As CS is moving into K-12 education, most PD, curriculum, and research efforts have focused on teaching and learning CS content: concepts to understand principles and processes of computers and software (Seehorn et al., 2011). Yet injustices amplified by computing applications (e.g., Benjamin, 2019) demand that equal attention be paid to computing's "impact on society" (Seehorn et al., 2011, p. 1). This call has initiated a critical turn in K-12 CS education (Kafai & Proctor, 2022;Ko et al., 2022) that advocates for moving beyond the often-adopted valuefree stance on computing. Through pedagogical frameworks, researchers recently have proposed similar critical shifts in K-12 curriculum (Kapor Center, 2021;Madkins et al., 2020). Integrating critical stance with technical is important when racist logics are embedded in computing abstractions and algorithms in computing tools often employed to 'solve' societal issues around policing and justice (e.g., Benjamin, 2019).
While many studies concerning K-12 student learning have added discussion units or whole courses to examine computing critically (e.g., Vakil, 2018), far fewer have focused on teacher preparation to teach social justice issues within computing classrooms. Most current efforts in computing teacher preparation concentrate on inducting existing teachers, focusing on building teachers' technical knowledge (Menekse, 2015). A few recent studies highlight the challenges teachers face as they integrate computing with societal issues such as race and gender (Everson et al., 2022) and how White teachers may evade and deflect conversations around race within teacher PD (Goode et al., 2020). These findings suggest the need to better understand how computing teachers can learn to engage with and integrate critical ideas while learning to teach computing.
In this paper, we turn our attention to teachers (rather than students) as learners of critical computing. We analyzed one PD session where experienced high school computing teachers learned to teach an electronic textiles (e-textiles) unit. This session, with six teachers, focused on integrating technical aspects of designing sensor-based physical computing artifacts with societal concerns, such as racism, that are embedded within inadequately designed sensor-based devices. Inspired by interaction analysis (Erickson et al., 2017;Jordan & Henderson, 1995) and sociocultural theories of teacher learning (Vygotsky, 1987), we conducted a collaborative video analysis of the teacher PD session to answer the research question: how do teachers learn technical and societal aspects of computing while participating in an electronic textiles PD?
The design of the PD and the ensuing data analysis were informed by sociocultural theories of teacher learning, PDs as sites for teacher learning, and the teacher learning of critical aspects within computing.
A sociocultural perspective of teacher learning highlights how situated, social and cultural aspects shape professional identity development (Fishman et al., 2014;Shulman, 1987). Shulman (1987) viewed teaching to be viewed as a practice and teachers as a part of communities of practice with shared understandings of teaching. He further framed teacher learning as developing fluency with practices of teacher communities, in interactions with one another and in a distributed fashion with expertise distributed across teachers (Fishman et al., 2014). These theories posit that teachers do not learn discrete, disconnected facts or skills but learn concepts as they relate to their teaching contexts, classroom dynamics, and personal experiences (Bukor, 2014;Enyedy et al., 2006).
Classrooms and teacher PDs are two prominent sites of teacher learning studied within learning sciences (Fishman et al., 2014). Studies that have explored teacher learning in classroom settings have examined teachers' interactions with both resources (such as curricular materials) and students in shaping their practice (e.g., Sherin & Han, 2004). Similar efforts involving PDs have either focused on different facets of teacher knowledge (e.g., Shulman, 1987) or on the social nature of teacher learning and the development of communities of practice (Fishman et al., 2014). Though attention to processes of teacher learning during PDs is important, scarce research details processes of how teachers learn to critically engage with disciplines or ask questions about the relationship between disciplinary knowledge and its interactions with communities and societies. This gap is especially significant in the case of in-service, experienced teachers who rely on PDs for professional learning opportunities to grow their ability to teach critical computing (Fishman et al., 2015;Goode et al., 2020). While studies related to student learning have highlighted the importance of, for instance, politicized trust-a race-conscious way of understanding, respecting, and being in solidarity with one another-in shaping learning of critical aspects (Vakil & de Royston, 2019), similar examinations of learning processes within teacher PDs are lacking.
