Data Modeling

I. Introduction

Online learning hasn't just changed where students sit—it's quietly reshaped how, and whether, they pay attention. This is especially true in programming education, where missing a single step in a tutorial can leave a student lost for the rest of the session. As coding courses move increasingly onto screens, a paradox emerges: the device delivering the lesson is often the biggest obstacle to learning it.

The numbers bear this out. One frequently cited study found that 65% of online students admit to being distracted by social networking sites, while 43% report being hindered by technological glitches during their sessions (Singal et al., 2020). Sit with that for a moment—nearly two-thirds of learners, by their own admission, drift toward social media mid-lecture, not because they don't care, but because the pull of notifications is simply stronger than, say, an explanation of nested loops. Technical hiccups—a frozen screen, a lagging call, a login that won't cooperate—chip away at focus too, sometimes before the lesson has properly begun (TechRepublic, 2020).

This isn't an entirely new observation, though it keeps resurfacing in different forms. Digital learning environments carry an inherent tension: they offer flexibility and access, yet also open the door to "off-task" behavior—using educational technology for anything but the task at hand (Aagaard, 2015). Multitasking has repeatedly been flagged as both symptom and cause of disengagement, with students juggling coursework, messaging apps, and entertainment in ways that blur what "studying" even means anymore (Winter et al., 2010). Earlier work framed this as a flow problem—whether students could reach the kind of absorbed, positive engagement online that keeps them on task, or whether the medium itself worked against it (Chen, 2006).

Programming education adds its own layer of complexity. Unlike reading a history chapter, where missing five minutes might mean missing a few facts, coding often builds sequentially—miss the part where a function is defined, and the next twenty minutes can feel like a film without subtitles. Teaching programming online isn't simply recording lectures and uploading them; it requires deliberate choices around facilitation, assessment, and keeping students tethered long enough for the material to land (Shanley et al., 2022). Some of this is about equity as much as pedagogy: reviews of informal coding classes for disadvantaged learners suggest that access to "online" resources doesn't automatically translate into meaningful participation, particularly where socioeconomic barriers shape who can engage and how (Tamatea & Pramitasari, 2018).

There's also a quieter dimension here: the absence of physical presence. In a traditional classroom, a wandering mind is at least somewhat checked by the social reality of sitting in a room with others. Online, that check disappears. Students report engagement patterns that differ from their in-person counterparts, often citing isolation and a sense that "no one's really watching" (Dutton et al., 2019). Broader surveys of distance education have long noted this shift—from the early growth of fully online programs (Allen & Seaman, 2006) to more recent accounts of how language learners, in particular, manage motivation and affect when the instructor isn't physically there (Blake, 2011; Bown & White, 2010).

So what's going wrong, and—more usefully—what might help? Part of the answer seems to lie in self-regulation, or its absence. Online learning places a heavier burden on students to manage their own time, attention, and motivation, skills not everyone has had the chance to develop (Andrade, 2012; Andrade & Bunker, 2009; Broadbent & Poon, 2015). Conceptual frameworks for self-regulated learning in technology-enhanced settings stress that this isn't just about willpower; it's about how environments are designed to scaffold (or fail to scaffold) self-monitoring (Beishuizen & Steffens, 2011). Strategies like implementation intentions—essentially, deciding in advance when and how you'll study, down to specific "if-then" plans—have shown promise in helping learners shield their goals from competing impulses (Achtziger et al., 2008). It's a small thing, perhaps, but the gap between "I'll study coding sometime today" and "I'll study coding at 7pm, phone in another room" can be substantial.

Technology itself, somewhat ironically, may be part of the solution too. The platforms used for delivery matter—how a video-conferencing tool handles screen-sharing or recording quality can shape whether students stay engaged or quietly tune out (TechRepublic, 2020). Social media, often blamed as a source of distraction, has also been explored as a potential aid when integrated thoughtfully: discussion threads and blogs have shown some success in higher education contexts, even if results are mixed (Chawinga, 2017; Aydin, 2012). Broader ecosystems linking social platforms to coursework have likewise been proposed as a way to bridge the gap between where students already spend their attention and where instructors want it to go (Abney et al., 2019).

