Start free trial
Trending New Genres Authors
EnglishEnglish
EspañolSpanish
简体中文Chinese
繁體中文Chinese (Traditional)
FrançaisFrench
DeutschGerman
日本語Japanese
PortuguêsPortuguese
ItalianoItalian
한국어Korean
РусскийRussian
NederlandsDutch
العربيةArabic
PolskiPolish
हिन्दीHindi
Tiếng ViệtVietnamese
SvenskaSwedish
ΕλληνικάGreek
TürkçeTurkish
ไทยThai
ČeštinaCzech
RomânăRomanian
MagyarHungarian
УкраїнськаUkrainian
IndonesiaIndonesian
DanskDanish
SuomiFinnish
БългарскиBulgarian
עבריתHebrew
NorskNorwegian
HrvatskiCroatian
CatalàCatalan
SlovenčinaSlovak
LietuviųLithuanian
SlovenščinaSlovenian
СрпскиSerbian
EestiEstonian
LatviešuLatvian
فارسیPersian
മലയാളംMalayalam
தமிழ்Tamil
اردوUrdu
Searching...
SoBrief
Failure to Disrupt

Failure to Disrupt

Why Technology Alone Can’t Transform Education
Why do education technologies keep falling short, and what makes a tool genuinely useful?
by Justin Reich 2020 316 pages
4.23
252 ratings
Listen
Immersive
V2.3
Amazon Kindle Audible
Try Full Access for 3 Days
Unlock listening & more!
Continue

Key Takeaways

Treat education technology as a system component, not a system replacement.

Hype meets school life. Rainbow Loom seemed like a screen-free toy, until children used online tutorials to learn intricate designs. One bracelet video drew tens of millions of views, showing how naturally technology can support informal learning. Formal schools are different: they also provide relationships, meals, supervision, and social services, none of which a platform can simply reproduce. Predictions that online courses would replace half of secondary-school classes or transform universities gave way to more modest uses. Khan Academy, for example, settled into a supplemental role alongside ordinary classroom teaching, a model with precedents going back decades.

Change is cumulative. Reich’s central claim is not that educational technology is useless, but that tools work only within systems of people, schedules, incentives, and institutions. A useful technology may improve a particular practice without remaking the whole institution.

Analysis

The distinction between a tool and the system around it resembles a principle in complex-systems research: changing one component rarely produces a predictable change in the whole, because surrounding relationships shape the result. Schools are not simply delivery channels for lessons. They coordinate care, authority, identity, and public obligations. That makes sweeping replacement claims especially implausible. Still, institutional durability should not be romanticized. A school’s routines can preserve inequity as well as stability. The practical test is therefore not whether a tool disrupts the institution, but whether educators can use it to improve a valuable function without weakening others.

Classify who guides the next step before judging a learning platform.

Three routes through learning. Learning at scale means designing a learning experience for many people with relatively few experts available to guide them. Reich distinguishes three forms by asking who determines what learners do next. In instructor-guided systems, such as most MOOCs, a teacher or designer sets the sequence. In algorithm-guided systems, such as adaptive math software, a learner’s answers influence the next activity. In peer-guided systems, such as Scratch, learners navigate projects and contributions made by a community.

Different strengths follow. Instructor-guided platforms distribute prepared instruction; adaptive tutors personalize practice within measurable subjects; peer networks support exploration and shared creation. The categories are useful for predicting likely strengths and limits, but some tools blend them. A learning game, for instance, may combine a designed sequence, automatic feedback, and collaboration among players.

Analysis

This taxonomy is valuable because it shifts attention from flashy product features to the allocation of authority. Who controls sequence, pace, and goals determines what kinds of learning the platform can support. Similar questions appear in organizational design: centralized systems coordinate reliably, while decentralized systems can adapt locally but demand more judgment from participants. The categories should be treated as diagnostic lenses, not rigid boxes. Hybrid tools may combine their advantages, but they can also inherit several kinds of friction at once. Evaluators should inspect actual learner experience rather than rely on a product’s label.

Open enrollment does not create equal access or reliable completion.

