Treffer: Cracking the Code: Enhancing Teacher Self-Efficacy and Student Engagement through Professional Learning in Computer Science and Gifted Education

Code.org is a leading non-profit organization providing professional learning (PL) for K-12 teachers across the United States to advance computer science (CS) education. While PL is known to improve teaching practices and student outcomes in general education, CS, and gifted education, limited research has specifically examined how CS-focused PL influences teachers working with gifted students (GT). This study aimed to assess the effects of CS-focused PL on teachers' self-efficacy and classroom practices, particularly in supporting high-ability learners. We collected data through teacher surveys, classroom observations, and focus group interviews. Results showed notable increases in teacher self-efficacy, although classroom practices demonstrated variability. The study discusses potential modifications to the PL framework to enhance classroom implementation and better meet the needs of gifted students. These findings have important implications for sustaining improved instructional practices and effectively challenging GT students in CS education.

As Provided

AN0187842666;gct01oct.25;2025Sep11.06:02;v2.2.500

Cracking the Code: Enhancing Teacher Self-Efficacy and Student Engagement Through Professional Learning in Computer Science and Gifted Education 

Code.org is a leading non-profit organization providing professional learning (PL) for K-12 teachers across the United States to advance computer science (CS) education. While PL is known to improve teaching practices and student outcomes in general education, CS, and gifted education, limited research has specifically examined how CS-focused PL influences teachers working with gifted students (GT). This study aimed to assess the effects of CS-focused PL on teachers' self-efficacy and classroom practices, particularly in supporting high-ability learners. We collected data through teacher surveys, classroom observations, and focus group interviews. Results showed notable increases in teacher self-efficacy, although classroom practices demonstrated variability. The study discusses potential modifications to the PL framework to enhance classroom implementation and better meet the needs of gifted students. These findings have important implications for sustaining improved instructional practices and effectively challenging GT students in CS education.

Keywords: gifted education; professional learning; computational thinking; underserved populations; instructional strategies

Teachers highlighted the value of PLCs, where researchers modeled effective CS instruction using best practices and pedagogies."

Introduction

Code.org® is a leading non-profit organization that aims to make computer science (CS) accessible to K–12 students across the United States. With over 80 million students using its platform, Code.org has been instrumental in preparing new CS teachers, shaping educational policies, and helping students pursue computing careers ([9]; [36]). However, challenges persist in delivering equitable CS education, especially in rural and low-income areas, where disparities in access to resources and qualified instructors are pronounced ([3], [2]; [28]).

These equity-based challenges are particularly acute for gifted students in underserved regions, who often lack access to advanced coursework, leaving their potential untapped (e.g., [34]). Research shows that targeted professional learning (PL) can foster equitable learning environments, enhance teaching practices, and improve student outcomes ([17]; [53]). Moreover, PL in CS has been shown to positively influence teachers' learning and classroom practice ([30]). Little, however, is known about the design features, implementation, and outcomes of CS-focused PL as it relates to gifted students.

This research fills that gap by exploring the intersection of PL in CS education and best practices in gifted education. Focusing on rural and low-income areas, like West Virginia (WV), we examine how targeted PL can enhance teacher self-efficacy, improve classroom practices, and support gifted students in developing computational thinking (CT) skills.

Context

West Virginia is predominantly rural, with many counties classified as persistently impoverished, meaning the poverty rate has been at least 20% for more than 30 years ([4]; [14]; [48]). Rural states, including West Virginia, consistently underperform compared to non-rural states in standardized measures of achievement for students in grades 3 through 8 ([21]; [33]). This underachievement highlights the need for focused educational interventions that can bridge the gap for students in these regions.

Code.org collaborates with local partners, like West Virginia University (WVU) and the West Virginia Department of Education, to increase CS education access through initiatives such as Code WV. These partnerships offer free PL, curricula, and support for K–12 schools, with a special focus on underserved districts that face resource constraints and competing priorities.

Our research team, funded by a U.S. Department of Education Jacob Javits Gifted and Talented Education Award [grant # s206a190014], has spent four years studying the implementation of these initiatives in rural Appalachia. Our focus is on how PL can equip teachers to better meet the needs of gifted students, especially those with advanced CT abilities who require more than standard instructional approaches to thrive in CS education.

We studied two schools within the same district: a pilot school with 644 students in grades K–2 and a scale-up school with 600 students in grades 3–5. At the pilot school, 65% of students are eligible for free or reduced-price meals (FARM), while 61% of students at the scale-up school qualify ([33]). Both schools meet the criteria for high-poverty status under the Elementary and Secondary Education Act (ESEA). In 2022, West Virginia students in poverty who qualified for the National School Lunch Program (NSLP) scored, on average, 14 points lower in mathematics and 24 points lower in reading than their non-NSLP peers ([32]). This achievement gap highlights the need for targeted interventions in high-poverty areas.

Key Challenges and Focus Areas

Three primary challenges exist in integrating CS education into K–12 curricula:

(1) Teacher Preparation: Teachers need the knowledge and skills to effectively teach CT.

(2) Student Engagement: Sustaining student engagement with CS content is essential, especially for keeping gifted students interested.

(3) Curriculum Implementation: Successful CS curriculum implementation requires alignment with best practices in pedagogy and gifted education strategies.

In addressing these challenges, our research examines the influence of PL on teachers' beliefs about coding, their self-efficacy in teaching CT, and their classroom practices. We also explore how PL can better address the needs of gifted students, who often require differentiated instruction and intellectual rigor to remain engaged.