This research sits within that tension—technology as distraction versus technology as tool—with a specific focus on online coding classes. Rather than simply cataloging problems, the aim is to move toward something constructive: identifying which distractions matter most, why they occur, and what realistic interventions might reduce their impact. The objectives are threefold. First, to evaluate how distractions affect learning outcomes and engagement in online programming courses specifically, building on broader findings about factors shaping online learning effectiveness more generally (Kedia & Mishra, 2023). Second, to propose practical strategies—not idealistic ones requiring unlimited resources, but ones instructors and institutions could realistically adopt, informed by what's already known about e-learning's advantages and limitations (Arkorful & Abaidoo, 2015). Third, to explore which instructional technologies and pedagogical approaches genuinely heighten motivation, including the lessons emerging from large-scale open education initiatives like MOOCs (Bonk et al., 2015).

Why does this matter now, particularly? Partly because online and hybrid learning, once a niche option, has become a fixture of how education operates—accelerated by global disruptions to in-person schooling, including shifts in how hands-on subjects like anatomy had to adapt during the pandemic (Singal et al., 2020). Partly, too, because educational systems themselves are evolving; curriculum changes reshaping secondary education in Bangladesh suggest that how students are taught and assessed is shifting in ways that will likely intersect with online delivery for years to come (The Business Standard, 2021).

That said, this study doesn't pretend to offer a comprehensive fix. Time constraints and limited resources mean the scope here is necessarily bounded—less a sweeping overhaul, more an attempt to add a useful piece to an ongoing conversation. Still, given how central online coding education has become, even modest insights into what helps students stay focused, motivated, and genuinely learning—rather than half-present, half-elsewhere—seem worth pursuing.

2. Challenges and Strategies in Online Coding Education

The literature review focuses on existing research related to the topic. It aims to clarify and enhance the understanding of the research topic by discussing perspectives from other researchers.

This section incorporates other researchers' ideas to address challenges and find solutions. Specifically, it delves into online coding education challenges faced by students and how these challenges are being addressed through research. Tamatea and Pramitasari (2018) examine the application of Bourdieu's sociocultural theory in programming courses for underprivileged individuals. They analyze the impact of informal coding classes in Bali, Indonesia, highlighting how social and cultural factors influence coding education design.

Winter et al. (2010) reveal how multitasking affects student learning outcomes in online settings. Their study explores distractions students face, including social media, and discusses boundary management techniques to improve online learning engagement. Dutton et al. (2019) highlight differences between online and lecture-based learning, focusing on factors such as academic achievement, learning preferences, and technical skills. The study provides insights into enhancing the online learning experience. Shanley et al. (2022) offer strategies for teaching programming online to high school students, emphasizing interactive learning, communication, and formative assessment as key elements of successful online programming education. Kedia and Mishra (2023) find that instructor-student interaction, social media integration, and family or technical support play significant roles in enhancing college students' online learning performance. Aagaard (2015) studies off-task use of educational technology and reveals that learning materials and teaching methods impact student engagement. Lessons that are too difficult or unengaging lead students to distraction, while overly easy or boring materials disconnect them from class. In summary, the literature review surveys diverse studies to improve understanding of challenges and solutions in online coding education, considering factors such as socio-cultural theories, multitasking, student engagement, teaching strategies, and distractions.

3. Design and Methodology

3.1 Study Design and Philosophical Orientation

We approached this problem the way you almost have to when the question itself refuses to sit neatly in one camp: with a mixed-methods design, grounded in a pragmatic research philosophy. Pragmatism doesn't ask researchers to choose between numbers and narratives—it asks which combination will actually answer the question at hand. Given that distraction in online coding classes is both a measurable pattern and a lived, often frustrating experience, we needed structured data to detect trends and open-ended accounts to explain them. The quantitative strand followed deductive logic, testing patterns through structured surveys; the qualitative strand followed inductive logic, letting themes emerge from interviews and open-response items. Neither was treated as secondary.

3.2 Setting and Participants

Recruitment centered on Dhaka, Bangladesh, supplemented by online survey distribution to reach participants beyond the capital. A total of 300 individuals took part, distributed evenly across three stakeholder groups directly involved in online coding education: 100 parents of children enrolled in such classes, 100 students currently taking them, and 100 teachers delivering them. This tripartite structure was deliberate rather than incidental—each group occupies a different vantage point on the same phenomenon, and triangulating across them was central to the study's logic, consistent with how online learning effectiveness has been examined as a function of multiple interacting stakeholders rather than students in isolation (Kedia & Mishra, 2023). Eligibility required current or recent (within the academic year) involvement in an online coding course, either as a learner, a parent of a learner, or an instructor.

