Interest in incorporating computational thinking (CT) into K-12 curricula has increased in recent years (Wing, 2006). CT is a problem-solving skill that draws upon concepts found in computer science. Kjällander et al. (2021) posited that “computational thinking as described by Jeannette Wing… is a fundamental skill for everyone, not just for computer scientists. Reading, writing, and arithmetic should add computational thinking to every child’s analytical ability” (p. 1). Researchers have argued that, rather than teaching CT in isolation, this skill should be integrated into existing content area instruction, particularly in science, technology, engineering, and math (STEM), as an integrated approach offers benefits to students, including more equitable access, opportunities for sensemaking, and engaging with interdisciplinary and complex problems (Jocius et al., 2021).
While CT is often associated with secondary or postsecondary coursework, studies have indicated that younger learners would benefit from the inclusion of this skill in learning experiences (Cabrera, 2021; Kim et al., 2024; Rich et al., 2017). Kjällander et al. (2021) found that digital competencies such as programming and CT are essential skills for early elementary students so that they are better prepared to compete on a global stage. However, providing elementary-aged students with access to CT-infused instruction requires that teachers have access to professional development (PD) around CT and are confident and able to deliver CT-infused lessons.
This study was part of a larger research project designed to empower elementary teachers to integrate CT practices into elementary science instruction in ways that are culturally relevant for students (Mak et al., 2025). Teacher participants are introduced to a framework outlining CT practices such as using data, programming, computational simulations, and systems thinking (Cabrera, 2021) and provided support for incorporating these practices into elementary science lessons, including two summer workshops (PD1 and PD2), monthly PD meetings (community gatherings), classroom observations (lesson observations), and access to materials and equipment.
While existing research provides some guidance about how to prepare preservice and in-service teachers to integrate CT into content area instruction (Cabrera, 2021; Gilbert et al., 2016), additional research is needed to examine teacher self-efficacy in integrating CT (Avci & Deniz, 2022; Cadieux et al., 2020; Jaipal-Jamani & Angeli, 2017; Jocius et al., 2021; Şahin et al., 2024). The purpose of this study, which was part of a larger designed-based research project, was to examine teachers’ self-efficacy in implementing CT-infused science lessons as they participated in a 2-year PD program. Specifically, this study considered how participating teachers’ self-efficacy in implementing CT-infused science lessons changes as they engaged in activities with the potential to act as sources of self-efficacy (Bandura, 1977) to answer the following research questions:
- How and to what extent does teachers’ self-efficacy in implementing CT-infused science lessons change after engaging in a PD program focused on CT practices in science?
- How is teachers’ self-efficacy in implementing CT-infused science lessons impacted by teaching CT-infused science lessons?
- How does participation in academic year professional development (community gatherings) impact teachers’ self-efficacy implementing CT-infused science lessons?
Literature Review
Theory of Self-Efficacy
Bandura’s (1977, 1986) theory of self-efficacy was used as an analytical lens to determine how teachers’ practices of CT-infused science evolved and changed over approximately 1.5 years. Self-efficacy is defined as “beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments” (Bandura, 1977, p.3). According to Bandura, self-efficacy influences behavior by the challenges and goals an individual sets for themselves and the amount of effort they put into attaining that goal while overcoming obstacles and difficulties throughout the process. For teachers, self-efficacy can be understood as the beliefs that teachers hold to make the best decisions for instruction, student engagement, and classroom management (Pfitzner-Eden, 2016).
In addition, individuals form self-efficacy beliefs by interpreting capabilities from four sources: verbal persuasion, the judgments teachers make based on the verbal encouragement they receive; vicarious experiences, the failures or successes of other teachers who serve as examples; mastery experiences, the perceptions teachers make based on past experiences; and physiological and affective states, the emotional and physiological affect teachers experience as they anticipate instruction (Tschannen-Moran & McMaster, 2009).
According to Burger (2024) mastery experiences have the strongest effect on teachers’ self-efficacy beliefs, as they tend to be the best indicators of capabilities based on authentic experiences. Mastery experiences that are generated in the classroom provide teachers with opportunities to grow and change based on the evidence as they learn and apply new skills. Although mastery experiences may boost teachers’ self-efficacy, teachers require opportunities to engage in instructional-related success or practical training. Burger further posited, “Vicarious experiences from observations of a model teacher, and verbal persuasion via performance feedback from mentors may also emerge as strong sources of efficacy beliefs” (p. 108).
Verbal persuasion through coaching and mentoring from more experienced teachers provides opportunities for teachers to grow, and vicarious experiences provide teachers with opportunities to observe classrooms of more experienced teachers so that learning through modeling may occur. Last, physiological and affective states can negatively impact teachers, as levels of stress may be directly related to abilities and beliefs in learning a new skill or teaching the task (Pfitzner-Eden, 2016), in this case, CT.
An added layer of self-efficacy is efficacy expectations, which determine the effort individuals expend and persist on a given task or skill in adversity. Bandura (1977) described perceiving self-efficacy as something that not only shapes which activities and environments are chosen to engage with but the ways challenges are managed. Specifically, a person’s belief in capabilities determines the level of effort they are willing to invest and how long they might keep going when faced with discomfort. The more confident someone is, the more actively and persistently engagement occurs. Individuals who push through situations they find threatening, even when those situations are safe, will accumulate experiences that validate and strengthen confidence, gradually dissolving defensive tendencies. However, those who give up too soon hold onto limiting beliefs and anxieties far longer, preventing any real growth or change.
At the same time, if individuals do not eliminate defensive behaviors based on a perceived threat or fear, they will not be self-efficacious, which may have long-lasting effects (Bandura, 1977, 1986). When discussing self-efficacy, Bandura and Locke (2003) made clear the role of perceived personal control leading to stress and anxiety on a person’s ability to accomplish a goal. For instance, as teachers begin to imagine how they might engage in innovative ways outside everyday instructional practices, perceived fears may cause stress and anxiety, leading to low self-efficacy. Bandura and Locke postulated that individuals’ self-efficacy may have either a self-enhancing (positive) or self-debilitating (negative) control over the ability to motivate or persevere with a desired outcome. On the other hand, if teachers are motivated to push themselves beyond the fears, stress, and anxiety, performance becomes a “bidirectional relationship where higher self-efficacy leads to pursuing more challenging activities and succeeding in these activities leads to higher self-efficacy” (Akcaoglu et al., 2023, p. 214).
While the research is clear that teacher self-efficacy is the change required to move teachers’ beyond current practices (Haatainen et al., 2021; Handtke & Bögeholz, 2023; Rachmatullah et al., 2023; Rich et al., 2017); the challenges for teachers may be varied as teachers face curriculum and standards reform, lack of teacher support through PD opportunities, and time constraints in the content areas, to name a few. For teachers to be successful implementing innovative practices, they require PD opportunities that build confidence.
