Physics during the 2023–2024 academic year. The course consisted of three hours of weekly lectures on topics such as learning theories, teaching models, aims and forms of evaluation, and didactic transposition. The 27 in-service science teachers attended the Interdepartmental Master’s Program “Educational Sciences: Teacher Education in Innovative Approaches to Teaching and Learning” in the specialization of Science Education. All participants had completed a general course on the educational use of ICT in the first semester. The TPACK questionnaire was administered at the beginning of the second semester, prior to their engagement with subject-specific didactics, to capture their initial knowledge and perceptions regarding TPACK. Participants from both groups completed the digital questionnaire using Google Forms (Google LLC, Mountain View, CA, USA) in person, at the end of their respective courses. The completion time was approximately 45 min for all participants. Before beginning the questionnaire, participants were informed about the aims of the study, the voluntary nature of participation, and the anonymity of their responses. By submitting the questionnaire, participants indicated their informed consent. No missing data were recorded, as the digital format required a response for each item.

2.2. The Research Instrument

A questionnaire was developed to measure teachers’ self-assessed knowledge across the seven domains of the TPACK framework [1]: CK, PK, TK, PCK, TPK, TCK, and the integrated TPACK. The instrument comprised 21 items, with three items assigned to each domain. Responses were recorded on a five-point Likert scale (1 = “Strongly Disagree,” 5 = “Strongly Agree”), allowing participants to indicate the extent of their agreement with each statement.

The questionnaire items were primarily adapted from widely used and validated TPACK instruments [10,13,14,15,23,24,25,36,38,39]. Minor modifications were made to contextualize the items for physics education, while some new items were developed by the authors to align with the study’s context. To minimize uncertainty and enhance clarity, some items were enriched with specific examples, following concerns raised by Willermark [16] regarding the use of overly general statements in TPACK instruments. For instance, items referring to technologies included concrete examples such as simulations and multimedia, while items mentioning “teaching models” specified approaches like discovery, constructivism, and inquiry. Moreover, within the TPACK domain, two items addressed the design of worksheets and assessment exercises that explicitly integrate technology, pedagogy, and content knowledge, providing a concrete example of how CK, TK, and PK interact in practice. The complete set of questionnaire items, organized by TPACK domain with references to their sources, is presented in Appendix A.

2.3. Data Analysis

Data analysis was conducted using PS IMAGO PRO 10 (SPSS version 29; IBM Corp., Armonk, NY, USA). The consistency of the adapted TPACK questionnaire across teacher groups (pre-service and in-service) was assessed. Group differences and relational patterns within the TPACK framework were examined using three analytical strategies.

The internal consistency reliability of the seven TPACK domain scales was evaluated separately for the pre-service and in-service teacher groups using Cronbach’s alpha (α). The conventional threshold of α ≥ 0.70 [40] served as the benchmark for acceptable reliability. Group differences in self-assessed competencies were examined by conducting a series of independent samples t-tests, treating each TPACK item as a dependent variable and group membership (pre-service vs. in-service) as the independent variable. Although internal consistency was assessed at the domain level, group comparisons were conducted at the item level. This approach was adopted because each TPACK domain consisted of only three items, and therefore, domain-level comparisons would have relied on aggregates with limited internal consistency, particularly for the pre-service group. Conducting item-level t-tests allowed for a more fine-grained and statistically defensible examination of group differences, while also avoiding potential distortions introduced by averaging across weak subscales. For each analysis, the mean difference, t-value, degrees of freedom, corresponding p-value, and Cohen’s d were reported, accompanied by 95% confidence intervals. In addition, 95% confidence intervals were also computed for the correlation coefficients presented in the group-wise correlation matrices.

Separate Pearson correlation analyses were conducted for each group to examine interrelationships among the seven TPACK domains, producing a correlation matrix of domain scores for both pre-service and in-service teachers.

3. Results

3.1. Internal Consistency of TPACK Domains Across Pre-Service and In-Service Teachers

To examine the reliability of the adapted physics TPACK questionnaire across groups, Cronbach’s alpha coefficients were computed for each domain. As shown in

Table 1. Internal consistency reliability (Cronbach’s Alpha) for each TPACK domain in pre- and in- service teachers.