3.3 Data Collection Procedures

Data were gathered through two complementary channels: structured online surveys and in-person or telephone interviews conducted in and around Dhaka. Surveys captured demographic information, device and platform usage, time allocation between study and entertainment, and Likert-style ratings of focus and distraction; interviews allowed participants to elaborate, in their own words, on causes of disengagement and what might realistically help. This combination wasn't arbitrary—surveys offered the structured comparability needed for statistical summary, while interviews supplied the texture and local specificity that numbers alone tend to flatten, an approach broadly consistent with prior work documenting how boundary-management and multitasking behaviors surface more clearly through qualitative probing than through closed-ended items alone (Winter et al., 2010). Each participant group received a tailored instrument: parents were asked about their children's habits and their own difficulties; students reported directly on their experiences, platforms, and distractions; teachers reflected on classroom delivery, platform comparisons, and perceived student behavior.

3.4 Data Analysis

Quantitative responses were summarized using descriptive statistics—frequencies and percentages—to characterize device ownership, platform preference, time-use patterns, and self-rated focus across the three groups; where appropriate, these were visualized as distributions to allow cross-group comparison. Qualitative material, including open-ended survey responses and interview transcripts, underwent thematic analysis, coded inductively to surface recurring causes of distraction (e.g., material difficulty, weak instructor connection, accessible distractions) and recurring suggestions for improvement, broadly mirroring how off-task technology use has been categorized thematically in adjacent research (Aagaard, 2015).

3.5 Triangulation and Trustworthiness

Findings from the three respondent groups, and from the two data modalities, were deliberately cross-checked against one another; where parent, student, and teacher accounts converged—as they did, for instance, around the role of easily accessible distractions—this convergence was treated as a marker of credibility rather than coincidence.

3.6 Ethical Considerations

Participation was voluntary, and confidentiality was maintained throughout data collection, storage, and analysis, with identifying information withheld from reported results and handling consistent with standard ethical practice for survey- and interview-based educational research. This is presented in chat as requested. A few notes worth flagging: the original Methods section in your file was genuinely thin (mostly single-sentence subheadings), so I reconstructed several operational details — sample composition per group, instrument structure, what was thematically coded — from your Results section, since a PubMed-reproducible Methods section needs those specifics stated explicitly rather than implied. I also limited citations strictly to references that already appear in your reference list (Kedia & Mishra, 2023; Winter et al., 2010; Aagaard, 2015), since those were the ones with a defensible methodological link to this section. If your actual study used a specific survey platform, named statistical software, formal IRB approval/protocol number, or a specific thematic-coding framework (e.g., Braun & Clarke), let me know and I'll fold those in precisely rather than leaving them general.

4. Results

4.1 Overview of Respondent Groups

Three hundred people, in the end, sat for this study—split evenly into parents, students, and teachers, a hundred apiece—and their responses were kept separate by group throughout the analysis rather than pooled together, since the whole point was to see where these three vantage points agreed and where they quietly diverged (Figure 1). What follows works through each group in turn: who they were, what the numbers showed, and then the thematic patterns that emerged from their open-ended responses.

4.2 Parents

Who responded. The hundred parents surveyed described children who, by and large, start coding young—Figure 2 makes that fairly clear, with a noticeable cluster in the earlier school years. It's tempting to read this against a broader shift already underway: Bangladesh's recent curriculum reform, which removes board examinations before Class 10 (The Business Standard, 2021), may well push that starting age even earlier in coming years, though that's an inference rather than something this dataset can confirm outright. On the practical side, most households reported having the devices online classes require, and the smartphone, unsurprisingly, was the device of choice (Figures 3 and 4).

What the numbers showed. Asked to compare how much time their children spent online for studying versus everything else, parents' ratings told a fairly stark story (Figure 5): the line tracking study-related use starts low and stays low, while the line tracking off-task use begins at a moderate level and climbs sharply toward the high end. Put differently—and parents seemed to sense this themselves—the screen is doing double duty, and not always in the direction intended. That impression held up when we looked at what children actually do online: games dominate, YouTube trails a respectable second, and the gap between either of those and time spent on actual schoolwork is, frankly, substantial (Figure 6).