Teacher Self-Efficacy in Elementary Science
The quality of teachers and the opportunities for instructional improvements are scarce for elementary science teachers, often leading to low self-efficacy. The elementary science workforce has less than 5 years of science teaching experiences (Plumley, 2019), coupled with fewer educators in the computer science and engineering fields, elucidating the need for professional learning opportunities for elementary science teachers. Effective science teaching requires that teachers have a solid grasp on the science curriculum and scientific inquiry; however, fewer elementary science teachers hold college or graduate degrees in science, engineering, or science education, leading to low confidence and often lower self-efficacy when teaching science.
Lukitasari et al. (2019) posited that elementary science teachers will require the content and knowledge base to teach scientific inquiry effectively in the 21st-century learning environment. Therefore, as the classroom infrastructure begins to evolve, early grades science offers a good home for integrating CT (Cabrera, 2021). For example, CT exposes students to computer science skills, like abstraction, decomposition, pattern recognition, and debugging (Bouck et al., 2021; Luo et al., 2023).
Computational Thinking in Elementary Science
The 21st-century is increasingly more robust in careers and industries related to computational skills used to solve complex problems. As CT becomes more prevalent in the K-12 learning environment, districts have increased PD opportunities for in-service teachers, and teacher education programs are more widely offering CT and computer science courses to meet new graduation requirements (Boulden et al., 2021). The need still exists, however, for CT at the elementary level, as CT is more broadly associated with secondary or postsecondary education (Cabrera, 2021; Rich et al., 2017). Elementary students who are provided opportunities to engage in CT in the earlier grades have a foundation of learning not only for computer science (Cabrera, 2021) and can develop computational literacy, allowing more opportunities for diverse academic careers (Kafai & Proctor, 2022; Kim et al., 2024).
Computational thinking in elementary science may present challenges due to low self-efficacy in a specific subject area as well as technological challenges; however, the Next Generation Science Standards (NGSS, 2022) included using mathematics and CT as one of the eight science and engineering practices outlining the pathway for teachers to begin thinking of how to integrate this skill. Although implementing new ways to teach in the elementary science classroom may leave teachers feeling less confident, creating barriers to overcome, teachers will require multiple opportunities and exposure to CT before effective instruction can take place (Kong et al., 2020). Computational thinking ought to be embedded in varying contexts to show how this problem-solving skill can be used in different subject-specific domains, such as math and science (Akcaoglu et al., 2023; Luo et al., 2023; Rich et al., 2017; Kjällander et al., 2021).
As Kong et al. (2020) outlined, elementary teachers may not have enough confidence to teach CT “because they lack an academic background in computer science (CS),” leading to low self-efficacy (p. 2). Furthermore, some elementary teachers may need more confidence in teaching science (Gilbert et al., 2016), as they are not adequately prepared during in-service practices, with fewer completing majors in computer science and mathematics (Gray et al., 2022). Therefore, low self-efficacy may inhibit elementary teachers from taking risks to learn an innovative skill, and more specifically, a technological construct such as CT. Again, some teachers may struggle with the pedagogical, technological, and content-based knowledge for CT integration; however, through high-quality PD and with patience, CT-infused science can occur using a combination of plugged and unplugged activities (Cabrera, 2021; Luo et al., 2023). Teachers might start small with various plugged activities (using a computer) while integrating unplugged activities, where children learn concepts such as abstraction, decomposition, debugging, and pattern recognition (Luo et al., 2023; Ridgway, 2020).
Teacher Self-Efficacy in CT-Infused Science
Thomson et al. (2019) examined the development of elementary science teachers over 5 years, capturing the entire length of the teachers’ preparation into the 1st years of teaching. The research on self-efficacy in science teaching shows that most elementary science teachers tend to feel less self-efficacious while holding negative opinions about teaching science (Avci & Deniz, 2022; Rich et al., 2017) than secondary counterparts (Thomson et al., 2019). Secondary science teachers are specialists in the field; therefore, confidence in teaching science far outweighs that of elementary science teachers, who are considered generalists, often leading to lower self-efficacy while teaching science (O’Dwyer et al., 2023). The study’s results indicated a change in teachers’ self-efficacy while teaching science, with higher self-efficacious beliefs during preservice implementation, followed by a decline during the first 2 years of teaching (Thomson et al., 2019).
Further studies show the impact CT-infused science teaching might have on early elementary school students (Cabrera, 2021; Kim et al., 2024). As computer science education becomes increasingly more critical in today’s K-12 classrooms, CT is often integrated into core content areas such as science and math. However, teachers’ confidence in the technological and pedagogical skills for implementing CT-infused science waivers.
For example, Luo et al. (2023) found that providing teachers with opportunities to codesign lessons with instructional materials and resources with the tools to be successful while implementing CT offered potential. The research is consistent that CT integration in science should first be a part of the preservice curriculum (Cabrera, 2021), followed by in-service teachers making CT a part of daily lives (Akcaoglu et al., 2023).
Previous research indicates that self-efficacy is crucial for teachers integrating CT into science lessons. While elementary teachers may exhibit low self-efficacy for implementing CT-infused science lessons, PD can significantly increase confidence levels (Cabrera, 2021; Boeve-de Pauw et al., 2022). If teachers are not provided with the tools such as ongoing PD to develop mastery experiences (Bandura, 1997) in a given subject or learning construct, stress, anxiety, and fear (Bandura, 1977; Bandura & Locke, 2003) often play a role, and, unfortunately, elementary teachers are typically not given PD opportunities to utilize the 21st-century skill of CT (Cabrera, 2021; Rich et al., 2017).
Methods
PD Opportunities Using CT-Infused Elementary Science
Professional development designed to positively change teacher instructional practices includes five key components: “appropriate content, active learning, consistency, sufficient time, and collective participation” (Şahin et al., 2024, p. 3), as well as ongoing support through “coaching, modeling, observation, and feedback” (Wei et al., 2010, p. 12). In this study, the goal of PD1 and PD2 as well as the academic year PD was to build teacher confidence while integrating CT into elementary science through a community of practice (CoP; Lave & Wenger, 1991).
During PD1 and PD2 the teachers engaged in CT practices designed to build capacity for implementing CT in lessons. Through self-reflection and a CoP teachers created CT-infused science lessons for the academic year implementation. Furthermore, teachers were encouraged to attend weekly office hours and monthly hour-long PD in the form of online community gatherings. The online community gatherings were organized around topics related to CT in the elementary science classroom. For example, guest visitors showcasing CT tools such as micro:bits, lesson study through grade band discussions about CT-infused science, and opportunities for peer support using break out rooms.
Taken together, prior research on teacher self-efficacy, PD design, and computational thinking integration informs the rationale for the current study and its research questions. Drawing on Bandura’s theory of self-efficacy, literature on mastery experiences, verbal persuasion, vicarious experiences, and physiological and affective states provides the theoretical grounding for examining changes in teachers’ self-efficacy over time (Research Question 1; RQ1).