An open door still has steps. MOOCs attracted huge numbers of registrants, but commonly only about 5 percent completed a course. Among learners who said they intended to finish, roughly 20 percent did; completion was higher among those paying for verified certificates. Participation was also socially skewed: around 80 percent of learners in leading US platforms came from affluent countries, and one Coursera study found that about 80 percent already held a bachelor’s degree. Self-regulated learning, the skills of setting goals, managing time, monitoring progress, and seeking help, strongly predicted persistence.

The market found a niche. MOOCs became more useful as routes to professional credentials and additional degrees than as broad entryways to college. Georgia Tech’s lower-cost online computer science master’s, for example, reached working professionals who might not have pursued a campus degree. Access to a course is not the same as support to complete it.

Analysis

The MOOC experience illustrates a distinction between formal availability and effective opportunity. A course can be technically open while demanding spare time, stable internet, academic confidence, and familiarity with independent study. This resembles the capabilities approach in social policy, which judges opportunity by what people can actually do, not by whether a resource exists. Low completion rates alone can mislead, since some registrants only sample material. But the gap between enrollment and completion matters when providers claim to widen access to credentials. Support, advising, and recognized pathways are part of access, not optional extras.

Use adaptive tutors selectively, especially where math practice needs feedback.

A long-running idea. Adaptive tutors adjust practice in response to student answers. Many rely on item response theory, a statistical method that estimates how difficult each question is and uses performance to select a suitable next question. The underlying approach is not a recent breakthrough: computer-assisted instruction has been developed for decades. Research on reading software has generally found little average benefit, while results for math are mixed but increasingly promising.

Fit matters more than spectacle. In a large study, Cognitive Tutor produced no gains in its first year, then a positive effect of about 0.2 standard deviations in the second year. ASSISTments, a simpler online homework tool that gave students immediate feedback and teachers reports on common errors, found a similar gain, including benefits for lower-achieving students. These results support selective trials, not claims that software can replace teachers or transform every subject.

Analysis

The contrast between a complex tutor and a comparatively simple homework tool is a reminder to test the active ingredient, not the sophistication of the product. Immediate feedback, better visibility into student errors, and time for targeted review may explain gains as much as algorithmic personalization. The reported improvement also depended on implementation: teachers needed time to learn how to use the tools, and some systems worked best when students could progress at different rates. A school should compare the software with realistic alternatives, including teacher development and human tutoring, rather than with an imaginary unchanged classroom.

Build peer learning around interests, but design for safe entry.

Learning through networks. Peer-guided learning lets participants find resources, projects, and collaborators rather than follow one fixed lesson path. Early connectivist MOOCs asked learners to publish on personal blogs and connect through shared feeds and hashtags. Scratch applies a related idea to programming: learners can inspect, remix, discuss, and build on one another’s projects. The approach can turn an interest into a path toward skill and belonging, as Rainbow Loom enthusiasts did through shared tutorials and designs.

Openness has costs. Decentralized courses often bewildered newcomers, who had to learn how to participate before learning the subject. Online communities can also teach harmful ideas. Reich discusses how extremist groups use networks, media, and recommendation systems to recruit and deepen commitment. Peer learning is therefore not automatically inclusive or beneficial. Good design must make participation understandable and consider who moderates the community and what values it amplifies.

Analysis

Peer networks offer something centralized instruction struggles to reproduce: learners can find models, feedback, and a sense of membership around a personally meaningful goal. Communities of practice research helps explain why observing and gradually contributing can build expertise as well as identity. Yet network access is not neutral. The same routes that carry creative tutorials can spread propaganda, and the freedom to explore can overwhelm novices. This is not an argument for eliminating openness. It is a case for treating orientation, moderation, and inclusion as core parts of a learning environment rather than as cleanup work.

Expect games to teach game-specific skills unless transfer is deliberately built.

Practice does not travel automatically. Transfer is the use of learning in a new situation. Near transfer occurs when the new task resembles the original; far transfer involves substantially different tasks. Studies of chess and brain-training games show that practice often improves performance on similar activities without producing broad gains in general reasoning. Lumosity, which claimed wide-ranging benefits from its games, paid a $2 million settlement after the Federal Trade Commission found its advertising unsupported.