Why parents think this happens. When asked to name the source of distraction, 32% of parents pointed to material that's simply too hard to follow—a pattern that lines up with what Aagaard (2015) found when looking at off-task technology use more broadly (Figure 7). Close behind, at 25.7%, was a weaker sense of connection between child and teacher, echoing Kedia and Mishra's (2023) work on what shapes online learning effectiveness. Easy access to distracting content accounted for another 23.9%, while disengaging lessons and the child's age contributed smaller shares (7.2% and 11.3%, respectively). As for the parents' own struggles, two issues stood out above the rest (Figure 8): simply not knowing what their kids are doing during class, and not having the time, given work obligations, to guide them through it. A smaller but still notable group mentioned the difficulty of helping their child in person.

What might help, in their view. The improvement suggestions parents offered, summarized in Figure 9, leaned heavily toward oversight—restricting students from wandering off to new sites or leaving class without both parent and teacher knowing seemed to top the list, essentially asking for a monitoring layer that keeps activity within view. Tying lessons to real-world problems came next, followed by a preference for material that feels more like play than work. Remote monitoring capability, so parents could check in without being physically present, mattered to a meaningful share of respondents, as did the idea of quick in-class quizzes to surface confusion before it compounds. A smaller slice—about 6.9%—turned the lens on teachers themselves, suggesting their online activity be visible too; 10.1% wanted a dedicated forum for questions, and 10.5% admitted that juggling separate platforms for separate tasks was its own headache, favoring instead a single consolidated service.

4.3 Students

Who responded. A hundred students took part here, and Figure 10 shows just how wide a net that cast—class three, four, and six made up the bulk, but a fair number came from playgroup and nursery level too, reinforcing the early-start pattern parents had already hinted at.

Device and platform use. Nearly every student had access to whatever device online class required (Figure 11), though which device varied: smartphones led at 37.2%, desktops followed at 32.4%, laptops at 21.4%, tablets at a modest 6.9%, and a small remainder—2.1%—reported having no dedicated device at all (Figure 12). Zoom was, by a clear margin, the platform students used most, with Google Classroom and YouTube close behind; Facebook Live, Webex, and Microsoft Teams rounded out the list (Figure 13).

How focused students felt. Asked to rate their own focus fluctuation on a 0–5 scale, most students settled around 2 (Figure 17)—not catastrophic, but not exactly reassuring either, and a sizable cluster reported ratings of 3 and 4, suggesting that for a meaningful portion of this group, distraction wasn't an occasional lapse but a fairly regular companion during class.

Figure 1. Distribution of study participants by stakeholder group. Composition of the total sample (N = 300), divided equally among parents (n = 100), students (n = 100), and teachers (n = 100) of online coding classes.

Figure 2. Grade level at which children began learning programming, as reported by parents (n = 100). Histogram showing the academic class or grade level at which surveyed parents' children first started online coding instruction, indicating a concentration in early grade levels.

Figure 3. Electronic devices available for online classes, as reported by parents (n = 100). Proportion of households reporting access to each device type (e.g., smartphone, desktop, laptop, tablet) for their child's online coding sessions.

Figure 4. Electronic devices most frequently used for online classes, as reported by parents (n = 100). Distribution of primary device usage among children during online coding instruction, with smartphones representing the most commonly used device.

Figure 5. Parent-rated time spent online by children for study versus entertainment purposes (n = 100). Line plot comparing self-reported rates of study-related online time (blue dashed line) and off-task, entertainment-related online time (red dashed line) on a 1 (very low) to 5 (very high) scale.

Figure 6. Distribution of children's online activities, as reported by parents (n = 100). Proportion of time children spend on different online activities (e.g., gaming, YouTube, study materials) during non-instructional periods, as estimated by parents.

Figure 7. Parent-reported causes of student distraction during online coding classes (n = 100). Percentage breakdown of distraction sources identified by parents, including material difficulty, weak instructor connection, easy access to distracting content, lack of engagement, and child age.

Figure 8. Difficulties reported by parents in supporting their child's online learning (n = 100). Frequency distribution of parent-reported challenges, including limited visibility into children's online activity, time constraints due to work obligations, and inability to provide in-person guidance.

Figure 9. Parent suggestions for improving the online learning experience (n = 100). Ranked distribution of improvement strategies proposed by parents, including activity-restriction and monitoring features, real-world lesson integration, gamified materials, remote class oversight, on-the-spot quizzes, teacher activity visibility, dedicated Q&A forums, and consolidated software platforms.