TPACK Domain	Pre-Service Teachers (N = 48)	In-Service Teachers (N = 27)
CK	0.60	0.79
PK	0.61	0.77
TK	0.59	0.67
PCK	0.60	0.73
TPK	0.85	0.89
TCK	0.75	0.95
TPACK	0.77	0.90

Note: Nunnally’s [40] threshold of α ≥ 0.70 is used as a benchmark for acceptable reliability.

, the results reveal notable differences in internal consistency between pre-service and in-service teachers, which may be associated with differences in how the groups interpret the TPACK items.

For pre-service teachers, the reliability coefficients for the core domains of CK (α = 0.60), PK (α = 0.61), TK (α = 0.59), and PCK (α = 0.60) fell below the conventional threshold of 0.70 [40]. These values suggest that the items within these domains exhibited limited internal consistency for this group, or that the scales did not consistently capture a unified underlying construct, pointing to a more fragmented conceptualization of these knowledge bases among pre-service teachers. However, for in-service teachers, the questionnaire demonstrated good to excellent reliability across all domains (α = 0.67–0.95). This indicates that, while the instrument itself possesses adequate internal consistency, pre-service teachers may hold less coherent or less well-developed representations of these knowledge domains compared with their in-service counterparts. In contrast, for pre-service teachers, the domains involving the integration of technology (TPK, TCK) as well as the overall framework (TPACK) demonstrated good to excellent reliability (α = 0.75–0.85).

For in-service teachers, except for TK (α = 0.67), which was marginal but still within acceptable limits for exploratory studies, all domains exceeded the 0.70 threshold. These results reflect more consistent response patterns among in-service teachers, which may be associated with more coherent patterns in their self-reported TPACK.

The low reliability of the TK domain for pre-service teachers (α = 0.59) warrants closer examination. For pre-service teachers, self-assessed skills in “using tools such as email, Word, and the internet” and “using simulations with ease” did not show strong inter-item associations. One possible interpretation is that everyday digital tools and subject-specific simulations may function as distinct types of technological engagement, with the latter often involving more complex conceptual and pedagogical demands. The low alpha thus serves as an indicator that pre-service teachers’ technological knowledge remains emergent and highly context dependent and may not yet form a fully coherent construct. For in-service teachers, even with a marginally acceptable alpha, these diverse technological skills appear to be more integrated into a unified perception of technological competence.

3.2. Differences in TPACK Self-Perceptions Between Pre- and In-Service Teachers

To address the second research question, “To what extent do pre-service and in-service physics teachers differ in their self-assessed competencies across the TPACK domains?”, independent-samples t-tests were conducted. The analyses revealed a consistent pattern: pre-service physics teachers reported significantly higher self-assessed competencies across nearly all TPACK domains compared with their in-service counterparts. As shown in

Table 2. Comparison of Pre-Service and In-Service Teachers’ Self-Assessed TPACK Competencies.