Design for the bridge. Games can still support learning. Meta-analyses found modest gains in knowledge and motivation, especially with repeated play and richer designs. Duolingo uses spaced repetition, scheduling review over time, and one study found that 34 hours of use taught material comparable to a first semester of college Spanish. But vocabulary practice is not literary fluency. To encourage transfer, connect game activity explicitly to real tasks, explanations, or conversations.

Analysis

Transfer research disciplines the temptation to treat engagement as proof of broad educational value. A player’s persistence, strategy, or memory may improve within a game because the environment supplies familiar cues and clear goals. Those abilities may not appear in a different context without deliberate practice in applying them. This limitation does not make games frivolous: narrow expertise can be worthwhile, and motivation can keep learners practicing. The sharper question is what specific capability is being developed, and whether the design provides a bridge from the game’s rules to the real-world task being claimed.

Scale teacher communities, not just familiar apps and student accounts.

Familiar tools travel easily. The curse of the familiar describes a dilemma: technologies that resemble existing classroom routines are easier to adopt, but may only digitize old practices; genuinely novel approaches can confuse learners or be reshaped into conventional routines. Digital flash cards are easy to use, but they do not by themselves change teaching. Scratch, designed for open-ended creative projects, is often turned into a step-by-step assignment when introduced in schools.

Practice changes socially. Teachers typically learn new approaches from other teachers. Scratch and Desmos therefore invest not only in software, but in workshops, educator networks, and examples that help teachers understand the underlying pedagogy. Desmos delayed adding common platform features until they could be shaped to support its vision for mathematical discussion. Adoption figures show reach, but a community of educators can help sustain changes in classroom practice.

Analysis

The account connects educational change to implementation science, which emphasizes that an intervention is not a detachable package. Its effects depend on local capacity, professional learning, leadership, and adaptation. A community can translate abstract principles into credible classroom routines and help newcomers recover when an experiment falters. This also exposes a tension: communities take time and funding, while technology markets reward rapid growth in accounts and usage. Educator networks are not a guaranteed engine of reform, but without them, a platform’s design intentions have little power against the daily pressures that shape teaching.

Free tools can widen gaps when support and agency are unequal.

The Matthew effect in edtech. This term describes how new learning resources often benefit people who already have the money, skills, support, and infrastructure to use them. Technology access is only the first divide. A second divide concerns how it is used: students in affluent schools are more likely to use technology for creative work with adult guidance, while students in less affluent schools may be assigned routine practice with less support. Studies of wikis, MOOCs, and classrooms found versions of this pattern.

Design around real conditions. Equity-oriented programs involve learners and communities in design, connect home and school, make room for culturally varied interests, and measure outcomes for different subgroups. Family Creative Learning, for example, brings parents and children together to create with Scratch. Free textbooks can also matter greatly when a textbook’s price is a serious barrier. Free access helps most when it addresses a meaningful constraint and comes with relevant support.

Analysis

This argument challenges the assumption that reducing price automatically redistributes opportunity. In economics, a resource has value only when people can convert it into a useful outcome; conversion depends on time, skills, infrastructure, and social support. The distinction also warns against designing for an imagined universal learner who is detached from family, place, and history. Yet targeted programs need careful evaluation too. They can misread a community or treat groups as uniform. Co-design and subgroup data help correct those errors, provided measurement does not become a new form of surveillance.

Do not let easy grading define what students should learn.

Automated grading narrows the target. Computers can reliably score many tasks with structured answers, such as arithmetic, multiple-choice questions, and code that must meet defined requirements. They struggle to judge open-ended reasoning, persuasive writing, and other performances whose quality depends on meaning and context. The trap of routine assessment is that schools may emphasize what machines can measure, even when those tasks are not the most valuable human abilities to develop.