Review of Literature

Professional Learning and Gifted Education

Effective PL is characterized by being content-focused, active, sustained, ongoing, coherent, and collaborative ([17]). For teachers of gifted students, PL should emphasize advanced pedagogical strategies, content mastery, and the ability to provide appropriate challenges ([20]). Aligning PL with these needs is essential for bridging the excellence gap between low-income gifted students and their higher-income peers.

PL should also cultivate teacher leaders ([12]), align with professional standards and school priorities ([10]; [11]; [20]), emphasize subject matter content (e.g., [10]; [11]), and incorporate effective pedagogical strategies (e.g., [1]; [18]; [29]). These strategies help students master challenging content, engage in problem-solving, and develop skills in effective communication, collaboration, and self-direction ([10]).

Critical thinking is a fundamental aspect of gifted education programming that involves the ability to conceptualize, apply, analyze, synthesize, and evaluate information collected through observation, experience, reflection, reasoning, or communication ([42]). Key strategies to promote critical thinking include determining cause and effect, making decisions, comparing and contrasting, classifying, observing, planning, and predicting ([8]). Formative instructional strategies, such as questioning ([46]), providing instructional feedback, and engaging in reflective dialogue further support these processes. Research has consistently shown that fostering critical thinking through rigorous curricula has positive effects on both cognitive and affective outcomes of students (e.g., [26]; [27]; [47]). In our PL, we integrated critical thinking strategies to help teachers better support gifted students.

Common PL models include an initial workshop followed by ongoing support ([10]) through follow-up workshops, online learning modules, mentoring, coaching, and Professional Learning Communities (PLCs) (e.g., [10]; [13]; [51]). Research links sustained PL to improved teaching practices and student outcomes (e.g., [10]; [31]; [39]), particularly when PL aligns with teachers' instructional needs ([11]). Challenges, however, persist in designing PL that is accessible, relevant, and sustained.

Our study draws on these best practices in PL, CS education, and gifted education to propose enhancements to Code.org's PL offerings. These enhancements are designed to help teachers better support students with high computational abilities, fostering a more inclusive environment for all learners (see Appendix A for a timeline of PL interventions).

CT and Gifted Education

CT involves key concepts such as sequences, loops, events, and conditionals, whereas key CT problem-solving techniques include decomposition, pattern recognition, algorithmic thinking, and debugging ([6]). These skills are essential for problem-solving in CS ([52]) and are particularly well-suited for gifted students who thrive on intellectual challenges (e.g., [45]; [49]) and complex thought processes (e.g., [23]).

PL focused on CT improves teachers' understanding of CT concepts, strengthens pedagogical knowledge, and increases their self-efficacy in teaching CT ([24]; [41]). Effective PL also supports gifted students by helping teachers differentiate instruction and provide the intellectual rigor they need to stay engaged.

Research Questions

Our research addressed the following questions:

(1) How does PL impact teachers' self-efficacy in teaching CT?

(2) How does PL influence teachers' classroom practices?

(3) How does PL help teachers meet the learning needs of gifted and talented students?

Next, we describe our research efforts by year, including our research processes, methods, data collection, analysis, and findings. We also describe how our research evolved in response to changes in the environment, community needs, and advancements in the field of CT.

Our Initial Efforts: Professional Learning in Year 1 (January 2020 – May 2020)

During the first year, educators in the pilot school were offered PL led by trained Code.org facilitators. The core of these offerings was a one-day Introductory CS Fundamentals workshop. This workshop introduced foundational CS concepts, guided teachers through the curriculum, provided access to various resources, and offered hands-on support for classroom implementation.

Teachers were also provided with an online course for students, which included both plugged and unplugged activities to teach computational thinking, problem-solving, programming concepts, and digital citizenship. Plugged activities required a computer, whereas unplugged activities allowed students to learn CS through physical activities, games, and puzzles (see Photo 1).

Graph: Photo 1.Author, 2020. Unplugged coding activity for grade 1 students. Unpublished.

The goal was to foster a school-wide culture of computing by ensuring all stakeholders were computer literate. Following the PL, each classroom teacher was tasked with implementing six CT lessons in their classrooms. A follow-up PL session was planned for Fall 2020 to reinforce and expand upon the initial learning.

Participants

Participants were 48 educators from a rural primary school in Appalachia, serving 750 students in grades K–2. The cohort consisted of 34 classroom teachers, five teacher's aides, seven support professionals, one school-level administrator, and one state level administrator. Most participants identified as female, as reflected in previous studies (e.g., [7]; [38]; [37]; [54]). Teaching experience ranged between 1 and 21 years.

Data Sources

Data sources included a modified version of the Teaching Efficacy Beliefs Survey ([15]), researcher observation notes, and a focus group interview. The Teaching Efficacy Beliefs Survey assessed participants' self-efficacy for teaching coding and CS topics. To determine Self-Efficacy for Teaching Coding, respondents indicated their level of agreement with five statements, such as, "I am confident in my ability to help my students value learning how to code." The coefficient alpha values for the pre- and post-test scores were.99 and.97, respectively. To determine Self-Efficacy for Teaching Computer Science Topics, respondents indicated their level agreement with three statements, including, "I am confident in my ability to motivate students who show low interest in computer science." The coefficient alpha values for the pre- and post-test scores were.98 and.99, respectively. All scales were on a 1–7 scale, with 1 indicating strongly disagree and 7 indicating strongly agree.