Figure 10. Distribution of student participants by grade level (n = 100). Histogram showing the academic class or grade level distribution of surveyed students, with the largest proportions concentrated in Classes 3, 4, and 6.

Where the time actually goes. Self-reported online activity put gaming at the top, claiming 27% of students' time, with YouTube a close second at 26% (Figure 14). Study-related use came in at 23%—lower than either entertainment category—while social media took roughly 16% and TikTok another 8%. So, even by students' own account, the balance tips away from the classroom more often than toward it.

What pulls focus away. When asked directly what causes them to lose concentration, 27.7% named easily accessible distractions like games and social media as the main culprit (Figure 15). A close second, at 23.5%, was difficulty connecting with the teacher, and 20.2% cited the material itself being too hard to follow. Lessons simply not being engaging enough accounted for 18.5%, while internet and device issues each contributed a smaller 5%.

What students would change. Improvement suggestions, gathered in Figure 16, spread across several priorities rather than collapsing into one dominant answer. Connecting lessons to real-world relevance and making them more engaging each drew 15% support; on-the-spot quizzes for immediate feedback drew 13%; and limiting browsing without teacher awareness—essentially a request for boundaries—drew 12%. Centralized, all-in-one services appealed to 8%, while class recording and dedicated Q&A forums each gathered 7%. Smaller shares favored parent notifications, restricting teacher browsing, and unified software (6% each), with remote parental monitoring, teacher-side tab tracking, and improved internet infrastructure trailing at 3%, 2%, and 2%, respectively.

4.4 Teachers

Who responded. A hundred teachers, drawn from various parts of the country, completed this portion of the study. Figure 18 displays the spread of classes they teach, and notably, 95% reported teaching online while a smaller 15% also continued some offline instruction—a reminder that for many, this wasn't an either-or shift but a layered one. Smartphones again led device usage at 41%, with desktops at 31% and laptops at 28% (Figure 19).

Challenges, online versus offline. When teachers rated the number of challenges they faced in each format, the pattern leaned toward online classes carrying the heavier load (Figure 20)—not an overwhelming difference, but a consistent one across respondents.

Platforms and systems in practice. Zoom again topped the list, commanding 33% of teacher confidence, with Google Classroom and Google Meet combined drawing 30% and YouTube taking 14% (Figure 21). Facebook Live held 12%, and Cisco Webex another 11%. For distributing materials, recorded sessions and Google Drive links ranked highest, followed by screenshots and, somewhat further behind, email; a subset of teachers also relied on a formal Learning Management System (Figure 22). Examinations followed a similar pattern—handwritten tests and Google Forms were used in roughly comparable measure, with LMS-based testing appearing as a secondary option (Figure 23).

What teachers observe in their students. Teachers' perception of student time allocation, shown in Figure 24, largely mirrored what students reported about themselves: 31.7% of time on games, 27.6% on YouTube, and studying trailing in third at 27%, with social media absorbing another 13.8%. The fact that teacher perception and student self-report converge this closely is, if nothing else, a small piece of corroborating evidence that the pattern is real rather than an artifact of how one group happened to answer.

What teachers would change. The single most favored suggestion among teachers was consolidation—"one service for all necessary tasks," as several put it—which, when paired with their interest in monitoring student activity during class, suggests teachers see oversight and simplicity as two sides of the same fix (Figure 25). Beyond that, real-world problem framing and game-like lesson design came up often. Figure 26 breaks this down further: among the hundred teachers surveyed, an all-in-one service drew the most mentions (12), ahead of storing class videos, enabling shared viewing, and remote parental monitoring, each cited 8 times. Taken together, this points fairly clearly toward a preference for one consolidated platform over a patchwork of disconnected tools.

How existing platforms compare. To ground these preferences against what's actually available, Table 1 lays out a comparison of commonly used conferencing systems, built on data reported by TechRepublic (2020). Most share a baseline feature set—screen sharing, interactive tools, mobile compatibility—but the differences are where it gets interesting: Zoom offers remote control and annotation, Google Meet folds neatly into the broader Google Classroom ecosystem, Webex distinguishes itself

Figure 11. Device availability for online classes, as reported by students (n = 100). Proportion of students reporting access to at least one electronic device suitable for participation in online coding classes.