Tests are partial samples. A so-called math test may measure computation while missing problem framing, modeling, and explanation. Automated essay scoring can match patterns in writing to essays graded by people, but it may reward surface features rather than sound argument. A deliberately nonsensical essay once received a high score from an automated system. New assessment tools, such as software that evaluates students’ drawn calculus graphs, can widen the sample, but cannot make any test a complete measure of expertise.

Analysis

Measurement shapes behavior: when a system rewards a narrow proxy, people redirect effort toward the proxy. This is a familiar problem in management and public policy, not unique to education. Automated assessment can still be useful, particularly for quick feedback and routine checks, but its speed should not be confused with validity. The deeper issue is not whether software can grade, but whether the task being graded represents the capability schools claim to value. Human judgment remains costly, yet that cost may be necessary for assessing work that depends on interpretation, audience, and context.

Treat student data as risky material, not free fuel for innovation.

Useful data carry obligations. Online platforms record clicks, answers, video pauses, and other activity. Researchers can use these records to improve instruction, but large datasets can also expose private information, especially when combined with other sources. Contextual integrity is the idea that appropriate data use depends on the expectations and relationships surrounding the information. A student may answer a survey believing it is anonymous, while administrators combine it with residence data for decisions the student did not anticipate.

Experiments need legitimacy. Randomized tests can reveal whether a new approach helps, instead of mistaking engagement or coincidence for learning. But assigning students to different experiences without clear communication raises concerns about consent and power. Reich argues for transparent methods, public engagement, limits on vendor use, and sensible deletion or protection practices. Data should be collected for a defined purpose, not stockpiled on the assumption that future analysis will justify it.

Analysis

The chapter’s central tension resembles the governance of medical research: evidence can improve practice, but participants are not merely raw material for discovery. Digital systems make experimentation easier and can detect small effects, yet ease of experimentation does not settle whether a study is fair. Context matters because students often cannot refuse a platform required by their school. Stronger safeguards need more than consent forms. They require clear limits on collection, retention, access, and secondary use, along with public explanations of what research is intended to accomplish.

Interrogate every new platform's novelty, pedagogy, evidence, and fit.

Ask before adopting. Reich’s practical test for a learning technology begins with four questions:
1. What is genuinely new, compared with earlier tools?
2. Who sets the learning sequence: an instructor, an algorithm, or peers?
3. Does the design transmit established content or invite exploration and creation?
4. What existing technologies and routines does it depend on?

Then test the local match. Examine research on similar systems, which learners benefit, what teachers must change, and what the technology can actually assess. Consider infrastructure, staff time, student support, and the opportunity cost of choosing this project over another. A promising average result does not guarantee success in a particular school. The aim is not to predict a revolution, but to identify a useful contribution and improve it alongside the people who teach and learn.

Analysis

This checklist replaces technological mystique with comparative judgment. Asking what is new forces a product to be evaluated against its predecessors; asking who guides learning exposes its pedagogical assumptions; asking about assessment and local fit makes hidden tradeoffs visible. The approach resembles due diligence in public procurement, where total costs include training, maintenance, and the consequences of failure, not just the purchase price. Its limit is that historical analogies cannot forecast every innovation. They can, however, set a more credible starting point than a vendor’s promise of inevitable transformation.

Analysis

The book’s most important move is to replace the question, “Will technology transform education?” with a more useful one: “Under what conditions does this tool improve a particular learning practice, for particular people?” Its three learning-at-scale categories make that shift concrete. Instructor-led courses, adaptive tutors, and peer networks distribute control differently, and therefore demand different learner capacities, support structures, assessments, and institutional changes.

The four recurring dilemmas deepen the analysis. Familiar tools spread because they fit existing routines, while ambitious alternatives can be confusing or domesticated by schools. Open access often advantages learners already equipped to benefit. Automated assessment favors structured tasks, which can narrow the curriculum toward measurable routines. Finally, data and experimentation can improve design while creating privacy and autonomy risks. Together, these mechanisms explain why technical reach does not equal educational impact.