Data Collection

The survey was administered at three time points: before the initial PL, after the initial PL, and after classroom implementation. The goal was to measure changes in teachers' self-efficacy in teaching CT, with higher scores indicating better instructional outcomes. Of the 48 participants, 48 completed the first survey, 38 completed the second, and 29 completed the third.

Two researchers observed the implementation of Code.org lessons in 13 classrooms, taking detailed notes. Additionally, a focus group interview with 11 classroom teachers was conducted to gather insights for future interventions.

Data Analysis

We used paired sample t-tests to analyze the quantitative survey data. We used inductive thematic analysis to analyze qualitative data ([5]), including researcher notes and focus group interviews.

Findings

In Year 1, the Introductory CS Fundamentals workshop statistically significantly increased the self-efficacy of elementary school educators (n = 48) in teaching computer coding (d = 1.20) and computer science topics (d = 1.12).

Classroom observations conducted one month after the workshop indicated positive effects of the PL on teachers' classroom practices. Teachers actively engaged students in coding activities, reinforced coding vocabulary, and provided positive academic feedback to encourage students' emerging skills (C. Brigandi, Researcher Notes, February 28, 2020). Teachers challenged students to think critically, recognize errors, and explore multiple solutions to the same problem, effectively addressing the needs of students with high coding abilities by using pedagogies and questioning strategies that promote higher-order thinking (C. Brigandi, Researcher Notes, February 28, 2020). Teachers described students as "engaged," "interested," and "learning" (C. Brigandi, Focus Group Interview, February 28, 2020).

Due to a shift in priorities in response to the onset of the COVID-19 pandemic, end-of-year survey data captured only how many teachers implemented Code.org lessons. Data indicated 21 teachers implemented 57 lessons.

Limitations

The study encountered significant limitations due to the early closure of schools in the spring of 2020 because of the COVID-19 pandemic. This resulted in the cancellation of a planned family night of code in Year 1, as well as alterations to Year 1 data collection and planned interventions in Year 2.

Our Efforts during Pandemic-Related School Closures: Professional Learning in Year 2 (August...

In the second year, amid the COVID-19 pandemic and the shift to virtual schooling, educators at the pilot school were offered a series of virtual PL opportunities provided by the researchers. These included a 90-min refresher course on CS Fundamentals, a one-hour introduction to the Hour of Code online resources, and a one-hour training on integrating the Code.org Learning Management System (LMS) with their school's existing LMS. During the pandemic, the focus was shifted from acquiring new knowledge to leveraging online activities that would make coding accessible for students.

Participants

A total of 44 educators attended the CT refresher course and the Hour of Code training. Nineteen first- and second-grade classroom teachers participated in the session on linking LMSs.

Data Sources and Collection

We conducted surveys three times: after the refresher course, after the Hour of Code training, and at the end of the school year. Given the disruption caused by COVID-19, we modified our approach to focus on supporting teachers, rather than strictly adhering to our original research priorities. As a result, each survey focused on different aspects:

• Survey 1: Assessed teachers' self-efficacy for teaching coding (α =.98) and computer science (α =.98) using the same survey instrument from Year 1.

• Survey 2: Evaluated teachers' confidence in differentiating instruction (see results in [7]).

• Survey 3: Focused on which Code.org lessons teachers were able to implement in hybrid learning environments.

Survey response rates were as follows: 37 out of 49 for Survey 1, 43 out of 48 for Survey 2, and 30 out of 49 for Survey 3. We were unable to conduct classroom observations and focus group interviews in Year 2 due to pandemic restrictions and shifting district priorities.

Data Analysis

Because the surveys were not directly comparable, we present the data descriptively.

Findings

At the beginning of Year 2, teachers indicated above average self-efficacy in teaching coding (M = 5.51, SD = 1.22) and in teaching computer science concepts (M = 5.47, SD = 1.11). All scales were on a 1–7 scale, with 1 indicating strongly disagree and 7 indicating strongly agree.

By the end of Year 2, 18 classroom teachers indicated having taught a combined total of 33 CS lessons, with 12 teachers implementing at least some lessons, while six implemented no lessons.

Limitations

The COVID-19 pandemic significantly altered the learning environment and limited our ability to collect comprehensive data. As a result, we were unable to fully address our original research questions in Year 2.

Our Efforts in Wake of the Pandemic: Professional Learning in Year 3 (August 2021 – May 2022)

In the third year, with the return of face-to-face schooling post-pandemic, returning educators (n = 25) were offered a 4-hour Code.org Deep Dive workshop that had been originally scheduled for August 2020. Newly hired teachers (n = 10) attended a 4-hour CS Fundamentals workshop. Both sessions were led by Code.org facilitators. Twenty-five teachers also participated in grade-level PLC meetings led by the research team, and one teacher leader attended an advanced week-long summer CS training. Additional interventions included Hour of Code and Dr Seuss Night with Coding, which were family-friendly evening events. Teachers were again asked to implement six CT lessons with their students.

Data Sources

Data sources included the Teaching Efficacy Beliefs Instrument ([15]), researcher classroom observation notes, and a focus group interview. The coefficient alpha values for self-efficacy for teaching coding pre- and post-PL for Deep Dive respondents were.98 and.95, respectively. The coefficient alpha values for self-efficacy for teaching CS concepts were.93 and.98, respectively.

Data Collection

Of the 25 Deep Dive participants, 15 completed both the pre- and post-survey. However, only two of the ten participants in the Introduction to CS Fundamentals workshop completed the post-survey, rendering it insufficient for meaningful analysis. Classroom observations took place in Fall 2021 across nine classrooms serving 144 students, with two researchers observing.