TPACK Domains	Item	Pre-Service Mean (SD)	In-Service Mean (SD)	Mean Difference	t Value	df	p Value	Cohen’s d (95% CI)
CK	a. I have sufficient knowledge about my teaching subject	4.08 (0.61)	3.93 (0.68)	0.16	1.03	73	0.307	0.248 [−0.225–0.721]
	b. I am able to gain a deeper understanding of the content knowledge of my teaching subject on my own	4.40 (0.61)	3.96 (0.71)	0.43	2.67	48	0.010	0.670 [0.186–1.154]
	c. I am confident in teaching the subject matter	4.04 (0.90)	3.67 (0.78)	0.37	1.81	73	0.074	0.436 [−0.041–0.913]
PK	a. I can use a wide range of teaching approaches in a classroom setting (discovery, constructivism, inquiry)	3.96 (0.82)	3.48 (0.70)	0.48	2.53	73	0.013	0.610 [0.128–1.092]
	b. I can adapt student assessment based on the teaching approach	4.06 (0.70)	3.41 (0.69)	0.65	3.91	73	<0.001	0.942 [0.446–1.438]
	c. I am familiar with common student understandings and misconceptions	4.06 (0.81)	3.33 (1.00)	0.73	3.44	73	<0.001	0.827 [0.337–1.317]
TK	a. I am comfortable using tools such as email, Word, and the internet	4.92 (0.35)	4.74 (0.45)	0.18	1.77	44	0.042	0.456 [−0.021–0.933]
	b. I can use simulations with ease	4.60 (0.76)	3.74 (0.94)	0.86	4.31	73	<0.001	0.833 [0.343–1.323]
	c. I can comfortably use the internet to search for images/experiments/simulations	4.92 (0.35)	4.56 (0.51)	0.36	3.30	40	0.002	0.879 [0.386–1.372]
PCK	a. I know how to select effective teaching approaches to guide student thinking and learning in my teaching subject	3.92 (0.58)	3.37 (0.63)	0.55	3.81	73	<0.001	0.916 [0.422–1.410]
	b. I can formulate goals (or adjust goals) according to the teaching approach I use	4.02 (0.67)	3.59 (0.57)	0.43	2.80	73	0.007	0.674 [0.190–1.158]
	c. I know how to develop exercises with which students can consolidate their knowledge of my teaching subject	4.25 (0.60)	3.63 (0.74)	0.62	3.94	73	<0.001	0.947 [0.451–1.443]
TPK	a. I can choose technologies (e.g., multimedia, simulations) that enhance the teaching approaches for a lesson	4.40 (0.64)	3.81 (0.83)	0.58	3.37	73	<0.001	0.810 [0.321–1.299]
	b. I can choose technologies (e.g., multimedia, simulations) that enhance student learning for a lesson	4.40 (0.71)	3.78 (0.75)	0.62	3.55	73	<0.001	0.855 [0.364–1.346]
	c. I know how to develop assessment exercises using new technologies	4.13 (0.73)	3.63 (0.93)	0.49	2.39	44	0.021	0.614 [0.132–1.096]
TCK	a. I can use appropriate technologies (e.g., multimedia resources, simulation) to represent the content of my teaching subject	4.44 (0.68)	3.89 (0.85)	0.55	3.06	73	0.003	0.737 [0.251–1.223]
	b. I can choose the appropriate technological tool depending on the content to be taught	4.17 (0.81)	3.81 (0.83)	0.36	1.79	73	0.078	0.431 [−0.046–0.908]
	c. I know how to develop assessment exercises using different technological tools	3.92 (0.85)	3.52 (0.98)	0.40	1.85	73	0.068	0.445 [−0.032–0.922]
TPACK	a. I can choose the appropriate technology depending on the content to be taught and the teaching approach	4.06 (0.67)	3.30 (0.61)	0.77	4.93	73	<0.001	1.187 [0.678–1.696]
	b. I can develop worksheets taking into account the technology, the teaching approach, and the content to be taught	4.17 (0.69)	3.44 (0.70)	0.72	4.32	73	<0.001	1.038 [0.537–1.539]
	c. I can develop assessment exercises taking into account the technology, the teaching approach, and the content to be taught	4.10 (0.78)	3.41 (0.75)	0.70	3.77	73	<0.001	0.908 [0.414–1.402]

, statistically significant differences with medium-to-very-large effect sizes were observed for most items, particularly those requiring integration of multiple TPACK components.

Within the core domains (CK, PK, TK), differences in CK were less pronounced. Both groups reported high confidence in their subject-matter knowledge (item CKa). However, a significant difference appeared in CKb, with pre-service teachers reporting higher confidence in independently deepening their understanding of content (p = 0.010, d = 0.67). This pattern may suggest that recent teacher preparation places greater emphasis on self-regulated content learning. A marginal difference also emerged for CKc (p = 0.074), showing that confidence in teaching the subject matter was relatively similar between the two groups.

In the PK domain, pre-service teachers demonstrated notably higher confidence overall. They reported higher confidence in using a wide range of teaching approaches (PKa: p = 0.013, d = 0.61). They also showed higher confidence in adapting assessment based on the teaching approach (PKb: p < 0.001, d = 0.94). Finally, they reported greater confidence in understanding student misconceptions (PKc: p < 0.001, d = 0.83). These large effects show substantial group differences in self-assessed pedagogical preparedness.

Within TK, both groups reported high confidence in using basic tools (TKa), with pre-service teachers showing a marginally higher mean. The largest differences appeared for applied technological skills, particularly using simulations (TKb: p < 0.001, d = 0.83) and searching for digital resources online (TKc: p = 0.002, d = 0.88). This pattern may be consistent with the emphasis on technological competencies commonly found in con-temporary teacher education programs.

The PCK domain revealed significant advantages for pre-service teachers across all items: selecting effective teaching approaches (PCKa: p < 0.001, d = 0.92), formulating instructional goals (PCKb: p = 0.007, d = 0.67), and developing consolidation exercises (PCKc: p < 0.001, d = 0.95). These large effects show notable differences in how the two groups self-assessed their pedagogical preparedness.