The argument is strongest when it combines historical comparison with evidence about implementation. The MOOC forecasts, the uneven record of adaptive tutoring, and examples such as ASSISTments show why technologies should be judged by outcomes and context rather than novelty. The emphasis on educator communities also offers a plausible mechanism for change: teachers adapt tools, interpret evidence, and establish practices that software alone cannot supply.

One limitation is that the book’s caution can make incremental change sound more dependable than it is. Small improvements may accumulate, but they can also leave intact the unequal funding, tracking, and social obligations that shape schools. Nor should average effects erase exceptional uses or distinctive local needs. The strongest reading is neither technological optimism nor reflexive skepticism. It is disciplined conditional optimism: identify a concrete need, examine who benefits and who bears costs, fund the human support that makes adoption possible, and revise the intervention when evidence or community experience contradicts the plan. Technology is most valuable not as a shortcut around educational institutions, but as one resource in the harder work of changing them.

Summary in 30 Seconds
Digital tools improve particular learning activities, but schools also provide relationships, support, and services that software cannot replace. Online courses often attract far more registrants than completers: about 5 percent finish MOOCs, with persistence tied to self-management and support. Adaptive math tools can help when feedback fits the task, while practice rarely transfers automatically to unrelated skills. Teacher communities, equitable access, sound assessment, and careful data safeguards shape whether a tool benefits learners.
Contains spoilers

Last updated:

Report Issue

Review Summary

4.23 out of 5
Average of 252 ratings from Goodreads and Amazon.

Failure to Disrupt: Why Technology Alone Can't Transform Education receives mostly positive reviews, with readers praising its well-researched and balanced approach to educational technology. Many appreciate Reich's critical analysis of edtech hype and his insights into why technology has failed to revolutionize education. Readers find the book informative, thought-provoking, and relevant, especially in light of recent pandemic-driven online learning experiences. Some criticize the book for being too pessimistic or lacking in proposals for improvement, but overall, it's considered a valuable read for those interested in education and technology.

Your rating:
4.55
149 ratings
Want to read the full book?

Glossary

learning at scale

Learning for many people

A learning environment designed to serve many learners while relying on relatively few experts for direct guidance. Such environments may be online, blended with classrooms, or embedded in large courses. The central challenge is to provide useful learning experiences across many participants without assuming that technology alone can supply the support, judgment, and social relationships learners need.

instructor-guided learning at scale

Instructor-sequenced learning

A large learning environment in which an instructor or instructional designer determines the sequence of activities and materials. MOOCs are a common example: learners follow an arranged course, often using recorded instruction and automated quizzes. This approach can distribute prepared content widely, but learners generally need to manage their own pace and seek support when the platform offers little personal guidance.

algorithm-guided learning at scale

Algorithm-sequenced learning

A learning environment in which software uses a learner’s performance to select subsequent problems or activities. Adaptive tutors are a common example. The approach can personalize practice when responses are measurable and content can be divided into assessable steps. Its usefulness depends on the quality of the assessment, the subject area, and the human support around the software.

peer-guided learning at scale

Peer-networked learning

A learning environment in which learners navigate resources, projects, and interactions created by a community rather than following a single prescribed sequence. Connectivist MOOCs and the Scratch community illustrate this approach. It can support interest-driven exploration, collaboration, and identity development, but newcomers may need help understanding how to participate, and open networks can circulate harmful as well as useful material.

curse of the familiar

Adoption without transformation

The dilemma that familiar educational technologies are easier to adopt because they resemble established classroom routines, but often reproduce those routines rather than improve them substantially. More novel designs may offer different possibilities, yet learners and teachers can find them confusing or reshape them to fit familiar practices. Meaningful change therefore requires support for educators and institutions, not just new software.

edtech Matthew effect

Advantages compound through technology

The tendency for educational technologies to benefit learners who already have resources, skills, support, and reliable infrastructure. Free or widely available tools can still disproportionately help affluent or academically experienced learners. Inequality can arise not only from unequal device access, but also from differences in how technology is used and how much adult guidance accompanies it.