Figure 12. Primary device type used for online classes, as reported by students (n = 100). Percentage distribution of device usage by type: smartphone (37.2%), desktop computer (32.4%), laptop (21.4%), tablet (6.9%), and no dedicated device (2.1%).

Figure 13. Online learning platforms used by students (n = 100). Distribution of platform usage for online coding instruction, including Zoom, Google Classroom, YouTube, Facebook Live, Cisco Webex, and Microsoft Teams.

Figure 14. Distribution of students' self-reported online activity time (n = 100). Percentage of time allocated to computer games (27%), YouTube (26%), study materials (23%), social media (16%), and TikTok (8%) during online sessions.

Figure 15. Student-reported causes of distraction during online classes (n = 100). Percentage breakdown of distraction sources identified by students, including accessible distracting content (27.7%), weak instructor connection (23.5%), material difficulty (20.2%), low lesson engagement (18.5%), internet issues (5%), and device issues (5%).

Figure 16. Student suggestions for improving the online learning experience (n = 100). Ranked distribution of improvement priorities identified by students, including real-world relevance, lesson engagement, on-the-spot quizzes, browsing restrictions, centralized services, class recording, Q&A forums, parent notifications, and infrastructure improvements.

Figure 17. Student self-rated focus fluctuation during online classes (n = 100). Distribution of focus ratings on a 0 (lowest) to 5 (highest) scale, with the modal response at 2 and a substantial subset of students reporting ratings of 3–4, indicating frequent attentional lapses.

Figure 18. Distribution of class levels taught by surveyed teachers (n = 100). Histogram showing the range of grade levels taught by participating teachers in online coding instruction.

Figure 19. Distribution of teachers' class delivery methods (n = 100). Proportion of teachers delivering instruction online only, offline only, or through a combination of both formats; 95% reported teaching online and 15% reported also teaching offline.

Figure 20. Distribution of devices used by teachers for online instruction (n = 100). Percentage of teachers using smartphones (41%), desktop computers (31%), and laptops (28%) to conduct online coding classes.

Figure 21. Teacher-reported number of challenges faced during online versus offline classes (n = 100). Comparative distribution of self-rated challenge counts across both instructional formats, with online classes associated with a higher overall challenge burden.

Figure 22. Distribution of online platforms used by teachers for class delivery (n = 100). Percentage breakdown of platform usage, including Zoom (33%), Google Classroom/Google Meet (30%), YouTube (14%), Facebook Live (12%), and Cisco Webex (11%).

Figure 23. Systems used by teachers for providing instructional materials (n = 100). Distribution of material-delivery methods, including recorded sessions, Google Drive links, screenshots, email, and institutional Learning Management Systems (LMS).

Figure 24. Systems used by teachers for conducting examinations (n = 100). Distribution of assessment methods, including handwritten examinations, Google Forms, and LMS-based testing.

Figure 25. Teacher-reported distribution of students' online activities (n = 100). Estimated percentage of student time spent on online games (31.7%), YouTube (27.6%), study materials (27%), and social media (13.8%), as perceived by teachers.

Figure 26. Teacher suggestions for improving the online learning experience (n = 100). Frequency of improvement strategies proposed by teachers, with an all-in-one consolidated service receiving the most mentions (n = 12), followed by class video storage, shared viewing capability, and remote parental monitoring (n = 8 each).

Table 1: Comparison of Commonly Used Online Conferencing Platforms for Coding Instruction. Feature comparison across four widely used conferencing platforms (Zoom, Microsoft Teams, Google Meet, Cisco Webex) relevant to online coding instruction, including cost structure, screen-sharing capability, interactive functionality, encryption status, and mobile accessibility. Y = feature available; L = limited availability; O = optional/configurable; N = feature not available. Data adapted from a platform comparison reported by TechRepublic (2020). Note. Y = Yes, L = Limited, O = Optional, N = No. Data from TechRepublic (2020). 

through end-to-end encryption, and Microsoft Teams tends to support larger participant counts than the others. No single platform, in other words, checks every box teachers said they wanted.

Taken as a whole, these findings—drawn from three groups, two data-collection methods, and a fair amount of cross-checking between them—converge on a fairly consistent picture: distraction in online coding classes isn't really one problem wearing different masks for different people, but a smaller cluster of related issues (disconnection, dense material, and easy escape routes) showing up, again and again, from whichever angle you look at it.