Similarly, in the TPK domain, pre-service teachers reported higher confidence in choosing technologies to enhance teaching (TPKa: p < 0.001, d = 0.81) and student learning (TPKb: p < 0.001, d = 0.86), along with developing technology-based assessments (TPKc: p = 0.021, d = 0.61).

In the TCK domain, a significant difference appeared in the ability to use technology to represent content (TCKa: p = 0.003, d = 0.74). Differences in choosing technological tools (TCKb) and developing technological assessments (TCKc) did not reach statistical significance. However, both items showed medium effect sizes.

The largest effect sizes appeared in the integrated TPACK domain. Pre-service teachers reported substantially higher confidence in choosing appropriate technology based on content and pedagogy (TPACKa: p < 0.001, d = 1.19). They also expressed higher confidence in developing integrated worksheets (TPACKb: p < 0.001, d = 1.04). Finally, they showed greater confidence in creating comprehensive assessment exercises (TPACKc: p < 0.001, d = 0.91). These large effects constitute the most pronounced differences between groups.

The pattern of results consistently shows that pre-service physics teachers perceive themselves as more competent across multiple TPACK domains compared to their in-service counterparts. The most substantial differences appear in areas requiring the integration of technology with pedagogy and content knowledge.

3.3. Interrelations of TPACK’s Domains and Comparison Between Pre- and In-Service Teachers

Pearson correlation analyses were conducted separately for pre-service and in-service teacher groups to address the third research question: “How are the different domains of the TPACK framework interrelated within each teacher group, and what similarities or differences emerge in the patterns of these relationships?”. Composite scores were calculated for each TPACK domain, although the relatively low reliability indices in the pre-service group (Cronbach’s α < 0.70 for CK, PK, TK, and PCK) warrant caution in interpreting these domains.

For pre-service teachers, as shown in

Table 3. Pearson Correlation Coefficients for the TPACK Domains for Pre-Service Teachers (N = 48).

	1. CK	2. PK	3. TK	4. PCK	5. TPK	6. TCK
1. CK	-
2. PK	0.362 *	-
3. TK	0.193	0.170	-
4. PCK	0.434 **	0.461 **	0.067	-
5. TPK	0.303 *	0.401 **	0.027	0.327 *	-
6. TCK	0.345 *	0.159	0.087	0.366 *	0.602 **	-
7. TPACK	0.399 **	0.347 *	0.123	0.616 **	0.681 **	0.567 **

* p < 0.05, ** p < 0.01. 95% confidence intervals (CIs) for correlations (Fisher’s z-transformed CIs, back-transformed to r): CK-PK [0.087, 0.586]; CK-TK [–0.096, 0.452]; CK-PCK [0.171, 0.639]; CK-TPK [0.021, 0.541]; CK-TCK [0.067, 0.573]; CK-TPACK [0.130, 0.614]. PK-TK [–0.120, 0.433]; PK-PCK [0.204, 0.659]; PK-TPK [0.132, 0.615]; PK-TCK [–0.131, 0.424]; PK-TPACK [0.070, 0.574]. TK-PCK [–0.221, 0.345]; TK-TPK [–0.259, 0.309]; TK-TCK [–0.202, 0.362]; TK-TPACK [–0.167, 0.393]. PCK-TPK [0.047, 0.559]; PCK-TCK [0.091, 0.589]; PCK-TPACK [0.402, 0.766]. TPK-TCK [0.383, 0.757]; TPK-TPACK [0.492, 0.809]. TCK-TPACK [0.337, 0.733].

, the correlation matrix revealed significant positive correlations among most TPACK domains. However, the role of TK was notably distinct, showing weaker correlations with the other domains.

CK was significantly positively correlated with all other domains except TK. Specifically, CK showed significant relationships with PK (r = 0.362, p = 0.011), PCK (r = 0.434, p = 0.002), TPK (r = 0.303, p = 0.036), TCK (r = 0.345, p = 0.016), and TPACK (r = 0.399, p = 0.005). PK was significantly correlated with CK, PCK (r = 0.461, p = 0.001), TPK (r = 0.401, p = 0.005), and TPACK (r = 0.347, p = 0.016). A critical finding was that TK was not significantly correlated with any other domain (all p > 0.05), a pattern that may suggest that TK operates somewhat independently within this group’s framework. The integrative domains of PCK, TPK, and TCK showed relatively strong interrelations. PCK was significantly associated with TPK (r = 0.327, p = 0.023) and TCK (r = 0.366, p = 0.011). Most notably, a strong positive correlation was observed between TPK and TCK (r = 0.602, p < 0.001). Finally, the synthesized TPACK domain showed strong, significant correlations with all core domains (CK, PK, PCK) and integrative domains (TPK, TCK), with the strongest associations observed with TPK (r = 0.681, p < 0.001) and PCK (r = 0.616, p < 0.001).