Research on CT-infused instruction and documented challenges in PD implementation further informs investigation of how teachers enacted CT practices in elementary science lessons and engaged with ongoing PD supports (Research Question 2 [RQ2]). Last, studies examining teacher learning trajectories and confidence development following PD provide context for exploring changes in self-efficacy across PD phases, including the observed decline following classroom implementation (Research Question 3 [RQ3]).
This study examines teacher self-efficacy in implementing CT-infused science lessons following participation in two PD workshops, academic year lesson observations, and community gatherings (Figure 1). Bandura’s theory of self-efficacy serves as the analytic lens used to determine whether the four sources of self-efficacy were addressed through the PD design, PD workshops, academic year PD, and community gatherings. The following section presents background information about the study, participant demographics, and procedures for the methods, data collection, and data analysis.
Figure 1
Professional Development Progression
A concurrent triangulation mixed-methods research design was employed. Concurrent triangulation designs “use both qualitative and quantitative data to more accurately define relationships among variables of interest” (Castro et al., 2010, p. 345). Qualitative data were derived from two focus group interviews, while quantitative data included participants’ T-SelECTS (Teacher Self-Efficacy for Integrating Computational Thinking in Science; Cabrera et al., 2021) scores, academic year PD participation, and lesson implementation. Together, these data were used to characterize teachers’ self-efficacy in implementing CT-infused science lessons. Data were collected for two summer PD workshops held over 2 years (PD1 and PD2) and during additional PD activities and lesson observations conducted in the academic years following each summer PD.
During PD1, participants examined CT through an existing CT framework (Figure 2) and engaged in CT-modeled activities while exploring CT tools such as Scratch, Arduino Science (computational data collection using a mobile phone), micro:bits, Edison Robots, Sage Modeler, PhET Simulations, and so forth. Throughout the 5-day PD, instruction about CT was scaffolded to provide teachers with the background knowledge to begin thinking more about how to integrate CT into elementary science lessons. Further, the PD functioned as a CoP in which teachers collaborated and offered one another support while learning this complex instructional skill. As a culminating activity, teachers were asked to use an existing science lesson to integrate CT using the CT framework as an instructional scaffold. At the conclusion of PD1, teachers presented lessons to one another in a gallery walk, while considering when and how they would use CT-infused science lessons during the academic year. Teachers also participated in a focus group interview on Day 5 of PD1.
Figure 2
CT Framework
A similar structure was followed for PD2, although teachers worked in grade-band teams rather than individually to create CT-infused science lessons, and the PD lasted only 3 days. As in PD1, the workshop included demonstrations of CT tools, and opportunities for encouragement and support as teachers experimented with the CT tools. Teachers again participated in a focus group interview on the final day of PD2.
During the academic years that followed PD1 and PD2, teachers were asked to attend monthly, virtual PD sessions (community gatherings), and to implement at least one CT-infused science lesson with students. When teachers implemented a CT-infused science lesson in the classroom, one or more trained observers observed the lesson using the CT framework. Pre- and postobservation conferences followed implementation. Throughout the academic years, teachers had the option of attending virtual office hours.
Teachers’ self-efficacy for integrating CT in science lessons was measured using the T-SelECTS survey during each summer PD. To minimize the Dunning-Kruger effect (Dunning, 2011), the survey was administered on the final day of each PD. The PD design provided teachers with access to all four sources of self-efficacy beliefs (Bandura, 1997). Designing and presenting CT-infused science lessons during the summer PDs, as well as implementing CT-infused science lessons during the academic years, offered sources of mastery experiences. Observing modeled CT tool use and engaging with fellow participants’ CT-infused science lessons provided a source of vicarious experience. Verbal persuasion was abundant during the summer PD workshops and academic year community gatherings; however, the PD intentionally withheld verbal persuasion during lesson observations, as well as during pre- and postobservation conferences to ensure that participants did not feel they were being evaluated during observations.
Last, when planning the PD activities of the project, meticulous attention was given to accessibility, ease of participation, and comfort to support teachers’ positive physiological and affective states. For example, PD sites were centrally located, PD dates were selected in consultation with participants, and frequent breaks and refreshments were provided throughout PD activities.
CT Framework
The CT framework (Figure 1) used throughout the PD summer workshops, lesson planning, and academic year implementation acted as both a guide and scaffold for teachers when considering integrating CT into elementary science lessons. The CT Framework was designed for and used in a previous study (Cabrera et al., 2021) for CT in elementary science. Teachers used the framework along with state science standards to create a lesson plan incorporating one or more of the CT practices (data, programming, computational simulations, and systems thinking).
Participants
Participants include elementary science teachers of first through sixth grade. The teachers represent two diverse school districts in the southwestern US. Of the 24 teachers who attended PD1, 23 completed the T-SelECTS survey (one teacher left the PD early and did not complete the survey, citing family reasons). All 18 teachers who returned for PD2 completed the T-SelECTS survey; however, the results from the teacher who left PD1 early are not included in this analysis. The resulting sample includes 17 teachers who completed the T-SelECTS survey during both summer PDs. Most of these teachers also attended at least one community gathering (14 of 17) and implemented one or more CT-infused science lessons that trained observers documented (16 of 17).
The demographics of the teachers as well as teaching experience vary; however, all teachers were currently teaching elementary science at the time of the study. Most of the teachers had taught over 10 years, with an average of 14 years teaching experience. The majority of the teachers (nine of 17) identify as White and female. The sample also includes two females who self-identify as Hispanic or Latino/a/x/é, two females who self-identify as Black, one female who self-identifies as Asian, two self-identifying White males, and one self-identifying White nonbinary.
Research Instruments and Procedure
As described previously, data sources include T-SelECTS scores, academic year PD participation, CT-infused science lesson implementation, and focus group interviews, used to determine if teacher participants experienced an improvement of self-efficacy for CT-infused science lessons. The decision to draw upon multiple data sources, both quantitative and qualitative, is influenced by Holzberger et al. (2013), who claimed that teacher self-assessment bias for self-efficacy may impede accurate teacher reporting, concluding that teacher self-assessment with additional data points such as classroom observations and student outcomes may yield the best results when focusing on teacher self-efficacy.
T-SelECTS Scale
The T-SelECTS scale (Figure 3) was developed to better understand how teachers, both preservice and in-service, learned to integrate CT into elementary science by measuring teachers’ self-efficacy for integrating CT in science (Cabrera et al., 2021). The T-SelECTS, a valid and reliable survey, comprises six items using a 100-point sliding scale (Figure 2). Bandura’s (2006) sliding scales were used to create the survey (0 = no confidence, 100 = high confidence in 10-point increments).