Data Analysis

We used paired sample t-tests to analyze the quantitative survey data. We used inductive thematic analysis to analyze qualitative data ([5]), including researcher notes and focus group interviews.

Findings

In Fall 2021, observations revealed that although teachers engaged students in coding activities, the implementation was often superficial. Teachers generally did not encourage deeper thinking, critical assessment, or meaningful connections, leading to a lack of challenge and resulting in boredom for some high-ability students (C. Brigandi, Researcher Notes, September 28, 2021). This was a missed opportunity to cultivate higher-level thinking skills, which had been a key objective for the grant team.

In response, we modified the Year 3 PL to include grade-level PLCs provided by the research team. During the PLCs, we showed teachers how to navigate the code platform and access features such as hints, audio, and "for teachers only." We demonstrated how to teach tough concepts, like creating loops with multiple repeat functions. We modeled quality pedagogy by giving teachers time to play, think, and code ([24]) and encouraging them to find multiple solutions to each problem rather than rushing through the puzzles. We incorporated critical thinking strategies to encourage higher-order thinking. Teachers compared the block code to the hidden JavaScript, observed the workspace code, predicted outcomes, and debugged the code. For example, in Code.org Course B Lesson 5: Programming with Harvester, first grade students assist the harvester in gathering corn in a maze. Students are given an algorithm—a step-by step sequence of coding commands—and instructed to "add one block [of code] to help the harvester gather the corn." Before solving the puzzle, teachers can encourage both computational and critical thinking by asking students to:

• Observe the puzzle. What do you see and what needs to happen?

• Make a Plan for the harvester to gather the corn.

• Observe the code. Can you spot any Bugs in the code?

• Predict what will happen when you press "run." Will the harvester gather all the corn?

• Compare & Contrast the block-based code with JavaScript. How are they the same? How are they different? When and why is it better to use a block code? When and why is it better to use JavaScript?

• Do you see a Pattern in the code? Describe the pattern.

• Debug the puzzle.

• Write an Algorithm that allows the harvester to gather all the corn.

• Write an Algorithm that allows the harvester to gather all the corn using the repeat block.

• Play, Think, & Code. Who can get the harvester to gather the corn using the most steps? Who can get Harvester to gather the corn using the fewest steps? When is it better to use many steps? When is it better to use fewer steps?

Due to the small group format and individualized instruction from the research team and PL providers, teachers reported that PLCs gave them more time to engage with the coding curriculum, collaboration time with grade-level team members, awareness of coding resources, and meaningful curriculum connections (C. Brigandi, Focus Group Interview, October 13, 2022). Teachers expressed that the grade-level PLCs provided a supportive community for sharing ideas and best practices, leading to a more collaborative teaching environment. Teachers thus saw PLCs as an important supplement reinforcing the Code.org large group trainings.

The January 2022 Deep Dive workshop, designed to reinforce coding skills among returning teachers, resulted in statistically significant changes in teachers' confidence in coding and computer science [d = 1.20 and d = 0.93, respectively].

By the end of Year 3, seven teachers reported having implemented a total of 27 coding lessons. Of the four teachers that completed all surveys, there were no statistically significant differences between their pre-survey and end of semester means for self-efficacy in teaching coding (M = 0.20, p =.42) or self-efficacy for teaching CS (M = 0.5, p =.39). Similarly, no significant differences were found between their post-workshop and end-of-year means (self-efficacy code difference: M = −0.40, p =.33, self-efficacy CS difference: M = −0.67, p =.25).

Limitations

Teacher attrition was a limitation, as was the delay of the Deep Dive workshop from August 2021 until January 2022 due to challenges faced by school administrators in onboarding new teachers amid pandemic-related challenges. Perceptions of CS relevancy waned as districts prioritized math and English language arts (ELA), and gaps in schooling led to gaps in data collection affecting our findings. For example, we were unable to conduct a Focus Group Interview in Year 3.

Our Revised Efforts: Professional Learning in Year 4 (August 22 – May 2023)

In year 4, we expanded our initiative to include a second school serving students in grades 3–5, building on insights gained from the first three years. Motivated by the success of PLCs in Year 3, we redesigned the original Code.org PL plan to be more comprehensive, sustained, ongoing, and collaborative ([17]) by incorporating four key PL components:

(1) Workshops: We offered two workshops—Introductory Code.org CS Fundamentals and the Deep Dive Workshop—to provide foundational knowledge and advanced exploration of the curriculum.

(2) Grade-Level PLC Sessions: These one-hour sessions allowed teachers to navigate the code platform, explore features, and tackle complex concepts like loops with multiple repeat functions.

(3) Demonstration Videos: A library of 30 researcher-created, 10-min videos offered targeted support for teachers on specific coding challenges.

(4) Teacher Leader: A school-level teacher leader, trained through multi-day workshops, provided ongoing peer support.

Grade-level PLCs: Standards-based Cross-Curricular Coding Activities

In the Year 4 PLCs, we focused on enhancing teachers' skills by demonstrating LMS features, reinforcing CS and GT content knowledge, and modeling CT pedagogies. To sustain the PL and address the challenge of infrequent visits due to the 6-hour round trip to our rural schools, we also utilized a researcher-created video library and school-based teacher leaders. Additionally, we worked to increase the relevance of coding instruction by showing how it could be aligned with state math and ELA standards (see Appendix B).