Previous research within computing education has highlighted key PD aspects that shape teacher learning: duration of the PD, connections to classroom practice, focus on learning methods and pedagogical content knowledge, and relationships within school districts (Menekse, 2015). This aligns with Fishman and colleagues' (2014) review of teacher learning within learning sciences that emphasizes extended durations, opportunities for engagement with content knowledge, practice-related aspects such as student learning, and reflection. But most research on computing teacher learning relies on teacher surveys and interviews as methods, revealing very little about the processes of computing teacher learning (Menekse, 2015;Yadav et al., 2016). Further, prior studies within computing teacher PDs focus primarily on disciplinary knowledge as a collection of technical aspects (Menekse, 2015), barely interrogating how teachers learn to expand the disciplinary boundaries and integrate critical societal concerns with technical aspects. An exception is Goode and colleagues' (2020) examination of teacher learning of critical issues in computing through an analysis of teacher discussions and interactions within PDs. With increased computing tools around us and heightened implications for marginalized communities (e.g., Benjamin, 2019), teacher preparation efforts should support teachers in critically engaging with societal impacts of computing. Recent equity-centered computing pedagogical frameworks underscore the need to "situate technology ideas within their sociopolitical context and give students opportunities to critique and explore issues that are relevant to them" (Madkins et al., 2020, p. 13). Buttressing this call, Goode and colleagues' (2020) study surfaces the need to support experienced high school computing teachers in engaging with societal aspects such as race during PDs. While limited prior studies highlight the struggles of teachers as they integrate technical and societal aspects (e.g., Everson et al., 2022;Goode et al., 2020), there is a need to further understand how to support teachers in critically analyzing the connections of societal and technical dimensions of computing.
Mathematics and science education have explored how to engage teachers critically (Bianchini et al., 2015), laying a path for examining learning of critical computing. Questioning the relationships between computing, people, communities, and societies can support students who belong to historically excluded groups find their voices, engage with the discipline in personally meaningful ways, and contribute agentically to the discipline (Goode et al., 2020;Vakil, 2018). With concerns of teachers evading or deflecting race-related issues or blaming individual students for failures that have roots in historical and systemic racial oppression (Segall & Garrett, 2013), teachers should engage with societal issues such as race and racism, gender, and ability in computing.
While prior studies have proposed equity-centered pedagogies (Madkins et al., 2020) and critical pedagogical frameworks in computing (Kapor Center, 2021), we know very little about how teachers learn or develop critical perspectives at the intersection of technical and societal aspects of computing. Most efforts have been additive in nature-i.e., introducing the societal aspects in addition to the technical aspects-rather than integrating social and technical dimensions (e.g., DeHart, 2022). In other efforts where race-based conversations were purposefully integrated into teacher PDs, race-related issues were discussed in terms of recruitment and retention of students from marginalized groups-i.e., working on strategies to remedy the structural lack of opportunities-while computing concepts and constructs were discussed separately from how computing algorithms may encode racism and have implications for users from marginalized communities (Goode et al., 2020). Integrative approaches that consider programs and algorithms hand in hand with societal aspects that shape them, rarely studied, are limited to researcher suggestions of potential classroom activities and examples (e.g., Ko et al., 2022) rather than explored in action in PDs. Thus, we know very little about what goes on within such PDs, despite recent calls to integrate the societal and technical concerns and shift away from an additive approach of disconnected modules on equity and social justice within computing (Goode et al., 2020;Ko et al., 2022).
This study was conducted with the Exploring Computer Science (ECS) teacher community, developed over 10+ years to support teachers to teach ECS curriculum (an introductory computing course) built on three pillars: computing concepts, equity, and pedagogy (Goode et al., 2012). PDs involve a weeklong session for two summers with four day-long quarterly sessions between the two. Previous research has found this model useful to develop teacher communities (Goode et al., 2014). With electronic textiles (e-textiles) offered as an optional unit within ECS (Kafai et al., 2019), the e-textiles online PD was designed along similar lines: ECS teachers with prior etextile teaching experience facilitated summer and quarterly sessions for ECS teachers new to e-textiles. As a part of a research-practice partnership (RPP), the PD had a diverse group of participants: university researchers and a non-profit partner, e-textiles-experienced teachers who were teacher-facilitators, and e-textiles-beginner teachers who were teacher-learners. Annie, Davon, Elisa, Julie, Leah, Maggie, Maria, and Leisha (all pseudonyms) were the teacher-learners with a wide range of experience teaching ECS (see Table 1 for their racial and gender identities, teaching experience and contexts). We analyzed teacher interactions during one of the quarterly PD sessions where teachers, except Maria and Leisha (unable to attend this session), worked on a sensor-based etextiles Human Sensor project and discussed technical and societal topics around it.