Figure 3
T-SelECTS Survey Items (Cabrera et al., 2021)
Lesson Observations
To document how teachers incorporated CT in science lessons, the study developed an observation protocol aligned with four CT practices: using data, programming, computational simulations, and systems thinking. Each observed CT practice was rated using 3-point Likert scale: Evident, Partially Evident, or Not Evident. The study established a 3-point ordinal rating system: “Evident” (E), “Partially Evident” (P), and “Not Evident” (N) rating system. The protocol uses the term “evident” because each data source was a partial window into how the teachers’ practice evolves. Early on, teachers’ practices were clearly multifaceted, and one piece of data was not sufficient to summarize the entire classroom practice. Under this system, an Evident rating indicated that there were unambiguous elements of the practice embedded in the lesson. However, a Partially Evident rating was applied when there was evidence of an ambiguous or incomplete attempt to embed a computational thinking practice in the lessons that were observed. A Not Evident rating indicated that no elements of CT were identified (see Bernier et al., 2025).
The protocol intentionally selected the term evident to reflect that observations represent a snapshot of instruction and may not capture the full extent of implementation. Five trained observers conducted lesson observations. The observers were members of the research team who completed training on the CT observation protocol prior to data collection. Before lesson observations, observers coded a sample lesson to ensure interrater reliability. Lesson observation ratings were not shared with participants.
Community Gatherings
Throughout Academic Year 1 (AY1), teachers were invited to participate in monthly virtual meetings, where they were joined by members of the research team. The community gatherings were designed to emphasize the shared commitment of both teachers and researchers, teachers received ongoing PD and a venue to ask questions, receive support, and encouragement. In total, seven community gatherings were held during AY1; attendance was taken at each gathering.
Focus Group Interviews
On the final day of each summer PD workshop, teachers were invited to participate in focus group interviews. Participants were divided into six focus groups. Each group was asked the same questions, although different interview protocols were used during PD1 and PD2. The purpose of using multiple interview protocols was to incorporate feedback and responses from the interviews to create PD that would ultimately provide teachers with sources to build confidence and identify any gaps in learning. During the focus group interviews, two facilitators asked questions designed to reveal teachers’ beliefs about CT. Focus groups were video and audio-recorded, and observational notes were taken.
Data Analysis
For the T-SelECTS analysis, mean scores were calculated for each participant, and item-level mean scores were compared to assess whether the trajectory of individual T-SelECTS items differed from that of the overall mean score. Although the quantitative analysis was limited by the small sample size, analyses were conducted using IBM SPSS Statistics (Version 29) to evaluate differences between participants’ PD1 and PD2 T-SelECTS scores with a paired-samples t test.
For lesson observations, observers used the CT framework to determine whether lessons contained evidence of at least one of the four CT practices; lessons demonstrating evidence of at least one CT practice were classified as CT-infused science lessons. For community gatherings, attendance was recorded for each participant, with scores ranging from 0 (no attendance) to 7 (attendance at all gatherings). Audio and video recordings from focus group interviews were transcribed using Rev transcription services. Tra nscript analysis focused on discussion related to self-efficacy for implementing CT-infused science lessons.
Results
Quantitative Results and Analysis
T-SelECTS Scale
The quantitative results reported in this section address RQ1, which examined changes in teachers’ self-efficacy for implementing CT-infused elementary science lessons across PD phases. When first surveyed at the end of PD1, participants’ mean T-SelECTS score was 88.89 out of 100 possible points, with scores ranging from 65.17 to 100, indicating that all teachers felt moderately to highly confident in abilities to implement CT-integrated science lessons. By the end of PD2, participants’ mean T-SelECTS score declined to 83.79, with scores ranging from 56.5 to 94 (Table 1). Overall, 13 out of 17 participants demonstrated a decrease in self-efficacy in implementing CT-integrated science lessons (Table 2). The results of a paired samples t-test provided evidence that the mean difference between PD1 and PD2 T-SelECTS scores is significantly different from 0 (t = 2.569, df = 16, p =0.021).
Table 1
Descriptive Statistics of PD1 and PD2 T-SelECTS Scores
Table 2
Participant T-SelECTS Scores
Lesson Observations
Findings in this section address RQ2, which focused on how teachers enacted CT practices during lesson observations. Of the 17 participants included in the sample, 16 implemented at least one science lesson that they considered to be CT infused, which was observed and rated by one or more trained observers (Table 3). Although 16 participants designed and implemented at least one science lesson that they believed was infused with at least one CT practice, only eight of those teachers’ lessons were rated as having one or more evident CT practices based on observation ratings. Although participants were asked to implement two CT-infused science lessons during the first academic year, less than half of the participants (seven total) met this requirement.
Table 3
Participant T-SelECTS Change and Potential Sources of Self-Efficacy Beliefs
Community Gatherings
This section also addresses RQ2 by describing teachers’ participation in ongoing PD support through community gatherings. During the academic year that followed the first summer PD workshop, participants were invited to attend seven community gatherings. Fourteen of the participating teachers attended at least one community gathering, while only one participant attended all seven community gatherings (Table 3). The average number of community gatherings attended per teacher was slightly less than half (3.29/7).
Analysis of Quantitative Data
This analysis addresses RQ3, which explored whether teachers’ classroom implementation experiences and PD participation were associated with changes in self-efficacy over time. Across the quantitative data sources (Table 3), no clear patterns emerged to explain the decline in participants’ self-efficacy for integrating CT-infused elementary science. Because a decline in T-SelECTS scores at the end of PD2 was not anticipated, analyses examined whether changes in self-efficacy were related to participants’ engagement with potential sources of self-efficacy beliefs during AY1. Specifically, analyses explored whether changes in T-SelECTS scores were associated with the presence of CT practices in lesson observations or participation in community gatherings. However, the data did not provide evidence of these relationships. For example, participants with the most significant increase in T-SelECTS scores (Participant 15, Δ = 14) attended six of the seven community gatherings but were unable to incorporate CT practices into an elementary science lesson successfully. Conversely, participants with the most significant decrease in T-SelECTS scores (Participant 6, Δ = -19.67) also attended six community gatherings but were able to demonstrate the incorporation of a CT practice during a lesson observation. A correlation analysis revealed no significant relationship between the quantitative variables. The correlations between the T-SelECTS score and the number of lesson observations, the number of CT practices evident, and community gathering participation were all weak and statistically insignificant (r = -0.19, -0.39, and 0.24, respectively; all p > 0.05).
Qualitative Results and Analysis
Focus Group Interviews
Qualitative findings in this section further addressed RQ2 by examining teachers’ perceptions and experiences related to CT implementation and PD participation. Thematic analysis was employed, utilizing both predetermined and emergent codes (Fereday & Muir-Cochrane, 2006) derived from the analyzed transcripts. The predetermined codes or themes, based on a preliminary review of the transcripts, were time constraints, district curriculum requirements, equating CT with tools, and a lack of feedback and exemplars. Upon reading the transcripts, reflexive thematic analysis was employed to facilitate easy access and flexibility in interpreting the qualitative data, as patterns and themes emerged.