Whereas all Code.org lessons align with the Computer Science Teachers Association (CSTA) K–12 Computer Science Standards, some lessons also include optional cross-curricular activities aligned with Common Core State Standards (CCSS) for mathematics and ELA and Next Generation Science Standards (NGSS). The Corn Crazy Choice Board Activity in Course B Lesson 7: Loops with Harvester, for example, integrates biomimicry and communication ([9]), addressing standards such as:

• CCSS.ELA-LITERACY.SL.1.1.A: Follow agreed-upon rules for discussions (e.g., listening to others with care, speaking one at a time about the topics and texts under discussion).

• CCSS.ELA-LITERACY.SL.1.1.B: Build on others' talk in conversations by responding to the comments of others through multiple exchanges.

• CCSS.ELA-LITERACY.SL.1.1.C: Ask questions to clear up any confusion about the topics and texts under discussion.

• NGSS.1-LS1-1: Use materials to design a solution to a human problem by mimicking how plants and/or animals use their external parts to help them survive, grow, and meet their needs.

In the optional Code.org Treasure Tracker activity included in Course B Lesson 8: Loops with Laurel, students are tasked with helping Laurel track her treasure as she counts and categorizes her loot. Cross-curricular standards addressed include:

• CCSS.ELA-LITERACY.L.1.5.A: Sort words into categories (e.g., colors, clothing) to gain a sense of the concepts the categories represent.

• CCSS.MATH.CONTENT.1.OA.A.1: Use addition and subtraction within 20 to solve word problems involving situations of adding to, taking from, putting together, taking apart, and comparing, with unknowns in all positions, for example, by using objects, drawings, and equations with a symbol for the unknown number to represent the problem.

• CCSS.MATH.CONTENT.1.OA.A.2: Solve word problems that call for addition of three whole numbers whose sum is less than or equal to 20, for example, by using objects, drawings, and equations with a symbol for the unknown number to represent the problem.

Despite completing the CS Fundamentals and Deep Dive workshops, many teachers in our study were unaware of these optional cross-curricular activities, although this may be a situational finding. Teachers at the pilot school also received specialized training in critical thinking to differentiate instruction for high-ability learners. Both schools benefited from after-school enrichment clusters and a Family Night of Code.

Participants

Pilot School

Twenty-two classroom teachers in grades K-2 and one pre-service intern attended a 4-hour PL session on critical thinking. Sixteen of these teachers also participated in a subsequent one-hour grade-level PLC session. Participants were mostly female (78% female, 9% male, 9% missing, 2% other).

Scale-Up School

Forty-eight educators in grades 3–5, including 25 classroom teachers, attended a 4-hour Code.org CS Fundamentals session, with 35 also attending a 4-hour follow-up Code.org Deep Dive workshop. Thirteen of the 25 classroom teachers participated in an interactive grade-level PLC session. Participants were mostly female (76% female, 6% male, 17% missing).

Data Sources and Collection

We employed a mix of quantitative and qualitative methods to gather data, including:

• A modified version of TOP-Science Protocol ([35]): Captured how teachers use technology to support learning, with inter-rater reliability (IRR) of 0.76 for observation one and 0.86 for observation two.

• TBaCCT Survey ([41]): Assessed teachers' value beliefs, coding self-efficacy, and teaching efficacy related to CT. The instrument has shown strong validity (e.g., CFI >.90, factor loadings > |.40|; [41]). Reliability across subscales in our study ranged from α = 0.68 to 0.99.

Survey and observation data were collected at multiple points throughout the school year, supplemented by a focus group interview and analysis of teacher support.

Data Analysis

An introductory workshop was conducted at the pilot school during the 2021–2022 academic year. Consistent with established guidelines in the literature, we expanded the PL program for the 2022-2023 school year by incorporating a deep-dive workshop, PLCs, videos, and teacher leaders' support. Additionally, we employed the TBaCCT as a tool to assess teachers' self-efficacy. The pre- and post-surveys that the pilot school teachers completed during the 2021–2022 academic year were not included in this study since they were different. When an analysis involved data collected from both schools, we used regression. When an analysis used data from only the scale-up school, we used a paired sample t test. We used inductive thematic analysis to analyze qualitative data ([5]).

Findings

Following the introductory workshop, teachers showed statistically significant improvements in value beliefs about coding (p =.02, d =.57), coding self-efficacy (p <.001, d = 1.72) and CT teaching self-efficacy (p <.001, d = 1.07). The Deep Dive workshop further increased coding self-efficacy (p <.001, d = 1.07) and CT teaching efficacy (p <.001, d = 1.05). Additionally, although teachers' self-efficacy in teaching CT improved after each training, this improvement wasn't sustained through the end of the year, signaling a potential need for continued support (see [54] for more detail).

Professional Learning

To examine how the PL influenced teachers' self-efficacy in coding and teaching CT, we asked teachers to rate the helpfulness of each PL component on a scale of zero–3. The mean helpfulness ratings ranged between 2.24 and 2.56, including the introductory workshop (M = 2.44), deep-dive workshop (M = 2.24), PLCs (M = 2.31), videos (M = 2.25), and teacher leader support (M = 2.56) ([54]). Focus group comments reflected this, with teachers emphasizing the importance of the resources, PLCs, and teacher leader support ([54]).