The correlation pattern for in-service teachers (

Table 4. Pearson Correlation Coefficients for the TPACK Domains for In-Service Teachers (N = 27).

	1. CK	2. PK	3. TK	4. PCK	5. TPK	6. TCK
1. CK	-
2. PK	0.196	-
3. TK	0.114	0.293	-
4. PCK	0.175	0.638 **	0.430 *	-
5. TPK	0.145	0.646 **	0.606 **	0.756 **	-
6. TCK	0.313	0.565 **	0.671 **	0.631 **	0.839 **	-
7. TPACK	0.200	0.574 **	0.591 **	0.750 **	0.758 **	0.795 **

* p < 0.05, ** p < 0.01. 95% confidence intervals (CIs) for correlations (Fisher’s z-transformed CIs, back-transformed to r): CK-PK [–0.199, 0.536]; CK-TK [–0.278, 0.474]; CK-PCK [–0.220, 0.520]; CK-TPK [–0.249, 0.498]; CK-TCK [–0.076, 0.619]; CK-TPACK [–0.195, 0.539]. PK-TK [–0.098, 0.606]; PK-PCK [0.341, 0.819]; PK-TPK [0.353, 0.824]; PK-TCK [0.236, 0.778]; PK-TPACK [0.248, 0.783]. TK-PCK [0.060, 0.696]; TK-TPK [0.294, 0.801]; TK-TCK [0.391, 0.837]; TK-TPACK [0.272, 0.793]. PCK-TPK [0.528, 0.882]; PCK-TCK [0.330, 0.815]; PCK-TPACK [0.517, 0.879]. TPK-TCK [0.674, 0.924]; TPK-TPACK [0.531, 0.884]. TCK-TPACK [0.595, 0.902].

) revealed a different configuration, with stronger and more widespread associations among the TPACK domains. In this group, TK showed significant correlations with multiple domains, possibly suggesting a more interconnected structure of self-reported knowledge.

In contrast to the pre-service group, CK was not significantly correlated with any other TPACK domain (all p > 0.05). In this dataset, subject-matter knowledge does not appear to show statistical associations with pedagogical, technological, or integrative knowledge for experienced teachers. In the in-service group, PK functioned as a central hub, showing strong and significant associations with the integrated domains: PCK (r = 0.638, p < 0.001), TPK (r = 0.646, p < 0.001), TCK (r = 0.565, p = 0.002), and TPACK (r = 0.574, p = 0.002). Unlike the pre-service group, TK was significantly and positively correlated with all integrative and blended domains: PCK (r = 0.430, p = 0.025), TPK (r = 0.606, p = 0.001), TCK (r = 0.671, p < 0.001), and TPACK (r = 0.591, p = 0.001). The associations among the integrative domains (PCK, TPK, TCK) were particularly strong. The strongest correlations in the entire matrix were observed between TPK and TCK (r = 0.839, p < 0.001), and between PCK and TPK (r = 0.756, p < 0.001). The TPACK domain was strongly correlated with all other domains except CK, with the strongest links to PCK (r = 0.750, p < 0.001), TPK (r = 0.758, p < 0.001), and TCK (r = 0.795, p < 0.001).

In both groups, the integrative domains (PCK, TPK, TCK) and the synthesized TPACK domain were strongly interrelated, with the strongest associations consistently observed between TPK and TCK. The comparison of the two correlation structures, however, reveals noteworthy differences. Among pre-service teachers, CK and PK were the most central domains, whereas TK showed weaker associations with the rest of the framework, indicating that TK may not yet be fully connected to their developing pedagogical and content-related understandings. For in-service teachers, PK, TK, and the integrative domains formed a more tightly connected network, with CK functioning as an independent knowledge base. Overall, the two groups displayed distinct patterns of correlations. Pre-service teachers showed a less integrated configuration of TPACK-related domains. For in-service teachers, the domains exhibit stronger and more widespread associations, pointing to a more interconnected pattern of self-reported TPACK-related knowledge.