Byrne (2021) cited Braun and Clark (2006), who stated that reflexive thematic analysis is about “the researcher’s reflective and thoughtful engagement with data and reflexive and thoughtful engagement with the analytic process” (p. 1393). The initial predetermined codes were organized with the transcripts, and a second phase of coding was performed using NVivo qualitative analysis software as patterns emerged. Next, Phase 3 of the coding process generated themes, while Phases 4 and 5 were used to outline and define themes. The five-phase reflexive thematic analysis used by Braun and Clark allows a cogent and concise narrative of the data.
Themes
Gift of Time. The first theme, both predetermined and emergent from the data, is time. Teachers were cognizant of the time commitment to plan and implement a CT-infused science lesson. As T1 stated, when asked about comfort levels with designing and implementing CT-infused science lessons, “I think you gave us the gift of time for it… because we’re so overwhelmed and teaching in the younger grades that are not subject teachers. We’re teaching every subject.… It was nice … to look at all my science units.”
To further note, T17 stated when asked the same question, “So, it’s something that I’m excited about, but it is going to be something that I’m going to have to persevere through and go in with an understanding that it’s going to take myself practice and time…”
When considering efficacy expectations and how much effort an individual will expend and persist on a given task, both T1 and T17 described time as a factor; however, there is an implied degree of optimism coupled with persistence in both responses. Interestingly, T1 and T17 each were observed twice during AY1 with at least one CT practice during each lesson observation. As Bandura (1986) argued, self-efficacy beliefs are as important as the skills related to learning new concepts, which influences and strengthens motivation leading to higher levels of self-efficacy often through mastery experiences.
CT Is Technology. The next theme emerging from the data, CT is technology, was evident in teachers’ responses about levels of comfort teaching this construct. Teachers equated CT as a skill most closely related to technology implementation, even though they were given a definition of CT as a skill that can be used with computers “plugged” or with computers “unplugged.” For instance, T7 said,
I feel like I have different levels of the comfortability of teaching it. For example, the computational thinking, I feel like I can go in there, but the different levels of technology is what I’m trying to say, like, with the robots. And I struggle to use the robot or to program the robots, whereas scratch is still difficult for me. So, I know that’s something that I can probably go home and teach myself and practice and become comfortable…
Moreover, T14 stated,
You wouldn’t even know that scratch existed. Same thing for every single gadget that we’ve used here, every single program that we’ve used, most of the concepts that we’ve tried out. I feel like some of us are comfortable instructing using those only because we had practical hands-on experience with them ourselves. And so, I’m not sure if I could teach my fellow teachers what I learned here today because I don’t have an Edison robot to show them. I would rate an eight.… Because I’m very technologically literate myself, and I find it easy to share that with my students.
Teachers who had higher technology efficacy showed higher confidence in how comfortable they would be implementing CT-infused science lessons, while teachers who understood CT as the tech tools shown during the summer PD workshops showed levels of doubt. However, both T7 and T14 implemented at least one lesson observation with at least one CT practice using computational simulations.
CT Is Hard to Implement. The next theme that emerged from the data is that CT is hard to implement and is difficult to plan due to limitations with curriculum initiatives put in place by the district. For example, T5 explained, “I would probably say for me, computational thinking is harder for me. I did see when we did things with Scratch… and then when I think about the concepts that I’m supposed to teach in science…”.
T14 went on to say,
Computational thinking, I think there’s a lot of barriers in computational thinking.… So I think computational thinking is not necessarily accessible to all teachers, and as a result they can’t instruct their students in it.… And I think the vast majority of teachers can definitely benefit from computational thinking.
Although T5 and T14 both asserted that CT may be difficult to implement, T14 could see the benefit in using CT in the classroom. While T5 and T14 both had reservations about how easily CT might be implemented in the classroom, both engaged in at least one lesson observation during AY1 with T14 implementing at least one CT practice, while T5 did not implement any CT practices. Considering district curriculum initiatives, T4 stated that both time and district constraints might impact lesson implementation:
We had a little bit of time to develop lessons where we were trying to implement those things and then kind of putting it all together when we came here this summer. Now I understand it. I found that I had more confidence creating a lesson that was my own, versus trying to use district curriculum and add computational thinking into it.
Coupled with teachers’ belief that implementing CT may be hard and with district curriculum constraints, teachers may have benefited from vicarious experiences through model observations of a CT-infused science lesson leading to the next theme.
Lesson Exemplars. The last predetermined and emergent theme from the data is the desire for lesson exemplars. Teachers discussed the need for reflective feedback and lesson exemplars so that they might know how well or not implementation was occurring. For example, T2 explained,
I feel pretty comfortable. The codebook is helping. Yeah. I’m the type of person that I like to know what I’m supposed to do right up front. And I feel like that would’ve helped me last summer just to see it laid out that way.
Upon completion of the first academic year, the need for an analytic tool became evident during data analysis, leading to the development of codebooks that were later made available to teachers. The codebooks were introduced during PD2 to support lesson planning and subsequent lesson observations. Teacher 2 referred to the codebook introduced during PD2 and was among many teachers who felt the tool was useful and helped with both lesson planning and lesson delivery.
Teacher 7 went on to say,
We were talking about, and I don’t know if this is something you could do or not, but it’d be cool to see if snippets about other people’s lessons, of pieces that you thought, maybe just a piece, that might be went well or something that was a good exemplar of CT.… And it even could be like, if you want it to be educational, the teacher talking, it could be a part of a PD where we decide what lens is being used here. It’d be nice to see other people’s little pieces of their lessons, too.
Although T12 did not implement an AY1 or AY2 CT-infused science lesson, they did address the need for lesson sharing or exemplars during the focus group interviews and via email:
Focus group interview: One really dynamic element to this whole thing. This is kind of a tangent, but that’s what I do. I just really like the collective. It was really neat kind of picking brains. And I wish there had been more, and people didn’t seem as into it, but I really wish there had been more of a think tank in terms of dropping it all into a document where you could get all 35 people.
Email exchange: I just had an idea regarding your upcoming visit to my classroom. Would you, or one of your fellow scientists, want to model a lesson in my classroom? I just have so many questions regarding CT and CR theory in classrooms, and thought “How cool would it be to observe a lesson with my group of students?” My thoughts are that, say you wanted to model a Coding or Robotics lesson….
Whether verbal persuasion in the form of effective feedback following lesson observations or during lesson design positively influenced teachers’ self-efficacy beliefs remains open to debate. Burger (2024) noted that verbal persuasion via performance feedback can enhance self-efficacy; however, the study design did not include formal evaluation of teacher performance. The study design also did not include lesson exemplars with teachers to avoid encouraging replication rather than original lesson design.
