An Article for Teachers & Trainers
When "Just Watch This Video" Becomes an Empty Assignment
YouTube holds arguably the richest library of educational video ever assembled. Teachers assign it constantly — and, without a way to verify that students actually engage with it, that trust can quietly backfire.
The Argument
Six ideas that explain why YouTube homework needs a safety net
Each idea builds on the last. Read them in order — together they form the case for tracking comprehension, not just assigning content.
YouTube is the largest library of educational video on the internet
No single platform hosts more instructional content, in more subjects and languages, than YouTube. Research by Oxford Economics, commissioned by YouTube, found that 84% of European teachers who use YouTube have used its content in their lessons and/or assignments [1]; the equivalent figure for US teachers is 94% [2]. In June 2025 alone there were more than 5.5 billion views of learning and how-to content in the US [2]. A separate peer-reviewed survey of 1,150 teachers in Spain found video to be a common instructional resource — though that study also concluded that use varies significantly by educational stage and school type [3].
Teachers routinely assign videos for students to watch outside class
Flipped classrooms, homework review, and self-paced revision all lean on the same assumption: send the link, and the student will watch and absorb it. In a 2024 national survey of 5,259 US college instructors, 42% of social scientists and 41% of scientists said they ask students to watch pre-recorded lectures so that class time can be reserved for discussion and problem-solving [4]. A survey of 12,070 US faculty found that a third (33%) had used a flipped classroom model — in which the lecture and homework elements of a course are reversed — in at least a few of their courses [5]. In K-12, the Speak Up national survey found 16% of teachers regularly creating videos of their lessons or lectures for students to watch [6]. In the US, 79% of teachers who use YouTube agree that it helps students learn [2].
The share varies by sector and by year. The mechanics do not: the video is watched somewhere the teacher cannot see.
The logs say a great deal of it never gets watched
This is the part that self-report surveys cannot reach. When researchers stop asking students and start reading the platform logs, the picture changes sharply.
In a flipped dental course, learning-management-system records showed that just 40% of students watched more than half of the assigned videos — and 13% watched none at all [7]. A study of 51 videos in an undergraduate medical course found a mean viewing rate of 41%, with 56% of registered students never interacting with the videos at all [8]. Across 77 videos and roughly 1,800 students at an Australian university, only 43% of those who started a video watched it to the end [9]. In a Norwegian engineering-mathematics course, students watched an average of 57% of each video [10]. And the largest study of its kind — 6.9 million viewing sessions across four edX courses — found that median engagement time was at most six minutes regardless of how long the video ran, with students often not making it halfway through anything longer than nine minutes [11].
Different countries, different disciplines, different platforms, a decade apart — and the same finding each time. Assigning a video and receiving a watched video are two different events.
The reason: once online, students drift toward entertainment, not education
The moment a student opens a browser or app to "watch the assigned video," they are one tap away from an entirely different feed. Research using real device-tracking rather than self-report is the clearest evidence of this: a University of Trento study logged the smartphone activity and time diaries of 72 students and found that social media was checked more often — and for longer stretches — during periods those students explicitly labelled as "studying" than during actual class lectures. Heavier and more frequent use correlated with both lower GPA and fewer credits earned [12]. Self-report surveys, even though they understate the true gap, still show it clearly: 1,000 British university students in one poll reported roughly 55 hours a week online but only about 14 hours of that on coursework [13], while Gallup found US teens report averaging 4.8 hours a day on social media — with 51% at four or more hours — far exceeding the time they report spending on homework [14].
"Time spent on a screen and time spent learning are not the same thing — and the gap between them is where good intentions quietly go to waste."
Asking the students is not the same as checking
A YouTube link sent home with no accountability mechanism relies entirely on the honour system — and the honour system does not survive contact with the data. The dental study above did something unusually valuable: it matched each student's LMS log against that same student's questionnaire response. 71% reported watching more than half of the videos. The logs showed 40% had [7].
That is a 31-point gap between what a teacher would hear and what actually happened — not because students are dishonest, but because nobody accurately remembers their own screen time. If a student opens the tab, mutes it, and scrolls a different app instead, neither the teacher nor the student reliably registers it. The American Academy of Pediatrics identifies multitasking as one mechanism behind this, noting that using screens while studying reduces focus and is associated with lower grades [15]. An observational study of 263 school and university students studying at home found they lasted under six minutes on task before switching to a technological distraction — social media, texting, or another open window [16] — a sobering number given that most video homework happens completely unsupervised.