4. Discussion

The analysis of internal consistency reliability revealed marked differences between pre-service and in-service teachers, underscoring the contextual nature of the TPACK framework. For pre-service teachers, reliability coefficients concerning the core domains (CK, PK, TK, and PCK) fell below the conventional threshold. This pattern is consistent with the view that these domains may not yet represent coherent or integrated constructs for novices. This interpretation aligns with prior scholarship characterizing TPACK as a situated and evolving form of teacher knowledge shaped by educational and professional contexts [4,5,34,41]. In contrast, in-service teachers exhibited good to excellent reliability across most domains, which may reflect more stable or more consistently interpreted knowledge structures, potentially related to their broader professional experience. The lower alpha values observed for TK in the pre-service group point to a possible distinction between general digital skills and subject-specific technological applications, which may not yet be internalized as a unified competence. These results may reflect the importance of contextual factors when interpreting TPACK assessment results, as the same instrument may capture partially developed or more cohesive knowledge structures depending on participants’ professional stage.

Results addressing the first research question, “To what extent do pre-service and in-service physics teachers differ in their self-assessed competencies across the TPACK domains?”, showed significant differences across most domains, with pre-service teachers consistently reporting higher confidence. These differences were especially pronounced in domains involving the integration of technology, pedagogy, and content (TK, TPK, and TPACK). By contrast, fewer differences were found to be associated with certain TCK items, suggesting more comparable perceptions in content-related technological applications. The largest effect sizes were observed in integrative items such as TPACKa (d = 1.19) and TPACKb (d = 1.04), followed by PCKc (d = 0.95). These findings appear to indicate that pre-service teachers report comparatively higher confidence in tasks involving the coordination of multiple knowledge domains. This pattern is consistent with recent studies showing that pre-service teachers often report strong technological competencies and substantial confidence in TPACK-related domains [28]. Studies also mention that teacher education programs increasingly emphasize digital technologies and inquiry-based methods, which may enhance pre-service teachers’ confidence in integrative TPACK competencies [13,14]. Research also suggests that pre-service teachers may overestimate these competencies, whereas in-service teachers tend to report more calibrated self-assessments aligned with classroom demands [33]. At the same time, differences in CK and some TCK items were small or nonsignificant, possibly suggesting that both groups feel relatively secure in content-related areas. Collectively, the evidence points to differing confidence profiles across professional stages. Beyond group differences, the findings also emphasize the technology’s role in shaping pre-service teachers’ emerging professional knowledge. Their higher self-assessed competencies in TK, TPK, TCK, and TPACK, along with the strong associations between technological and integrative domains, suggest that technology-rich coursework may play a substantial role in fostering confidence in designing technology-supported lessons. Strengthening technological components within pre-service teacher education may therefore contribute to preparing future teachers who are pedagogically ready to integrate technology meaningfully in their classrooms.

According to the second research question, “How are the different domains of the TPACK framework interrelated within each teacher group, and what similarities or differences emerge in the patterns of these relationships?”, the correlation analysis revealed two distinct patterns of associations. For pre-service teachers, TK appeared independent from pedagogy and content. This pattern is consistent with the idea that novice teachers often view technology as an external add, rather than an integral component of instruction [32]. CK and PK functioned as central domains, while the integrative domains (TPK, TCK, PCK) showed emerging, but not yet strongly unified, patterns of association. In contrast, the in-service teachers displayed a highly interconnected TPACK structure in which TK was strongly related to PCK, TPK, TCK, and TPACK, which may reflect that technology had become embedded in their pedagogical and content-related reasoning. CK, however, was not significantly correlated with other domains in the in-service group, which may indicate that content knowledge operates as a relatively independent knowledge base for experienced teachers. A consistent finding across both groups was the strong correlations between TPK and TCK, supporting the view that understanding how technology reshapes pedagogy is closely tied to understanding how it represents content [3,7]. Recent studies confirm these trends, showing that in-service teachers’ TPACK domains form highly interconnected structures grounded in practice, while pre-service teachers’ structures remain more fragmented and show distinct gaps in TK-TPK integration [34,41]. However, the markedly stronger associations among in-service teachers may reflect the role of professional experience in developing a more coherent knowledge framework.