Classroom Practice

Researcher observation using the Technology Observation Protocol for Science (TOP-Science; [35]) revealed an improvement in teachers' classroom instruction. Specifically, teachers' use of computational concepts, practices, and pedagogies increased from 0.6 (out of 3), 1.0, and 1.3 at the beginning of year 4 to 1.6, 2.0, and 1.7 at the end of year 4. In surveys and focus group interviews, teachers noted teaching looping, pattern recognition, debugging, and using problem-solving strategies.

Student Engagement

In Year 4, 550 students in grades K–5 received CS instruction in their regular classrooms, while 1060 students participated in STEM labs. Optional activities, such as Enrichment Clusters and the Hour of Code, further sparked interest, with 58 students participating in after-school coding clusters and 48 people attending the Hour of Code event.

Discussion of Findings

This research provides valuable insights into the relationship between PL and teachers' self-efficacy in teaching CT, classroom practices, and their ability to address the needs of gifted students. Several key findings emerged from the analysis of PL's impact on CT and gifted education.

Increased Teacher Confidence and Self-Efficacy

Over multiple years, PL workshops consistently improved teachers' self-efficacy in coding and teaching CT and CS, especially when supplemented by regular PLC sessions. These extended engagements allowed teachers to meaningfully integrate coding into their classrooms. In contrast, shorter or less frequent PL sessions were less effective in sustaining these gains (e.g., [12]; [20]). Extending the PL duration and aligning coding lessons with local standards in math and ELA could improve the curriculum's relevance and long-term impact.

Influence on Classroom Practices

The PL interventions led to observable changes in classroom practices. In the first year, teachers actively engaged students in coding activities, utilizing strategies like critical thinking and higher-order questioning that promoted intellectual challenge, particularly benefiting gifted students. However, by Year 3, the coding activities in the classroom became superficial, failing to sufficiently challenge high-ability students. This decline suggests that the positive effects of PL on classroom practices diminish over time. Modifications to the PL program that emphasized critical thinking and problem-solving helped address these gaps, encouraging deeper exploration of coding concepts. Coaching teachers to use formative assessments and critical thinking strategies could further enhance classroom practices, particularly for high-ability students.

PLCs and Collaborative Learning

Teachers highlighted the value of PLCs, where researchers modeled effective CS instruction using best practices and pedagogies. These PLCs provided valuable time for teachers to deeply engage with coding materials, actively participate in learning, and share ideas. By fostering sustained engagement and building supportive communities, PLCs enhanced lesson planning efficiency, increased awareness of available coding resources, and aligned instruction with local standards, reinforcing the relevance of coding education. Teachers also appreciated the additional instructional videos and guidance from teacher leaders. These findings underscore the need for PL to be relevant, comprehensive, sustained, and collaborative (e.g., [12]; [50]).

Limitations

COVID-19 Pandemic

The pandemic disrupted planned interventions, making it challenging to measure the full impact of PL on teaching practices and student outcomes. Interruptions to schooling, staff shortages, teacher turnover, high rates of absenteeism, mental health challenges, and associated declines in learning during the pandemic all influence findings. These challenges disproportionately affected low and high poverty schools, like those in our study, with students in the poorest districts falling behind, worsening pre-existing educational and socioeconomic inequalities ([16]; [25]).

Teacher Attrition

High turnover rates, particularly in Year 3, affected the continuity and effectiveness of the PL interventions. This instability limited the program's ability to produce long-term, consistent improvements.

Demographic Data

The absence of demographic data on race is a limitation of this study. Although the population of West Virginia is predominantly White, without specific data on racial and ethnic groups, it is difficult to assess whether the findings are generalizable across diverse populations. This lack of data limits the ability to examine potential variations in experiences and outcomes based on racial identity.

Cross-Temporal Comparisons

Comparisons across years were complicated by the varying contexts, including disruptions caused by the pandemic and the introduction of new instructional tools. These factors made it difficult to isolate the impact of PL interventions from external variables that also influenced teaching and learning during this time.

Future Directions and Implications

Providing interdisciplinary and multifaceted PL opportunities that blend CS instruction with strategies for gifted education holds promise for supporting both teachers and students. Schools can better address the needs of high-ability learners by offering sustained, differentiated PL that integrates coding, critical thinking, and formative assessment strategies. Further investigation into the role of local teacher leaders, instructional videos, and their relationship to teacher self-efficacy and classroom practice could provide valuable insights into improving the effectiveness of future PL efforts. Future research should aim to collect and analyze demographic data to ensure more comprehensive and inclusive conclusions.

Conclusion

By tailoring CS instruction to high achieving students and educating teachers to properly support these students, we can better encourage students to think critically and engage with complex concepts. Additionally, PL needs to be collaborative, sustained, and ongoing, as well as meaningful to participants. Teachers, like students, need time to explore and engage with content, or to play, to think, to code. This multifaceted, interdisciplinary approach to PL can bridge gaps in educational opportunities, helping teachers to support all students, particularly those with high academic potential.

ORCID iDs

Carla B. Brigandi https://orcid.org/0000-0002-1152-872X

Karen E. Rambo-Hernandez https://orcid.org/0000-0001-8107-2898

Maryann R. Hebda https://orcid.org/0000-0003-0600-235X

Appendix

Graph: Appendix A. Timeline of professional learning.

Appendix B. PL Components Aligned With PL, CT, and GT Best Practices.