The fix: put the check for understanding directly on top of the video
There is direct evidence that attaching a check changes the behaviour, not just the record of it. In a pharmacy therapeutics course that logged actual view time, median video viewing rose significantly when students knew a quiz was attached to the material — 80 minutes versus 69 [17]. The check does not merely reveal who watched; it increases how much gets watched.
Rather than adding a separate app, login, or platform that competes for a student's attention, that accountability layer should live exactly where the video already is. That is the founding idea behind KFQ (Kung Fu Quiz) — an interactive quiz layer that sits directly on top of any YouTube video, so teachers can see, question by question, whether a student actually watched and understood the material, with no extra tool required.
The Evidence
What the logs say about unmonitored video assignments
These figures come from peer-reviewed academic research and from industry studies commissioned by YouTube — none of them from KFQ. Where a figure comes from YouTube-commissioned research, it is labelled as such. The studies at the top of the table are the most important, because they measure what students did rather than what students said.
Tracking-based research is the most damning because it removes the guesswork. Self-report surveys are known to understate real screen behaviour — in a pilot study of 35 college students, those unaware of their own overuse underestimated their usage by roughly 40% relative to logged data, while actually using their phones about 15% more than average [18] — which suggests the true gap between assigned viewing and actual engagement is wider than most surveys reveal.
| Study | Method | Finding |
|---|---|---|
| Scott et al. (2021), dentistry, US (n≈104) | LMS logs matched to each student's own self-report | 40% watched more than half the assigned videos and 13% watched none — while 71% reported watching more than half [7] |
| Gutiérrez-González et al. (2024), medicine, Spain (n=109) | Platform analytics, 51 videos | Mean viewing rate 41%; 56% of registered students never interacted with the videos at all [8] |
| Maloney et al. (2022), mixed disciplines, Australia | LMS + video-player logs, 77 videos | Of students who started a video, only 43% watched it to the end [9] |
| Guo, Kim & Rubin (2014), edX MOOCs | 6.9M viewing sessions, clickstream logs | Median engagement at most 6 minutes regardless of video length; students often stop before halfway on videos over 9 minutes [11] |
| Patanwala et al. (2017), pharmacy, US (n=100) | Logged video view time | Median viewing time significantly higher when a quiz was required — 80 min vs 69 min [17] |
| SmartUnitn, University of Trento (n=72) | Smartphone logs + real-time diaries | Social media checked more often during declared "study" time than during lectures; correlated with lower GPA and fewer credits earned [12] |
| Hyperoptic / OnePoll, UK students (n=1,000) | Self-report survey | 54h47m/week online in total, but only 13h56m on coursework — incl. 14h36m social media, 10h55m streaming, 6h16m gaming [13] |
| Gallup, US teens (2023) | Self-report survey | Teens report averaging 4.8 hrs/day on social media, 51% at 4+ hrs/day — far exceeding self-reported homework time [14] |
| Oxford Economics / YouTube, US | Commissioned survey + platform data | 94% of US teachers who use YouTube use its content in lessons and/or assignments; 5.5B+ views of learning and how-to content in one US month [2] |
The takeaway: the same openness that makes YouTube such a rich teaching resource is also what makes an unverified assignment risky — the platform that hosts the lesson also hosts the distraction, one click away.
The Solution
KFQ: a comprehension check that lives on the video itself
KFQ (Kung Fu Quiz) overlays interactive questions directly onto any public YouTube video. Students watch and answer inside the same viewing experience — there is no second app to open, no separate login to remember, and no ambiguity about whether the assignment was actually completed.
Attach a YouTube link
A teacher takes any public YouTube video — a lecture, a demo, a review clip — and pastes the link into KFQ to create a Video Challenge.
Add or auto-generate questions
Teachers write their own questions across six question types — single answer, multiple answer, drag words into box, fill in the blanks, sort sentence, and button — or let KFQ's AI (sensAI) generate video-aligned questions automatically at key timestamps, saving hours of prep [19].
Track, reuse, and repurpose
Every Challenge is saved to the teacher's account, ready to reassign to a new class or session, and results feed into dynamic, data-driven insights on who watched, who understood, and where to reteach.
The video stays exactly where it is. The accountability comes to it — not the other way around.
In Their Own Words
Teachers who put KFQ on top of their videos
These accounts come from educators across grade levels, subjects, and countries — from early years classrooms to MBA lecture halls — describing what changed once video assignments came with a built-in comprehension check.