Another source of self-efficacy evident in teacher discussions and email exchanges involved physiological and affective states related to stress about lesson implementation. At the conclusion of PD1 and PD2, teachers demonstrated relatively high confidence, likely supported by ongoing reassurance and positive feedback. In contrast, during AY1 PD, a nonparticipatory role was adopted (i.e., observing without intervening or altering the situation in any way; Barrios et al., 2024) during lesson observations, including pre- and postobservation conferences, which may have negatively influenced teachers’ self-efficacy beliefs.
Discussion
This concurrent triangulation mixed-methods study examined changes in teacher self-efficacy across an approximately 2-year period. Teachers were given opportunities to engage in two PD workshops beginning in the summer of Year 1 and the summer of Year 2. During the academic year, teachers were invited to attend monthly meetings during community gatherings. Through these gatherings, teachers were offered a variety of tools to provide support while considering how to integrate CT into elementary science lessons. A valid and reliable instrument was used to survey teachers’ confidence with CT-infused science lessons at two points, PD1 and PD2. In relation to RQ1, these findings documented how teachers’ self-efficacy for implementing CT-infused elementary science lessons changed over time and across PD phases.
The survey results showed that teachers had higher confidence in CT-infused science after PD1 than after the implementation of the AY1 and PD2. Several research studies showed that teachers experienced a peak after initial exposure to a concept or PD, leading to higher results of self-efficacy followed by a decline (Akcaoglu et al., 2023; Boeve-de Pauw et al., 2022; Pfitzner-Eden, 2016; Rich et al., 2017; Thomson et al., 2019). It was also possible that classroom implementation functioned as a challenging mastery experience that unsettled teachers’ initial confidence, while nonimplementation may have allowed some teachers to avoid experiences that could threaten self-efficacy (Bandura, 1977).
Qualitative themes related to time constraints, uncertainty about CT and stress associated with lesson implementation helped contextualize this quantitative decline in self-efficacy. The results of the data suggest that teachers’ confidence shifted as they encountered the realities of classroom enactment.
Bandura and Locke (2003) posited that teachers may have self-enhancing or self-debilitating beliefs that can lead to perceived abilities or inabilities to act. In response to RQ3, the observed decline in self-efficacy was interpreted through Bandura’s theoretical framework as a recalibration process shaped by mastery experiences and teachers’ physiological and affective states. Teachers whose confidence declined from PD1 to PD2, even while attending community gatherings and conducting lesson observations, may have had self-debilitating beliefs addressing physiological and affect states induced by stress or anxiety. As Bandura and Locke (2003) stated,
In managing challenges in performance situations, people need a resilient sense of efficacy that they can achieve desired results by efforts and try to remain unfazed by setbacks or failure. One cannot execute well-established skills while beset with self-doubt. In applying what one knows, a strong belief in one’s performance efficacy is essential to mobilize and sustain the effort necessary to succeed. (p. 97)
Addressing RQ2, findings related to lesson observations and participation in PD activities highlight how teachers’ enactment of CT practices and access to instructional support shaped self-efficacy beliefs. These findings prompt consideration of how effectively the PD design provided opportunities for verbal persuasion and mastery experiences. Considering verbal persuasion in the form of verbal encouragement, teachers may have needed and required this self-efficacy source in the form of feedback for delivery of a CT-infused science lesson during classroom observations. Offering teachers feedback was outside the scope of this study, but this may have led to the decline in teacher confidence from PD1 to PD2. The research on high-quality PD outlines critical points of high-quality PD that include “support through coaching, modeling, observation, and feedback” (Wei et al., 2010, p. 12).
A nonparticipatory observation role was adopted during lesson observations, and no lesson feedback was provided to teachers during the postobservation meeting. Qualitative findings further helped interpret the lack of clear quantitative relationships between self-efficacy change and participation measures, as teachers frequently described needing more targeted feedback and clearer instructional models rather than additional PD.
Moreover, as Bandura (1977, 1986) stated, without appraisals of sufficient or insufficient performance, the incentive for action to change becomes limited, elucidating the need for feedback. If teachers had been given feedback on performance, perceptions may have changed due to understanding expectations toward mastery of CT-infused science. Furthermore, the outcome expectancy may have been limited by teachers’ need for feedback to ascertain whether the outcome was executed, as CT-integrated science was a new pedagogical construct for many. The efficacy expectations were in line with Bandura’s theory, as teachers had higher self-efficacy scores after PD1. Teachers may have had a strong belief toward executing the outcome before self-debilitating beliefs, doubt, stress, anxiety, and even fear interrupted self-efficacious behaviors. In addition, annotating areas of success and improvement through feedback may have helped teachers’ perceptions of each subsequent lesson visit providing opportunities for mastery experiences.
Implications and Conclusions
As a result of the current study, teacher self-efficacy for integrating CT-infused elementary science offers several layers for consideration. Teachers’ self-enhancing and self-debilitating behaviors, particularly those related to physiological and affective states, may shape how teachers view themselves when implementing new and innovative instructional tools such as CT. High-quality PD that includes opportunities for feedback (verbal persuasion) may support teachers’ judgments and decisions about learning. Participating teachers were asked to integrate CT into science lessons while simultaneously learning new instructional tools. In several cases, this involved integrating unfamiliar technologies such as Scratch, micro:bits, Edison Robots, and PhET simulations. Therefore, this research raises new challenges and questions to consider when implementing CT-infused elementary science.
First, teachers not accustomed to using digital technology in the classroom will require additional support (vicarious expectations) to raise levels of self-efficacy when learning more complex instructional practices. Although CT can be used unplugged, as previously discussed, the expectation is that the unplugged version will eventually be used in service of the plugged. So, at some point, computers will be used to teach the skill to students, whether it’s using data, programming, computational simulations, or systems thinking. At some point, teachers and students will require the technological acumen to consider using these CT practices in classrooms. Integrating CT in science is an ongoing process that requires more than one or two PD experiences; sustained, high-quality PD with multiple opportunities to navigate this instructional skill is essential for building teacher capacity.
Therefore, teacher support through ongoing PD that includes peer models of CT-infused science lessons may be beneficial. CT-infused science lessons were deliberately not shared with teachers to avoid encouraging replication rather than original lesson design. In hindsight, and when considering vicarious experiences as a source of self-efficacy, access to instructional models may have supported teachers in bridging the gap between self-debilitating and self-enhancing beliefs related to physiological and affective states. Integrating CT-infused elementary science also involves a relatively high learning curve, particularly for teachers with lower technology efficacy. More deliberate support in distinguishing between plugged and unplugged approaches to CT may have provided additional opportunities for teachers to develop a clearer understanding of this problem-solving skill.