Taken together, these findings reinforce the view that TPACK must be understood as contextually situated [4,5,35]. In university pre-service programs, knowledge is often introduced through relatively discrete domains [30], which may help explain the isolation of TK in the pre-service group. In contrast, professional experience may function as an important integrating factor, supporting the development of more interconnected knowledge structures among in-service teachers. These insights underscore the need for teacher education programs to foster integration through technology-rich field experiences, and for professional development to continue supporting the consolidation of more sophisticated TPACK structures.

It is worth mentioning the divergence between perceived competence and structural coherence across the two groups. Pre-service teachers consistently reported higher self-assessed competencies across most TPACK domains; however, the reliability and correlation analysis revealed that their knowledge structures were less integrated. This pattern suggests that novice teachers may demonstrate a form of theoretical confidence, reflecting early stages of professional formation, without yet possessing an integrated understanding of how the domains of content, pedagogy, and technology interact in practice. In contrast, in-service teachers, despite reporting somewhat lower mean self-assessments, demonstrated stronger internal consistency and more interconnected knowledge structures. Perhaps this is related to the broader professional experience they have acquired over time. This discrepancy suggests that TPACK may function as situated, context-dependent construct [4,5,35], with different levels of coherence emerging at different stages of teachers’ professional trajectories. Our findings are consistent with recent research indicating that TPACK tends to evolve in interaction with both university-based preparation and practical teaching experience, which may support the development of more integrated knowledge structures over time [4,5,42].

5. Conclusions

This study confirms the context-dependent nature of TPACK, showing that while pre-service teachers report higher confidence in integrative domains, their knowledge structures remain inconsistent. In contrast, in-service teachers exhibit more coherent and interconnected knowledge structures, despite reporting lower self-assessments. These patterns may be linked to differences in professional experience between the two groups, although the cross-sectional design does not allow for causal inference. However, this divergence highlights the need to rethink how teacher education promotes the development of TPACK.

For university programs, the findings suggest that fostering theoretical understanding alone is not sufficient. Teacher education should provide structured, technology-rich, and practice-oriented experiences that enable pre-service teachers (university students) to integrate content, pedagogy, and technology in authentic contexts. Embedding modeling tasks, microteaching sessions, and field-based projects that explicitly require coordination of multiple knowledge domains can help bridge the gap between abstract knowledge and applied competence. Such a design would support pre-service teachers in developing not only confidence but also coherent and contextually grounded TPACK structures. At the same time, professional development for in-service teachers should continue to strengthen and refine their integrated frameworks, enabling them to adapt TPACK flexibly to diverse and evolving teaching contexts.

Overall, the study reinforces that TPACK should not be treated as a static framework but must be cultivated as situated knowledge. For universities, this requires designing teacher education programs that deliberately connect theory with practice, ensuring that future educators acquire both the confidence and the conceptual coherence needed to integrate TPACK effectively in real classrooms.

6. Limitations

The present study assesses teachers’ TPACK exclusively using self-reported questionnaires. While such instruments are appropriate for systematic comparisons across groups and provide insight into teachers’ perceived competencies, they capture perceptions rather than actual classroom practices or pedagogical enactments. This reliance on self-assessment introduces potential bias, particularly over- or underestimation of integrative competencies, and may not fully capture the complexity of teachers’ knowledge in authentic teaching contexts.

Additionally, given the sample size of the in-service group, formal testing of cross-group measurement comparability was not feasible

7. Suggestions for Future Research

Future research should complement self-reported measures with performance-based assessments, classroom observations, or other forms of evidence that capture teachers’ knowledge in action. Comparative studies across disciplinary and cultural contexts could further illuminate how educational environments shape teachers’ TPACK. Continued refinement of TPACK instruments, particularly with contextualized items, is essential for advancing the validity and reliability of questionnaire-based approaches.

Future research could further explore how practice-based components of teacher education, such as explicit engagement with the TPACK framework, structured workshops on technology integration, guided micro-teaching using digital tools, and opportunities for reflective lesson design, shape the development of TPACK. Investigating how such targeted interventions contribute to more coherent TPACK structures among pre-service teachers would help bridge the gap between theoretical preparation and classroom practice and could inform more effective models of teacher professional learning.

A methodological contribution of the present study concerns the decision to compare the two teacher groups at the item level rather than at the domain level. This approach may be useful for future TPACK studies employing short subscales, especially in contexts where domain-level reliability varies across participant group.