Students began paying closer attention while watching videos because they knew they would be assessed on key concepts. Their confidence grew stronger as they solved challenging questions.
Instead of passively watching, they actively participated throughout the lesson. Kung Fu Quiz turned video-based lessons into an active learning experience while making outcomes easier to track with report insights.
If a tool does not increase student engagement and reduce my workload, it does not belong in my teaching. Kung Fu Quiz passed this test immediately.
By embedding questions into the Arabic Language-related YouTube videos, students immediately answer the quiz while watching the videos. And I noticed that this creates a more positive and competitive classroom atmosphere.
Innovative tools like this reflect exactly the kind of innovation that modern educators are looking for. By turning assessments into live quizzes, the platform gave the instructor a practical way to encourage participation and spark classroom discussion.
With a simple QR present on the screen, more than 60 students joined in the quiz. With leaderboards updating after every round, it really brought the competitive spirit.
If you assign YouTube videos, try KFQ on the next one
Teachers and trainers already use KFQ across subjects, grade levels, and countries to know — not guess — whether a video assignment landed. It sits on top of the video you already planned to use. No new platform for students to learn.
Try Kung Fu Quiz free →References
Sources cited in this article
- [1] YouTube Official Blog (8 Sept 2025). “How European kids, parents, and teachers are using YouTube for education.” Teacher figures from research by Oxford Economics, commissioned by YouTube; covers EU27 and the UK. Base for the 84% figure is teachers who use YouTube. blog.youtube/news-and-events/eu-youtube-for-education-study-2025/ · Primary source: 2025 YouTube Europe Impact Report
- [2] YouTube Official Blog (28 Apr 2026). “The global campus: How YouTube is empowering learners everywhere.” US learning-view data and the 79% figure. The 94% figure is stated more precisely in the underlying research — 94% of teachers who use YouTube use its content in lessons and/or assignments. blog.youtube/news-and-events/youtube-empowering-learning-tools/ · Primary source: 2025 YouTube U.S. Impact Report (Oxford Economics)
- [3] Pattier, D. (2021). “Teachers and YouTube: The use of video as an educational resource.” Ricerche di Pedagogia e Didattica — Journal of Theories and Research in Education, 16(1), 59–77. Sample: 1,150 teachers working in Spain, academic year 2019–20. rpd.unibo.it/article/view/11584
- [4] Love, S. J., & Blankstein, M. (2024). US Instructor Survey 2024: Findings from a National Survey. Ithaka S+R, 22 August 2024. 5,259 completed responses from an invited sample of 135,384 instructors at US four-year institutions (3.9% response rate), fielded 7 Feb – 20 Mar 2024. DOI: 10.18665/sr.321165. sr.ithaka.org/publications/us-instructor-survey-2024/
- [5] Brooks, D. C. (2015). ECAR Study of Faculty and Information Technology, 2015. EDUCAUSE Center for Analysis and Research, October 2015. 13,276 respondents across 139 institutions in 10 countries; findings developed from 12,070 US responses (1% margin of error), fielded 4 Feb – 20 Mar 2015. Verbatim: “ECAR found that a third of faculty (33%) have used a flipped classroom pedagogical model,” used “in at least a few of their courses.” ERIC no. ED593427. eric.ed.gov/?id=ED593427 · full report PDF
- [6] Project Tomorrow & Flipped Learning Network (2014). Speak Up 2013 National Research Project Findings: A Second Year Review of Flipped Learning. US K-12; opt-in national survey of 403,000+ students, parents, teachers, administrators and community members, fielded autumn 2013. Verbatim: “16 percent say that they are regularly creating videos of their lessons or lectures to students to watch.” flippedlearning.org — Speak Up 2013 Survey Results
- [7] Scott, J. M., Bohaty, B. S., & Gadbury-Amyot, C. C. (2021). “Using learning management software data to compare students' actual and self-reported viewing of video lectures.” Journal of Dental Education, 85(10), 1674–1682. LMS log data matched student-by-student to questionnaire responses. Verbatim: “Of the 104 students with data from the LMS, 40% of students watched more than half of the videos… LMS data showed 14 (13%) students did not watch any videos,” against 71% who self-reported watching more than half. PubMed 33942890 (free abstract) · publisher version (paywalled)