Yursa and Janine were senior researchers. Dana and Grace, also senior researchers, partnered with teacher-facilitators to design and implement the PD. Kate works for non-profit organization responsible for supporting PD online. Ben and Jesse, with Angela (not present during this session) were ECS teacher-facilitators with 4-6 years of experience teaching e-textiles. Ben facilitated the chosen session on the Human Sensor project. Leo, Mona (not present during this session), and Geeta were three graduate student researchers assisting data collection. All of us identify as cis-women or cis-men. Among cis-women, Grace, Janine, Dana, and Kate identify as White, Geeta as South Asian, Mona as Black, Angela as Asian-American, and Yursa as Indo-European. Among cis-male, Ben identifies as White, Jesse as Latino-European, and Leo as Latino. Shaped by our teaching, learning, and research experiences within CS across diverse contexts, we are committed to center justice in CS education. The teacher-learners were working on the Human sensor project as a part of the constructionist-driven e-textiles unit (Kafai & Fields, 2019). As shown in Fig. 1 (left), it required sewing a pair of aluminum patches to act as analog touch sensors when connected to the microcontroller pins and programmed using an Arduino programming environment. The teacher-facilitator team chose the lesson about computationally testing different sizes of aluminum patches as an opportunity for conversations about inclusivity of diverse users. By then, teachers had designed circuits and aesthetics to make 3-dimensional soft toys that would respond to different degrees of touch (no touch, soft, medium, and hard press). They tested sensors with family members and noted down the range of sensor values for different sizes of sensors (small, medium, big, Fig. 1, right). The session after sensor testing was chosen for analysis since Ben orchestrated a conversation involving social and technical aspects of computing.
A human sensor project template (left) and analog sensor readings (right).
Screen recording of the online PD session (45 min.) is the primary data source. PD materials such as session agenda, teacher project designs, and teacher pre-PD interviews were analyzed to provide the video with context. As a first step, Geeta (the first author) analyzed teacher pre-PD interviews for teacher backgrounds and teaching contexts and examined project designs and the session agenda to contextualize conversations. The teacher group was the unit of analysis with a focus on group meaning making as discussions ensued during the session.
Inspired by interaction analysis (Jordan & Henderson, 1995), Geeta structured collaborative video analysis (Erickson et al., 2017) supplemented by repeated individual viewing and interaction with the transcript. In this online PD session, interactions were multimodal, including verbal utterances, text-based chat, and screenshares, with embodied participation less visible. Geeta initially viewed the video to prepare a multimodal transcript to capture talk, gestures, facial expressions, chat entries, and entries on shared documents. The transcript was elaborated with comments during three ~2-hour long iterative group viewing sessions attended by the author team. This allowed for cross-expertise discussions and exchanges of perspectives during video analysis (Erickson et al., 2017). The multimodal transcript served as a shared artifact that viewers interacted with while watching the video together, pausing the video for notetaking whenever required. Joint viewing sessions helped us identify three key episodes where the technical and societal dimensions interweaved in teacher discussions. Geeta followed up with individual viewing sessions, rewatching the video in relation to the comments gathered, particularly to answer the research question. A visualization of the dynamics of the video was generated collaboratively by Mona and Geeta (Fig. 2) that represented the flow of the session, particularly how different aspects were initially individually discussed and then interweaved across the three episodes. The visualization was shared with the analysis group for further engagement with the emerging themes.
Teachers engaged with critical computing as they discussed inequities in technical designs and implications for diverse users within the context of their human sensor projects and real-world technologies, while anchoring discussions in their personal and classroom experiences (Fig. 2 for visualization). Below we elaborate three episodes to demonstrate how teachers integrated the technical and the social aspects of computing, and how their experiences tethered their learning, supported by a trimmed version of the verbal transcript due to space limitation. Episode 1: Do you think everybody will be able to use your device the same?
During this episode, teachers grappled with technical decision-making around sensor value ranges within their individual projects and the implications of those decisions on different users. Upon designing the aesthetics, teachers had to determine the size of the analog sensor patches for their projects and program them to allow users to interact with them and generate four different light patterns. Towards that end, teachers were tasked with testing sensor patches of different sizes (small, medium, and large) with any family members around and noting the lowest and the highest values (Fig. 1, right). Ben asked teachers to determine four sensor value ranges that could be included within their programs to correspond to the four lighting patterns caused by different degrees of touch.