trap of routine assessment

Teaching what machines score

The risk that automated assessment privileges routine tasks because computers can score them reliably, while complex communication, interpretation, and unstructured problem solving remain harder to assess. If curriculum follows what can be measured cheaply at scale, important human capabilities may receive less instructional attention, even when those capabilities are central to advanced work and civic life.

contextual integrity

Privacy shaped by context

A framework for judging whether information use fits the expectations and relationships in which the information was shared. Data use may feel inappropriate even when no explicit promise was broken, particularly when information is combined with other datasets or used for a new purpose. The framework asks whose expectations matter and whether participants understand how information may be used.

self-regulated learning

Managing one’s own learning

The ability to set learning goals, organize time, monitor progress, seek help, and persist without continuous external direction. These skills are especially important in self-paced online courses, where instructors may have little contact with individual learners. The book notes that they are often developed through previous experiences in formal education, which can disadvantage people with less educational opportunity.

item response theory

Modeling question difficulty

A statistical approach that estimates how likely learners at different ability levels are to answer a particular question correctly. The estimates allow test makers to compare items and enable adaptive systems to select questions based on previous performance. It models an item’s difficulty and response patterns; it does not itself determine what knowledge is valuable or provide a complete account of learning.

transfer

Applying learning elsewhere

The use of knowledge or skill acquired in one situation in a different situation. Near transfer involves a new task resembling the original practice; far transfer involves a more substantial change in context. Evidence discussed in the book suggests that improvement often remains close to the practiced domain, so claims of broad real-world benefits require careful testing.

FAQ

1. What is Failure to Disrupt: Why Technology Alone Can’t Transform Education by Justin Reich about?

  • Critical examination of edtech: The book analyzes why technology alone has not revolutionized education, focusing on the persistent gap between edtech promises and real-world outcomes.
  • Complexity of education systems: Reich argues that educational change is shaped by social, cultural, and institutional factors, making simple technological solutions insufficient.
  • Genres of learning at scale: The book categorizes large-scale learning technologies into instructor-guided, algorithm-guided, and peer-guided models, exploring their strengths and limitations.
  • Equity and systemic change: Reich emphasizes that technology often reproduces or worsens inequalities unless integrated with thoughtful, systemic reforms.

2. Why should educators, policymakers, and technologists read Failure to Disrupt by Justin Reich?

  • Debunking edtech myths: The book provides a research-based critique of common misconceptions, helping readers avoid repeating past mistakes in educational technology.
  • Guidance for realistic implementation: Reich offers a "tinkerer’s guide" for incremental, evidence-based improvements rather than expecting disruptive transformation.
  • Equity and ethical focus: The book highlights the importance of designing technology with equity, privacy, and ethical considerations at the forefront.
  • Practical design principles: Reich and Mizuko Ito propose actionable principles for developing edtech that serves diverse learners and communities.

3. What are the key takeaways and future directions from Failure to Disrupt by Justin Reich?

  • Incremental change over disruption: Reich stresses that meaningful educational improvement comes from long-term, collaborative tinkering, not sudden technological revolutions.
  • Integration with human systems: Technology must be thoughtfully embedded within complex educational ecologies, considering teachers, students, and institutions.
  • Equity and community engagement: Addressing digital divides and cultural relevance is essential; technology alone cannot democratize education.
  • Ethical data use: Responsible stewardship of student data and transparent research practices are critical for building trust and maximizing benefits.

4. What are the three genres of learning at scale described in Failure to Disrupt by Justin Reich?

  • Instructor-guided learning: Includes MOOCs and online courses where instructors sequence content and assessments, effective mainly for self-regulated learners in structured domains.
  • Algorithm-guided learning: Uses adaptive tutors and computer-assisted instruction to personalize learning paths, excelling in routine, structured tasks like math and early reading.
  • Peer-guided learning: Emphasizes community, collaboration, and creativity through platforms like Scratch and connectivist MOOCs, supporting social learning but challenging to integrate into formal education.
  • Distinct pedagogical models: Each genre has unique strengths, limitations, and challenges for scaling and equity.