- [8] Gutiérrez-González, R., Zamarron, A., & Royuela, A. (2024). “Video-based lecture engagement in a flipped classroom environment.” BMC Medical Education, 24, 1218. Platform analytics across 51 videos, 109 students. DOI: 10.1186/s12909-024-06228-x. link.springer.com/article/10.1186/s12909-024-06228-x
- [9] Maloney, S., Axelsen, M., Galligan, L., Turner, J., Redmond, P., Brown, A., Basson, M., & Lawrence, J. (2022). “Using LMS Log Data to Explore Student Engagement with Coursework Videos.” Online Learning, 26(4), 399–423. 77 videos, ~1,800 students. Verbatim: “Of those who did initiate a play, 43% watched the video to the end.” DOI: 10.24059/olj.v26i4.2998. olj.onlinelearningconsortium.org — article page · full text PDF
- [10] Nielsen, K. L. (2020). “Students' video viewing habits during a flipped classroom course in engineering mathematics.” Research in Learning Technology, 28, 2404. YouTube Analytics logs; 118 students, 104 videos. Verbatim: “The videos had an average view percentage… of 57% (SD = 14%).” DOI: 10.25304/rlt.v28.2404. journal.alt.ac.uk — article page · full text PDF
- [11] Guo, P. J., Kim, J., & Rubin, R. (2014). “How video production affects student engagement: An empirical study of MOOC videos.” Proceedings of the First ACM Conference on Learning @ Scale (L@S '14), 41–50. 6,902,358 viewing sessions across four edX courses. DOI: 10.1145/2556325.2566239. up.csail.mit.edu/other-pubs/las2014-pguo-engagement.pdf
- [12] Giunchiglia, F., Zeni, M., Gobbi, E., Bignotti, E., & Bison, I. (2018). “Mobile social media usage and academic performance.” Computers in Human Behavior, 82, 177–185 (SmartUnitn project, University of Trento; n = 72, correlational). arxiv.org/abs/2004.01392
- [13] Hyperoptic / OnePoll (Sept 2019). Poll of 1,000 UK university students. Breakdown: 54h47m total online, of which 13h56m university work, 14h36m social media, 10h55m streaming, 6h16m gaming, 3h44m shopping, 3h14m video chat, 2h06m adult content. hyperoptic.com — press release · full breakdown as reported
- [14] Gallup (2023). “Teens Spend Average of 4.8 Hours on Social Media Per Day.” Familial & Adolescent Health Survey, 1,591 US teens; self-reported. news.gallup.com/poll/512576/teens-spend-average-hours-social-media-per-day.aspx
- [15] American Academy of Pediatrics, Center of Excellence on Social Media and Youth Mental Health. “Effects of Screen Time on Academic Performance and Mental Health” — see the “Multitasking” section. The same page notes that research has found small or no associations between overall screen media use and academic performance. aap.org — Effects of Screen Time on Academic Performance and Mental Health
- [16] Rosen, L. D., Carrier, L. M., & Cheever, N. A. (2013). “Facebook and texting made me do it: Media-induced task-switching while studying.” Computers in Human Behavior, 29(3), 948–958. 263 middle-school, high-school and university students observed for a 15-minute study window at home. Verbatim: “Participants averaged less than six minutes on task prior to switching most often due to technological distractions including social media, texting and preference for task-switching.” DOI: 10.1016/j.chb.2012.12.001. doi.org/10.1016/j.chb.2012.12.001 · publisher version (paywalled)
- [17] Patanwala, A. E., Erstad, B. L., & Murphy, J. E. (2017). “Student use of flipped classroom videos in a therapeutics course.” Currents in Pharmacy Teaching and Learning, 9(1), 50–54. Logged view time; 100 PharmD students. Verbatim: “Median video view time was greater when a quiz was required [80 minutes (IQR: 38–114) versus 69 minutes (IQR: 3–105), p < 0.001].” doi.org/10.1016/j.cptl.2016.08.043 · PubMed 29180154 · University of Arizona repository
- [18] Lee, H., Ahn, H., Nguyen, T. G., Choi, S.-W., & Kim, D. J. (2017). “Comparing the Self-Report and Measured Smartphone Usage of College Students: A Pilot Study.” Psychiatry Investigation, 14(2), 198–204. Pilot study, n = 35. Verbatim: “The unconscious users underestimate their usage time by 40%, in spite of 15% more use in the actual usage” — the figure applies to that subgroup, not all users. DOI: 10.4306/pi.2017.14.2.198. psychiatryinvestigation.org (free full text) · PubMed Central PMC5355019
- [19] Kung Fu Quiz — product site and sensAI feature documentation. kungfuquiz.com · sensAI AI video quiz generator · question types (help centre)