Graph

<table><thead valign="top"><tr><th align="left">PL Component</th><th align="center">PL Best Practices</th><th align="center">CT Best Practices</th><th align="center">GT Best Practices: NAGC Teacher Standards</th><th align="center">GT Best Practices: NAGC PK-12 Standards</th></tr></thead><tbody valign="top"><tr><td align="left"><p>Code.org CS fundamentals workshop</p></td><td align="left"><p>&#8226; Effective PL should focus on the content that teachers need to teach and the pedagogy that can be used to support student learning and provide opportunities for teachers' active participation (<xref ref-type="bibr" rid="bibr10">Darling-Hammond et al., 2017</xref>; <xref ref-type="bibr" rid="bibr11">Desimone, 2009</xref>).</p></td><td align="left"><p>&#8226; Give teachers time <italic>to play, to think, to code</italic> (<xref ref-type="bibr" rid="bibr24">Kong et al., 2020</xref>).</p></td><td align="left"><p>&#8226; 5.2 Beginning gifted education professionals apply appropriate technologies to support instructional assessment, planning, and delivery for individuals with gifts and talents.<break />&#8226; 4.4 Beginning gifted education professionals use assessment results to develop long- and short-range goals and objectives that take into consideration an individual's abilities and needs, the learning environment, and other factors related to diversity.</p></td><td align="left"><p>&#8226; 6.4.1. Educators regularly reflect on and assess their instructional practices, develop professional learning plans, and improve their practices by participating in continuing education opportunities.<break />&#8226; 3.1.7. Educators integrate a variety of technologies for students to construct knowledge, solve problems, communicate and express themselves creatively, and collaborate with others in teams locally and globally<break />&#8226; 5.1.5. Educators leverage technology to increase access to high-level programming by providing digital learning options and assistive technologies.</p></td></tr><tr><td align="left"><p>Code.org Deep-Dive workshop</p></td><td align="left"><p>&#8226; Effective PL should focus on the content that teachers need to teach and the pedagogy that can be used to support student learning, provide opportunities for teachers' active participation, and align with school priorities (<xref ref-type="bibr" rid="bibr10">Darling-Hammond et al., 2017</xref>; <xref ref-type="bibr" rid="bibr11">Desimone, 2009</xref>).</p></td><td align="left"><p>&#8226; The cognitive CT framing that posits students need to develop an understanding of CT concepts and practices (<xref ref-type="bibr" rid="bibr22">Kafai et al., 2019</xref>).<break />&#8226; Key CT concepts like sequences, loops, events, conditionals, and data (<xref ref-type="bibr" rid="bibr6">Brennan & Resnick, 2012</xref>). Key CT practices like decomposition, pattern recognition, algorithms, and debugging (<xref ref-type="bibr" rid="bibr6">Brennan & Resnick, 2012</xref>; <xref ref-type="bibr" rid="bibr19">Jocius et al., 2020</xref>; <xref ref-type="bibr" rid="bibr40">Relkin et al., 2020</xref>).<break />&#8226; Effective CT pedagogical strategies like the PRIMM approach (<xref ref-type="bibr" rid="bibr43">Sentance et al., 2019</xref>), the problem-solving approach (<xref ref-type="bibr" rid="bibr44">Shute et al., 2017</xref>), and the pedagogy of <italic>to play, to think, to code</italic> (<xref ref-type="bibr" rid="bibr24">Kong et al., 2020</xref>).</p></td><td align="left"><p>&#8226; 5.3 Beginning gifted education professionals collaborate with families, professional colleagues, and other educators to select, adapt, and use evidence-based strategies that promote challenging learning opportunities in general and specialized curricula.</p></td><td align="left"><p>&#8226; 3.1.1. Educators use local, state, and national content and technology standards to align, expand, enrich, and/or accelerate curriculum and instructional plans.<break />&#8226; 3.4.1. Educators select, adapt, and use a repertoire of instructional strategies to differentiate instruction for students with gifts and talents.<break />&#8226; 5.2.3. Educators plan coordinated learning activities within and across a specific grade level, content area, course, class, and/or programming option.</p></td></tr><tr><td align="left"><p>Grade-level professional learning communities</p></td><td align="left"><p>&#8226; Effective PL should be of sustained duration and encourage participation of teachers from the same school or grade level (<xref ref-type="bibr" rid="bibr10">Darling-Hammond et al., 2017</xref>; <xref ref-type="bibr" rid="bibr11">Desimone, 2009</xref>).</p></td><td align="left" /><td align="left"><p>&#8226; 3.1 Beginning gifted education professionals understand the role of central concepts, structures of the discipline, and tools of inquiry of the content areas they teach, and use their understanding to organize knowledge, integrate cross-disciplinary skills, and develop meaningful learning progressions within and across grade levels.<break />&#8226; 3.4 Beginning gifted education professionals understand that individuals with gifts and talents demonstrate a wide range of advanced knowledge and performance levels and modify the general or specialized curriculum appropriately.</p></td><td align="left"><p>&#8226; 3.6.1. Educators use current, evidence-based curricular resources that are effective with students with gifts and talents.</p></td></tr><tr><td align="left"><p>Researcher- created video library</p></td><td align="left"><p>&#8226; Effective PL should focus on the content that teachers need to teach and the pedagogy that can be used to support student learning and be of sustained duration (<xref ref-type="bibr" rid="bibr10">Darling-Hammond et al., 2017</xref>; <xref ref-type="bibr" rid="bibr11">Desimone, 2009</xref>).</p></td><td align="left"><p>&#8226; Key CT concepts include sequences, loops, events, conditionals, and data (<xref ref-type="bibr" rid="bibr6">Brennan & Resnick, 2012</xref>).<break />&#8226; Key CT practices include decomposition, pattern recognition, algorithms, and debugging (<xref ref-type="bibr" rid="bibr6">Brennan & Resnick, 2012</xref>; <xref ref-type="bibr" rid="bibr19">Jocius et al., 2020</xref>; <xref ref-type="bibr" rid="bibr40">Relkin et al., 2020</xref>).<break />&#8226; Effective CT pedagogical strategies include the PRIMM approach (<xref ref-type="bibr" rid="bibr43">Sentance et al., 2019</xref>), the problem-solving approach (<xref ref-type="bibr" rid="bibr44">Shute et al., 2017</xref>), and the pedagogy of <italic>to play, to think, to code</italic> (<xref ref-type="bibr" rid="bibr24">Kong et al., 2020</xref>).</p></td><td align="left"><p>&#8226; 5.1 Beginning gifted education professionals know principles of evidence-based, differentiated, and accelerated practices and possess a repertoire of instructional strategies to enhance the critical and creative thinking, problem-solving, and performance skills of individuals with gifts and talents.</p></td><td align="left"><p>&#8226; 3.1.4. Educators design differentiated curriculum that incorporates advanced, conceptually challenging, in-depth, and complex content for students with gifts and talents.<break />&#8226; 3.4.3. Educators use models of inquiry to engage students in critical thinking, creative thinking, and problem-solving strategies, particularly in their domain(s) of talent, both to reveal and address the needs of students with gifts and talents.</p></td></tr><tr><td align="left"><p>School-level teacher leader</p></td><td align="left"><p>&#8226; Effective PL should provide follow-up support (<xref ref-type="bibr" rid="bibr11">Desimone, 2009</xref>).</p></td><td align="left" /><td align="left"><p>&#8226; 7.2 Beginning gifted education professionals serve as a collaborative resource to colleagues.</p></td><td align="left"><p>&#8226; 6.4.2. Educators participate in professional learning that is sustained over time, incorporates collaboration and reflection, is goal-aligned and data-driven, is coherent, embedded and transferable, includes regular follow-up, and seeks evidence of positive impact on teacher practice and on increased student learning.</p></td></tr></tbody></table>