5. How does Justin Reich in Failure to Disrupt explain why technology alone fails to transform education?

  • Conservatism of schools: Schools are durable institutions where new technologies are often absorbed into existing practices rather than driving transformation.
  • Complexity and multiple stakeholders: Education involves conflicting purposes and diverse actors, making rapid, tech-driven change difficult.
  • Need for human support: Effective technology integration requires professional development, peer networks, and systemic backing.
  • Domestication of innovation: Technologies are often adapted to fit familiar routines, limiting their transformative potential.

6. What is the "curse of the familiar" in education technology, according to Failure to Disrupt?

  • Definition and analogy: The curse of the familiar refers to the tendency for new technologies to mimic traditional classroom practices, making them easy to adopt but unlikely to drive significant improvement.
  • Skeuomorphism in edtech: Like digital tools that resemble their analog counterparts, education technologies often replicate worksheets or lectures, stifling innovation.
  • Barrier to transformation: Truly novel technologies can confuse or overwhelm educators and learners, slowing adoption and limiting impact.
  • Balancing act: Successful innovation requires finding a middle ground between familiarity and new possibilities.

7. What is the "edtech Matthew effect" and how does it impact educational equity in Failure to Disrupt?

  • Definition of the effect: The edtech Matthew effect describes how new educational technologies disproportionately benefit already-advantaged learners, widening existing inequalities.
  • Evidence from research: Affluent students use technology for creative, higher-order tasks, while marginalized students often have limited access and use tech for rote learning.
  • Implications for democratization: Free or open technologies do not automatically level the playing field; social, cultural, and infrastructural barriers must be addressed.
  • Design for equity: Reich and Ito propose four design principles to guide equitable edtech development, focusing on shared purpose, support alignment, relevance, and targeted measurement.

8. How does Failure to Disrupt critique the promises of disruptive innovation in education technology?

  • Disruptive innovation theory challenged: Reich argues that education systems tend to domesticate new technologies, integrating them into existing routines rather than enabling radical change.
  • Historical perspective: The book traces a century of failed predictions, from motion pictures to MOOCs, showing that educational change is incremental and complex.
  • Skepticism of grand claims: Reich encourages stakeholders to be wary of hype and focus on realistic, evidence-based improvements.
  • Community over heroism: Systemic change requires collaborative efforts among educators, researchers, and designers, not lone inventors.

9. What is the "trap of routine assessment" in education technology, as discussed in Failure to Disrupt?

  • Definition of the trap: Computers can only reliably assess routine, structured tasks, leading educational technologies to focus on what machines can grade.
  • Misalignment with workforce needs: As routine tasks are automated in the labor market, education risks overemphasizing computational skills at the expense of complex communication and problem-solving.
  • Attempts to escape: Advances like automated essay scoring and stealth assessment show promise but remain limited and incremental.
  • Human judgment required: Meaningful assessment of complex skills still depends on human evaluators, constraining the reach of automated systems.

10. How does Failure to Disrupt evaluate the impact of MOOCs, adaptive tutors, and peer-guided learning?

  • MOOCs: Initially hyped as disruptive, MOOCs mainly serve already-educated, affluent learners and have been absorbed as supplements or online program managers, with low completion rates.
  • Adaptive tutors: Systems like ASSISTments and Cognitive Tutor show modest positive effects in math, especially when implemented with fidelity, but are limited to domains suited for autograding.
  • Peer-guided learning: Platforms like Scratch foster creativity and engagement through community, but measuring effectiveness is complex and integration into formal schooling is challenging.
  • Limited transformative impact: None of these models alone have fundamentally transformed education; success depends on thoughtful integration and support.

11. What are some successful learning technologies or programs highlighted in Failure to Disrupt by Justin Reich?

  • ASSISTments and Cognitive Tutors: These adaptive systems have demonstrated positive effects on math learning, especially for struggling students, by supporting personalized instruction.
  • Scratch programming community: Scratch has engaged millions of young learners worldwide, promoting creativity and computational thinking through peer-guided learning.
  • MOOC-based master’s programs: Georgia Tech’s online master’s in computer science is cited as a scalable, effective model for serving working professionals.
  • Equity-focused initiatives: Programs like Tech Goes Home and Family Creative Learning address access, cultural relevance, and targeted support for marginalized learners.