During lesson observations, teachers frequently relied on CT tools without a clear delineation between the science standards being addressed and the role of the tool in supporting those standards, potentially limiting opportunities for mastery experiences. As emerging technologies such as generative artificial intelligence (AI) become increasingly present in K–12 classrooms, future research should examine how these tools influence teachers’ self-efficacy for CT-infused instruction. Generative AI may simultaneously introduce new sources of uncertainty while also serving as a support for lesson planning and implementation, potentially shaping teachers’ efficacy beliefs in complex and competing ways.
Next, this research raises challenges about high-quality PD for elementary science teachers integrating CT in science. As previously mentioned, high-quality PD includes supporting teachers through feedback inclusive of coaching, modeling, and observation, which the latter three were included in this study; however, offering the teachers feedback was not (verbal persuasion). Teachers may have required feedback as they began to integrate CT into science lessons. In addition to opportunities for coaching and modeling, teachers were reluctant to schedule and join office hours and community gatherings during the academic year, which might have yielded more robust efficacy results. In-person gatherings or in-person office hours might have been beneficial for teachers so that additional modeling of CT tools in classroom contexts could occur. This may have been especially helpful for teachers who may have held self-debilitating beliefs (physiological and affective states).
Last, results from the T-SelECTS show a decline in teacher confidence between PD1 and PD2, while participation in community gatherings and lesson observations did not appear to contribute meaningfully to this change. Drawing on survey data, lesson observations, community gathering participation, office hours, and email exchanges, self-enhancing and self-debilitating beliefs, particularly those related to physiological and affective states, emerged as influential factors shaping teachers’ self-efficacy for lesson implementation. In addition, high-quality PD for CT integration in science appears necessary to support teacher growth through mastery experiences and verbal persuasion, particularly when engaging with complex instructional practices.
Supporting teachers in identifying and addressing self-debilitating beliefs may be critical for fostering sustained engagement with CT-infused elementary science. Bandura (1977) stated, “Expectation alone will not produce desired performance if the component capabilities are lacking. Moreover, there are many things that people can do with certainty of success that they do not perform because they have no incentives to do so” (p. 194). Although teachers in this study were monetarily incentivized to participate, these findings underscore the importance of feedback and support, suggesting that, much like students, teachers may also require meaningful academic feedback to strengthen self-efficacy (Wei et al., 2010).
Acknowledgement
This material is based upon work supported by the National Science Foundation under Grants No. 2101039 and No. 2101526. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
References
Akcaoglu, M., Özcan, M. Ş., & Hodges, C. B. (2023). Exploring the relationship among motivational constructs and preservice teachers’ use of computational thinking in classrooms. Computers in the Schools, 40(2), 213–229. https://doi.org/10.1080/07380569.2023.2172987
Avcı, C., & Deniz, M. N. (2022). Computational thinking: Early childhood teachers’ and prospective teachers’ preconceptions and self-efficacy. Education and Information Technologies, 27(8), 11689–11713. https://doi.org/10.1007/s10639-022-11078-5
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
Bandura, A. (1986). The explanatory and predictive scope of self-efficacy theory. Journal of Social and Clinical Psychology, 4(3), 359–373. https://doi.org/10.1521/jscp.1986.4.3.359
Bandura, A. (1997). Self-efficacy: The exercise of control. W.H. Freeman.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares & T. Urdan (Eds.), Self-efficacy beliefs of adolescents (Vol. 5, pp. 307-337). Information Age Publishing.
Bandura, A., & Locke, E. (2003). Negative self-efficacy and goal effects revisited. Journal of Applied Psychology, 88(1), 87–99. https://doi.org/10.1037/0021-9010.88.1.87
Barrios, H. E. M., Roys, E. Y. M., Maestre, B. O., & Araujo, H. A. A. (2024). Contributions to the application of non-participant observation as a research technique. Power System Technology, 48(4), 174-185.
Bernier, J., Kramarczuk, K., Terrell Shockley, E., Figueroa, F., Yan, L., Xin, Y., Mak, J., Su, M., Ketelhut, D. J., & Nelson, B. (2025). Exploring culturally responsive teaching practices in computational thinking + science lesson planning. Discover Education, 4(1), Article 465. https://doi.org/10.1007/s44217-025-00897-6
Boeve-de Pauw, J., Olsson, D., Berglund, T., & Gericke, N. (2022). Teachers’ ESD self-efficacy and practices: a longitudinal study on the impact of teacher professional development. Environmental Education Research, 28(6), 867–885. https://doi.org/10.1080/13504622.2022.2042206
Bokhove, C., & Downey, C. (2018). Mapping Changes in Support: A Longitudinal Analysis of Networks of Pre-Service Mathematics and Science Teachers. Oxford Review of Education, 44(3), 383–402. https://doi.org/10.1080/03054985.2017.1400427
Bouck, S. P., Long, H., & Yadav, A. (2021). Preparing special education preservice teachers to teach computational thinking and computer science in mathematics. Teacher Education and Special Education, 44(3), 221–238. https://doi.org/10.1177/0888406421992376
Boulden, D. C., Rachmatullah, A., Oliver, K. M., & Wiebe, E. (2021). Measuring in-service teacher self-efficacy for teaching computational thinking: development and validation of the T-STEM CT. Education and Information Technologies, 26(4), 4663–4689. https://doi.org/10.1007/s10639-021-10487-2
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. https://doi.org/10.1191/1478088706qp063oa
Burger, J. (2024). Constructivist and transmissive mentoring: Effects on teacher self-efficacy, emotional management, and the role of novices’ initial beliefs. Journal of Teacher Education, 75(1), 107–121. https://doi.org/10.1177/00224871231185371
Byrne, D. (2021). A worked example of Braun and Clarke’s approach to reflexive thematic analysis. Quality & Quantity, 56(3), 1391–1412. https://doi.org/10.1007/s11135-021-01182-y
Cabrera, L. (2021). Computational thinking in the elementary classroom: How teachers appropriate CT for science instruction (Order No. 28865090). Available from ProQuest Dissertations & Theses Global. (2678809704). http://login.ezproxy1.lib.asu.edu/login?url=https://www.proquest.com/dissertations-theses/computational-thinking-elementary-classroom-how/docview/2678809704/se-2
Cabrera, L., Byrne, V., Jass Ketelhut, D., Coenraad, M., Killen, H., & Plane, J. (2021). Measuring teacher self-efficacy for integrating computational thinking in science (T-SelECTS). Educational Innovations and Emerging Technologies, 1(1).