2 Note. GT best practices are from https://nagc.org/page/National-Standards-in-Gifted-and-Talented-Education

By Carla B. Brigandi; Karen E. Rambo-Hernandez; Jiangmei Yuan; Maryann R. Hebda; Robin Spitznogle; Remeri Mestemacher-McClanahan and Sarah Wilson

Reported by Author; Author; Author; Author; Author; Author; Author

Carla B. Brigandi, PhD, is an associate professor at West Virginia University in the School of Education where she teaches graduate and undergraduate courses in Educational Psychology, Research Methods, and Special Education. Dr. Brigandi's scholarship is focused on improving educational opportunities for students who have high academic ability, particularly those from traditionally underserved populations. Her related interests include evidence-based enrichment practices and teacher professional learning.

Karen E. Rambo-Hernandez, PhD, is an associate professor at Texas A&M University in the College of Education and Human Development. Her research has been funded by National Science Foundation, the U.S. Department of Education, and others and focuses on the assessment of educational interventions to improve STEM education, and access for all students—particularly high achieving and underrepresented students—to high quality education.

Jiangmei Yuan, PhD, is an associate professor in the College of Education at Boise State University. She will become an associate professor in the College of Education at Boise State University, beginning in August 2025. She earned her PhD in Learning, Design, and Technology from the University of Georgia. Her research focuses on enhancing pre- and in-service teachers' learning and teaching of computational thinking and artificial intelligence, improving undergraduate engineering students' engagement and academic achievement in foundational courses, and designing technology-enhanced learning experiences. Her publications have appeared in multiple leading venues, including Computers & Education, Educational Technology Research and Development, Instructional Science, and Journal of Computer Assisted Learning.

Maryann R. Hebda, PhD, is a research specialist in Teaching, Learning, and Culture at Texas A&M University. She received her PhD in Educational Psychology from Baylor University. She received her MS in Special Education-Gifted Emphasis from Emporia State University and BS in Elementary and Special Education from Nebraska Wesleyan University. Her research focuses on STEM talent development and achievement motivation in gifted and twice-exceptional populations.

Robin Spitznogle, MA, is a doctoral student in the Learning Sciences and Human Development Program at West Virginia University and a graduate research assistant on Project Appalachian Coders, a U.S. Department of Education Grant. Her research interests include cognitive, metacognitive, and social-psychological processes that facilitate deeper, more meaningful learning and student persistence in first-year, at-risk, and marginalized racial and ethnic student groups underrepresented in college.

Remeri Mestemacher-McClanahan, BS, is currently a post-baccalaureate researcher at West Virginia University where they are working on a National Science Foundation-funded project investigating the synaptic connections and projection patterns of thalamocortical axons in the barrel cortex of the mouse brain. Their research interests span learning and development, neurodegenerative diseases, and sensory processing. Previously, they worked as a research assistant on Project Appalachian Coders, a U.S. Department of Education Grant Project. They plan to pursue a PhD in neuroscience.

Sarah Wilson is an undergraduate student at West Virginia University majoring in Psychology with a minor in Child Development and Family Studies. Her research interests include computational thinking and literacy within early education settings. She is a research assistant on Project Appalachian Coders, a U.S. Department of Education Grant. Her accolades include being a Gilman Scholar, the valedictorian of her graduating class, and a prize-winner for the West Virginia High School Business Plan Competition.