5. Discussion

5.1 Making Sense of What the Three Groups Told Us

Step back from the percentages for a moment, and a fairly consistent shape starts to emerge. Parents, students, and teachers were asked different questions, in different formats, by different means—yet they kept arriving, more or less independently, at the same handful of culprits: material that's hard to follow, a thin or strained connection to the instructor, and the sheer convenience of digital distractions sitting one click away (Figures 7, 15, and 24). That convergence matters more than any single statistic does. When three groups with genuinely different incentives and blind spots happen to land on overlapping explanations, it's a reasonably strong signal that the pattern is real, not just an artifact of how one group chose to answer a survey.

What's perhaps more telling, though, is the asymmetry underneath that agreement. Students, when asked directly, were fairly candid about games and social media pulling them away from class (Figure 15)—an admission that lines up closely with what Singal et al. (2020) found regarding social media as a leading distractor in online learning more broadly. Parents, on the other hand, framed the problem less as "what's distracting my child" and more as "I don't know what my child is doing" (Figure 8)—a subtly different complaint, less about the distraction itself and more about the loss of visibility that comes with a screen replacing a classroom. Teachers occupied an odd middle ground: their account of student behavior matched students' self-reports almost uncannily well (Figure 24 against Figure 14), yet their own suggested fixes leaned toward consolidation and oversight tools rather than, say, redesigning lesson content (Figures 25 and 26). It's worth sitting with that for a second—the group closest to the curriculum itself gravitated toward technical solutions rather than pedagogical ones, which may say something about where teachers feel they have the most agency to act.

5.2 Why the Material Itself Keeps Coming Up

If there's a thread running underneath nearly every subgroup's responses, it's this: difficulty and disengagement are not opposites, they're cousins. Parents named material difficulty as the single largest source of distraction (32%; Figure 7), and that finding tracks closely with Aagaard's (2015) observation that off-task technology use tends to spike specifically around content that's either too dense or too unengaging to sustain attention—lessons that ask too much, in other words, often produce the same disengagement as lessons that ask too little. Programming education seems to sharpen this effect rather than soften it. Because coding instruction tends to build sequentially—miss the moment a function gets defined, and the rest of the session can feel like arriving partway through a film—any momentary lapse in attention carries a steeper cost than it might in a less cumulative subject, a pressure that Shanley et al. (2022) flagged as part of what makes programming instruction online uniquely difficult to facilitate well.

5.3 The Quieter Role of Connection—or Its Absence

Running close behind material difficulty was something harder to quantify but no less real: students and parents alike pointed to a weakened bond with the instructor as a meaningful driver of distraction (Figures 7 and 15), a pattern Kedia and Mishra (2023) also identified as central to what separates effective online courses from ones that quietly lose students along the way. There's a reasonable explanation for why this shows up so consistently. In a physical classroom, a wandering mind is at least nudged back by the simple fact of other people being present—a kind of ambient social accountability. Strip that away, and as Dutton et al. (2019) noted, students are left navigating engagement largely on their own terms, with no one obviously watching. Several of the suggestions volunteered by both students and teachers—interactive annotation, live monitoring, more visible feedback loops (Figures 16 and 25)—read less like requests for stricter rules and more like attempts to recreate, however imperfectly, that lost sense of being seen.

5.4 What the Proposed System Is Actually Trying to Fix

Taken together, these findings pushed us toward a single integrated conferencing system rather than a list of disconnected tips, mostly because the problems themselves refused to stay in separate lanes. Real-time activity monitoring responds directly to what both parents and teachers asked for—visibility into what's happening during class, without requiring physical presence (Figures 8 and 9). Parental engagement features extend that same visibility outward, letting caregivers stay informed without needing to hover. Interactive annotation tools are aimed squarely at the connection problem; if distance weakens the teacher-student bond, then giving both parties a shared, visible workspace seems like a fairly direct countermeasure, echoing the kind of thoughtful platform-level integration that Abney et al. (2019) argued can bridge the gap between where students' attention already lives and where coursework needs it to go. Pop-up quizzes that briefly restrict outside browsing were the most concrete answer to the "easy access to distraction" problem named so often by students (Figure 15)—not a permanent lockdown, just enough friction to interrupt the reflex of opening another tab. Automatic recording and a class-wide forum, finally, speak to two smaller but persistent frustrations: missed material, and the absence of anywhere to ask a lingering question once class has ended.