Cadieux et al., 2020
Castro, F. G., Kellison, J. G., Boyd, S. J., & Kopak, A. (2010). A methodology for conducting integrative mixed methods research and data analyses. Journal of Mixed Methods Research, 4(4), 342–360. https://doi.org/10.1177/1558689810382916
Dunning, D. (2011). The Dunning-Kruger Effect: On being ignorant of one’s own ignorance. In [EDITOR NAME(S)?], Advances in experimental social psychology (Vol. 44, pp. 247–296). Elsevier Science & Technology. https://doi.org/10.1016/B978-0-12-385522-0.00005-6
Fereday, J., & Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A hybrid approach of inductive and deductive coding and theme development. International Journal of Qualitative Methods, 5(1), 80-92. https://doi.org/10.1177/160940690600500107
Gilbert, A., Bloomquist, D., & Czerniak, C. M. (2016). Using mathematical and computational thinking in early grades science classrooms: The need for high-quality professional development. School Science and Mathematics, 116(4), 175–176. https://doi.org/10.1111/ssm.12174
Gray, L., Taie, S., & Lewis, L. (2022). Qualifications of public school mathematics and computer science teachers in 2017-18 (Data Point, NCES 2022-026). National Center for Education Statistics.
Haatainen, O., Turkka, J., & Aksela, M. (2021). Science teachers’ perceptions and self-efficacy beliefs related to integrated science education. Education Sciences, 11(6), 272. https://doi.org/10.3390/educsci11060272
Handtke, K. & Bögeholz, S. (2023). Self-efficacy beliefs as well as perceived advantages and challenges of interdisciplinary science teaching from a longitudinal perspective. Education Sciences, 13(2), 212. https://doi.org/10.3390/educsci13020212
Holzberger, D., Philipp, A., & Kunter, M. (2013). How teachers’ self-efficacy is related to instructional quality: A longitudinal analysis. Journal of Educational Psychology, 105(3), 774-786. https://doi.org/10.1037/a0032198
Jaipal-Jamani, K., & Angeli, C. (2017). Effect of robotics on elementary preservice teachers’ self-efficacy, science learning, and computational thinking. Journal of Science Education and Technology, 26(2), 175–192. https://doi.org/10.1007/s10956-016-9663-z
Jocius, R., O’Byrne, W. I., Albert, J., Joshi, D., Robinson, R., & Andrews, A. (2021). Infusing computational thinking into STEM teaching: From professional development to classroom practice. Educational Technology ＆ Society, 24(4), 166–179. https://doi.org/10.30191/ETS.202110_24(4).0013
Kafai, Y. B., & Proctor, C. (2022). A revaluation of computational thinking in K–12 education: Moving toward computational literacies. Educational Researcher, 51(2), 146–151. https://doi.org/10.3102/0013189X211057904
Kim, J., Leftwich, A., & Castner, D. (2024). Beyond teaching computational thinking: Exploring kindergarten teachers’ computational thinking and computer science curriculum design considerations. Education and Information Technologies, 29, 15191–15227. https://doi.org/10.1007/s10639-023-12406-z
Kjällander, S., Mannila, L., Åkerfeldt, A., & Heintz, F. (2021). Elementary students’ first approach to computational thinking and programming. Education Sciences, 11(2), 80.
Kong, S.-C., Lai, M., & Sun, D. (2020). Teacher development in computational thinking: Design and learning outcomes of programming concepts, practices and pedagogy. Computers and Education, 151, 103872. https://doi.org/10.1016/j.compedu.2020.103872
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge University Press.
Luo, F., Ijeluola, S. A., Westerlund, J., Walker, A., Denham, A., Walker, J., & Young, C. (2023). Supporting elementary teachers’ technological, pedagogical, and content knowledge in computational thinking integration. Journal of Science Education and Technology, 32(4), 583–596. https://doi.org/10.1007/s10956-023-10045-0
Lukitasari, M., Purnamasari, I., Utami, S., & Sukri, A. (2019). Blended-problem-based learning: How its impact on students’ critical thinking skills? JPBI (Jurnal Pendidikan Biologi Indonesia), 5(3), 425–434. https://doi.org/10.22219/jpbi.v5i3.10048
Mak, J., Xin, Y., Yan, L., Kramarczuk, K., Figueroa, F., Nelson, B., … & Ketelhut, D. J. [LIST ALL AUTHORS UP TO 19 PER APA] (2025). Accessible computational thinking in elementary science: an aligning framework for teacher learning and implementation. Journal of Technology and Teacher Education, 33(1), 67-107.
NGSS Lead States. (2022). Next generation science standards: For states, by states. The National Academies Press. https://doi.org/10.17226/18290
O’Dwyer, A., Hourigan, M., Leavy, A. M., & Corry, E. (2023). “I have seen STEM in action and it’s quite do-able!” The impact of an extended professional development model on teacher efficacy in primary STEM Education. International Journal of Science and Mathematics Education, 21(Suppl 1), 131–157. https://doi.org/10.1007/s10763-023-10361-2
Pfitzner-Eden, F. (2016). Why do I feel more confident? Bandura’s sources predict preservice teachers’ latent changes in teacher self-efficacy. Frontiers in Psychology, 7, 1486–1486. https://doi.org/10.3389/fpsyg.2016.01486
Plumley, C. L. (2019). 2018 NSSME+: Status of elementary school science. Horizon Research
Rachmatullah, A., Hinckle, M., & Wiebe, E. N. (2023). The role of teachers’ self-efficacy beliefs and habits in differentiating types of K–12 science teachers. Research in Science Education, 53(2), 337–355. https://doi.org/10.1007/s11165-022-10060-y
Rich, P. Jacob., Jones, B., Belikov, O., Yoshikawa, E., & Perkins, M. (2017). Computing and engineering in elementary school : The effect of year-long training on elementary teacher self-efficacy and beliefs about teaching computing and engineering. ERIC Clearinghouse. [IS THERE NO PAPER NUMBER OR DOI?]
Ridgway, R. (2020). Building bridges with computational thinking. Science Scope, 43(8), 12–15.
Şahin, E., Sari, U., & Sen, O. F. (2024). STEM professional development program for gifted education teachers: STEM lesson plan design competence, self-efficacy, computational thinking and entrepreneurial skills. Thinking Skills and Creativity, 51, 101439. https://doi.org/10.1016/j.tsc.2023.101439
Thomson, M. M., Huggins, E., & Williams, W. (2019). Developmental science efficacy trajectories of novice teachers from a STEM-focused program: A longitudinal mixed-methods investigation. Teaching and Teacher Education, 77, 253–265. https://doi.org/10.1016/j.tate.2018.10.010
Tschannen‐Moran, M., & McMaster, P. (2009). Sources of self‐efficacy: Four professional development formats and their relationship to self‐efficacy and implementation of a new teaching strategy. The Elementary School Journal, 110(2), 228–245. https://doi.org/10.1086/605771
Wei, R. C., Darling-Hammond, L., & Adamson, F. (2010). Professional development in the United States: Trends and challenges. INCOMPLETE ENTRY
Wing, J. M. (2006). Computational thinking. Communications of the ACM, 49(3), 33–35. https://doi.org/10.1145/1118178.1118215