After the teachers determined the sensor ranges, Ben moved their attention towards the social aspects of the design. Connecting the numerical values to implications for users, he asked if "everybody will be able to use [the] device the same [way]?" (Table 2). While teachers had grappled with the dimensions of programming sensors at an individual level until now, they had to extend their individual observations and reasoning across their colleagues' datasets and discuss the human/social dimensions of these choices as they related to people who might interact with their projects. Design decisions, in this case, were couched inside the objective of not just making a functional project but one that includes diverse users. This brought in the broader societal issues of technology design in conversation with the technical details of programming, issues around who can and cannot use their projects. Teachers reasoned their choice of sensor value ranges in relation to their observations of their family members testing the sensors earlier and discussed implications of their decisions on such interactions. For instance, Davon noticed the effort it took him to realize a particular low value, reflected on how it might be difficult for Maggie's child who may not be able to exert the same pressure on the sensor. He said this led him to include the 0-500 range in his program to accommodate similar users. Maggie discussed her observations of her and her children's interactions with the sensor in terms of individual physiology and meaning making of possible interactions. Further, Davon acknowledged the limitation of sensor testing, i.e., he was the only person who tested his sensors (Fig. 1, right) and how that further shaped his decisions around sensor value ranges for his project. Overall, in bringing the technical aspect of programming analog sensors in relation to users' interactions and experiences, social aspects were no longer a modularized concern but deeply integrated with the project code.
Episode 1 transcript excerpt at 34:24.
Verbal transcript Ben: Do you think that everybody would be able to use your device the same? Elisa, Maggie, and Leah shake their head to tell no. Davon: Probably not [unmuted by mistake]. Oh! Ben: So why don't you talk to me, why not? What differences do you see?
How are the decisions that we make in these light patterns going to manifest in terms of how people use the thing that we're making? Davon: Well, I know for me like I'm using the regular one [medium-sized patch], so I don't know… I had to squeeze real hard just to get my lower numbers, somebody might not be able to squeeze. Ben: [reading from chat] Your light pattern four is from zero to 500 right.
Did anyone get down to below 500? Maggie: I didn't but my daughter did. And, so in our case, at home, we all did it differently. I was pressing on top of the Episode 2: Everyday technologies that are not inclusive
In addition to connecting sensor ranges to user interactions within e-textiles projects, teachers continued to weave societal and technical aspects as they discussed various sensor-based technologies that excluded or caused ineffective outcomes for certain user groups. Conversations around inclusive design of individual projects in the previous episode shifted to everyday technologies that "did not think through inclusiveness" (Table 3). Teachers drew from a range of personal connections and discussed how technologies in their lives did not include marginalized populations. For instance, Leah mentioned cell phone touch screens that assume certain physiological characteristics such as shape and sensitivity, and Elisa shared the lack of consideration of human aspects in the design of motion-sensor lights in her classroom. Further, Julie shared about her five-year-old with Down's syndrome and his struggle with voice assistants like Alexa.
However, lively conversations only occurred when teachers related their lived experiences. Otherwise, there was awkward silence. For instance, teachers barely engaged when Dana and Ben presented examples from outside personal experience, such as pulse oximeters that "don't work well for dark-skinned people" and airport scanners that have "a much higher significant hit on Black women's hair" respectively. Perhaps bringing in external sources disrupted the conversation or perhaps the lack of personal connections among teachers in the room triggered the silence. Except Davon, all other teachers in the room identified as non-Black and not darkskinned (two Black female teachers were absent: Maria and Leisha, see Table 1) which could have led to silence around topics related to Black community's interaction with technology. This alludes to the need for diverse teacher experiences within PDs and supporting non-Black teachers to engage with race-related topics. clearly their user sample's not very great. Ben: Or for example, those things in the airport, they go around you like that zzzz, to make sure that there's nothing on you. There's a much higher significant hit on Black women's hair. So, if you are a Black woman going through that security system, you have a significantly higher chance of being removed from line. [silence in the room] Episode 3: Experiences as a tether Throughout, personal and classroom experiences tethered teachers' weaving of societal with technical aspects. Making the project put the teachers in students' shoes and enabled reflections on their learning and pedagogy (Episode 1). Building on Episode 2, teachers drew from their personal experiences with everyday technologies to identify bias. Drawing these together, Ben prompted teachers to "link" between the two (Table 4). Coming back to technical implications around determining sensor ranges in their projects, teachers concluded the need for "inclusive sample for testing" and again grounded conversation in their personal experiences.