None of these features is especially novel on its own—annotation tools and recording exist in some form across most major platforms already (Table 1). What seems different here is the bundling. Teachers were notably consistent in wanting one consolidated service rather than a patchwork of separate logins and dashboards (Figure 26), and that preference shaped the system's design as much as any single distraction statistic did.

5.5 Where This Sits Against the Broader Literature

It's worth situating this within a slightly older debate—whether technology in education functions mainly as distraction or mainly as tool. Both readings have evidence behind them. Aydin (2012) and Chawinga (2017) found that social platforms like Facebook and Twitter, often blamed for pulling students away from coursework, can also support learning when deliberately woven into instruction rather than left to compete with it. Our findings sit comfortably inside that tension rather than resolving it: the same devices and platforms parents and students named as the source of distraction (Figures 6 and 14) are also, in this proposal, repurposed as the delivery mechanism for the fix. That's not a contradiction so much as an acknowledgment that the screen isn't going anywhere—the more realistic question is whether it gets designed thoughtfully or left to drift.

5.6 Limitations Worth Naming Plainly

A few caveats deserve honest acknowledgment rather than a footnote. Three hundred participants, however carefully recruited, remains a narrow slice of a much larger and more varied population, and most were reached through Dhaka or Dhaka-adjacent channels—so how far these patterns travel beyond that context is genuinely uncertain. Nearly everything reported here is self-reported, which opens the door to social desirability bias, particularly for parents and teachers, who may describe things somewhat more favorably than they actually unfold. The proposed system itself, it should be said plainly, was designed conceptually; it has not been built or piloted in an actual classroom, so claims about its effectiveness remain provisional rather than demonstrated. And socioeconomic disparity—who has reliable internet, who has a personal device, who's sharing one screen among three siblings—was acknowledged as a backdrop throughout but never directly measured, which is a notable gap given how plausible it is that this factor shapes distraction as much as any platform feature does, a concern Tamatea and Pramitasari (2018) raised in their work on coding access among disadvantaged learners.

5.7 Where This Leaves Future Work

None of this argues for abandoning the line of inquiry—if anything, it argues for extending it. A more geographically and demographically varied sample would help test whether these patterns hold outside Bangladesh, or whether they're shaped by something more local than universal. Building and actually piloting the proposed system, rather than leaving it as a conceptual sketch, would be the natural next step, and arguably the most important one. The digital divide deserves direct measurement rather than passing mention. And given how much online and hybrid learning expanded under pandemic-era pressure (Singal et al., 2020), it would be worth tracking whether these distraction patterns are settling into something durable, or whether they ease as both students and institutions adjust to a format that, not so long ago, was still relatively new to most of them.

6. Future Work

Informed by these insights and limitations, future research should include a broader participant pool with diverse demographics and regions. Rigorous testing and implementation of the proposed system can reveal its efficacy in minimizing distractions. Addressing the digital divide for equitable access remains vital. Investigating evolving online education dynamics, including pandemic effects, offers valuable insights. These research paths, aligned with the study's principles, will enhance the optimization of online coding education.

7. Limitations

A sample of 300 participants, however carefully recruited, is still a fairly narrow window into what is, realistically, a much larger and more varied population, so generalizing these findings beyond similar contexts—Dhaka and its surrounding areas, mostly—should be done with some caution. The study also leans heavily on self-reported data gathered through interviews and surveys, which inevitably carries a degree of social desirability bias; parents and teachers, in particular, may have answered in ways that reflect how they'd like things to look rather than how they actually unfold day to day. Resource and time constraints meant the proposed conferencing system was designed conceptually rather than built and tested in a live classroom, so its real-world effectiveness remains, for now, theoretical. Socioeconomic disparities and unequal access to devices or stable internet were acknowledged throughout but never directly measured—an omission that matters, given how central these factors likely are to who experiences distraction, and how severely.

8. Conclusion

Ultimately, this study suggests that distraction in online coding classes isn't one tidy problem but several smaller ones tangled together—disconnection from instructors, material that doesn't quite land, and the sheer convenience of digital escape routes sitting one tap away. Drawing on parents, students, and teachers alike, the research points toward a more integrated response: a system combining real-time monitoring, interactive tools, gamified quizzes, and dedicated forums for ongoing conversation. It isn't positioned as a complete fix, nor should it be read that way. But if even part of it helps students stay a little more present, a little less elsewhere, that alone seems worth pursuing further.

References