12. What are the best quotes from Failure to Disrupt by Justin Reich and what do they mean?

  • “People who do stuff do more stuff, and people who do stuff do better than people who don’t do stuff.” Reich’s Law highlights that engagement correlates with success, but cautions against simplistic interpretations of learning analytics.
  • “The trap of routine assessment is that computers can only assess what computers themselves can do, so that’s what we teach students.” This quote encapsulates the risk of narrowing education to what machines can grade, potentially neglecting vital human skills.
  • “Educational technology is implicated in perpetuating inequality.” Reich warns that without intentional design and policy, edtech often reinforces existing disparities.
  • “We need villages, not heroes.” This phrase summarizes the book’s call for collaborative, community-driven educational change rather than reliance on individual innovators.

About the Author

Justin Reich is a professor at MIT and the director of the Teaching Systems Lab. He was the first full-time researcher hired to work on the infrastructure for EdX Massive Open Online Courses (MOOCs). Reich's expertise in educational technology is evident in his book, which draws from his firsthand experience and extensive research in the field. He hosts the 'TeachLab' podcast and is known for his balanced and critical approach to analyzing the impact of technology on education. Reich's work focuses on understanding why technological innovations have not led to the revolutionary changes in education that were often predicted, and he advocates for a more realistic and incremental approach to integrating technology in learning environments.

Download PDF

To save this Failure to Disrupt summary for later, download the free PDF. You can print it out, or read offline at your convenience.
Download PDF
File size: 0.47 MB     Pages: 23

Download EPUB

To read this Failure to Disrupt summary on your e-reader device or app, download the free EPUB. The .epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.
Download EPUB
File size: 1.45 MB     Pages: 34
Want to read the full book?
Follow
Listen1 mins
Home
Swipe
Library
Get App
Try Full Access for 3 Days
Listen, bookmark, and more
Compare Features Free Pro
📖 Read Summaries
Read unlimited summaries. Free users get 3 per month
🎧 Listen to Summaries
Listen to unlimited summaries in 40 languages
—
❤️ Unlimited Bookmarks
Free users are limited to 4
—
📜 Unlimited History
Free users are limited to 4
—
📥 Unlimited Downloads
Free users are limited to 1
—
Risk-Free Timeline
Today: Get Instant Access
Listen to full summaries of 26,000+ books. That's 12,000+ hours of audio!
Day 2: Trial Reminder
We'll send you a notification that your trial is ending soon.
Day 3: Your subscription begins
You'll be charged on Nov 25,
cancel anytime before.
Consume 2.8× More Books
2.8× more books Listening Reading
Our users love us
1M+ readers
Trustpilot Rating
TrustPilot
4.6 Excellent
This site is a total game-changer. I've been flying through book summaries like never before. Highly, highly recommend.
— Dave G
Worth my money and time, and really well made. I've never seen this quality of summaries on other websites. Very helpful!
— Em
Highly recommended!! Fantastic service. Perfect for those that want a little more than a teaser but not all the intricate details of a full audio book.
— Greg M
Save 62%
Yearly
$119.88 $44.99/year/yr
$3.75/mo
Monthly
$9.99/mo
Start a 3-Day Free Trial
3 days free, then $44.99/year. Cancel anytime.
Unlock a world of fiction & nonfiction books
26,000+ books for the price of 2 books
Read any book in 10 minutes
Discover new books like Tinder
Request any book if it's not summarized
Read more books than anyone you know
#1 app for book lovers
Lifelike & immersive summaries
30-day money-back guarantee
Download summaries in EPUBs or PDFs
Cancel anytime in a few clicks

Why one summary?

Scanner
Find a barcode to scan

⭐
⭐
⭐
We have a special gift for you
Open
38% off
DISCOUNT FOR YOU
⭐
⭐
⭐
$79.99
$49.99/year
Only $4.16 per month
Continue
2 taps to start, super easy to cancel
Settings
General
Widget
Loading...