When Ben encouraged teachers to connect classroom practice and think about supporting their students to do the same, teachers drew from their prior teaching and learning experiences to derive lessons for classroom practice. Of particular note, Annie shared how her students in a different class designed a pair of sunglasses without considering gender diversity among testers (Table 4). Narrating this instance, Annie reflected on her knowledge of her students and how she "ignorantly… thought [her students] would get inclusive feedback." She instead concluded that "it's very important to help kids understand who should be in the sample size." Not visible in the transcript, other teachers like Leah discussed other pedagogical approaches such as asking "who's missing in the sample data" to support students through critical computing. Overall teachers' experiences-technical, personal and classroom-acted as a tether to learn to integrate technical with societal concerns during the PD. Ben: Talking about sample, when we're talking about design. Annie: Right. Very much so it's very important to design and helping the kids understand who should be in that sample size. Because I mistakenly thought they would have a broad representation, and then they didn't on their own. And I think some of these companies too, you know, you think you're good. And then all of a sudden, oh wait, we're not good.
A limitation of this study is the context which involved a particular group of teachers-experienced ECS teachers with an established sense of community, developed while attending ECS PD sessions together in the past. Further, they were attuned to equity and justice issues in computing education as they read and discussed Stuck in the Shallow End (Margolis et al., 2017) in ECS PDs and developed politicized trust (Goode et al., 2020). The findings presented above are couched in this context and further research may be needed to understand how these findings relate to PDs with teachers not necessarily having already established relationships and commitment to equity.
This analysis sheds light on dimensions of PD activities that can support teachers to engage critically with STEM disciplines like computing. Across the three episodes, teachers made connections with their lived and classroom experiences as they engaged with issues of technology design with implications for diverse users. The significant role played by teachers' personal experiences is comparable to earlier studies (Bianchini et al., 2015;Bukor, 2014). In cases where such connections were absent, as seen in episode two, teachers struggled to engage with critical issues, just as prior research revealed the struggles of White teachers in engaging with race-related conversations within computing PDs. They often expressed discomfort and adopted colorblind terminologies and discourses of "othering" when discussing aspects of Black culture such as cornrow braids in relation to computing (Goode et al., 2020). Given the demographics of K-12 teachers and the prominence of White teachers, it is important to support them to engage with disciplines critically. Future PDs need to build on teachers' lived experiences, draw in diverse teacher voices, and develop politicized trust among teachers.
Having diverse perspectives in PDs helps build thought-provoking interactions (as noted in episodes two and three), just as diverse technology design teams mitigate biases in technology design. As teachers build on their personal experiences from their lives and classrooms during PDs, having diversity along those lines will expand the accounts discussed in PD settings, just as Julie's personal experience with her child's use of voice assistants brought in a particular perspective to discussions around inclusive design of technologies and projects in episode two or Annie's classroom experience added a particular perspective in episode three.
Yet another key contribution of this analysis are the contextual factors that shaped the findings, in particular, creating a common project (like the human sensor e-textile project) and building on existing teacher relationships with politicized trust (from earlier shared PD experiences). Sensor-based devices, by their design, can encode inequities based on who's included in user-testing and who's left out (e.g., Benjamin, 2019). The etextile human sensor project, which involved programming analog sensors, enabled teachers to discuss and consider programming aspects with broader societal issues around sensor testing and to connect with the larger issue of myopically designed sensor-based devices such as voice assistants, enabling conversations grounded in but extending beyond e-textiles. At the same time, this focused hour of PD built on six prior days of relational work, with shared crafting times, reflections, family interruptions (characteristic of long video calls) and other shared experiences. Combined, these shared technical and social experiences made room for teachers to grapple with issues of inclusive design while trusting the space as safe to discuss their perspectives. Doing similar work with other teachers in other contexts will mean designing and creating such spaces for teachers to engage in similar ways. Sharing materials like a book, e.g., Stuck in the Shallow End (Margolis et al., 2017), that allow for political conversations in relation to the discipline (Goode et al., 2020) and creating technical projects, such as the human sensor project, lay the groundwork for disciplinary integration. Building relationships over many days, as the ECS model does across 14 days over two years, can provide the grounding for tethering technical, personal and societal aspects of computing. These combined technical and relational underpinnings will allow for integration beyond the more traditional additive or modular means of bringing up equity and justice-related issues in computing-like disciplines (DeHart, 2022). Many more future PD and teacher-learning studies are needed to explore integrating societal issues with computing content in the effort to bring critical computing to the fore in K-12 education.
(*) Teachers could not be present during the session analyzed; ( # ) RM == Racial Minority; FRL = Free and Reduced Lunch.