The landscape of English language proficiency testing has undergone a profound and accelerating transformation over the past decade, driven by technological innovation, shifting institutional expectations, and the growing demand for accessible and affordable assessment options among the millions of international students, skilled migrants, and professionals who need to demonstrate their English ability each year. At the center of this transformation stand two tests that represent fundamentally different philosophies about what language assessment should look like, how it should be administered, and who it should serve. The TOEFL, which stands for Test of English as a Foreign Language, has been the dominant standard in academic English assessment for more than half a century, while the Duolingo English Test has emerged in recent years as a disruptive challenger that has forced institutions and test-takers alike to reconsider long-held assumptions about what a valid and reliable language proficiency test requires.
Understanding the relationship between these two assessments requires appreciating both the historical context that shaped the TOEFL into the institution it has become and the technological and social forces that created the conditions for the Duolingo English Test to gain the kind of acceptance that would have seemed implausible just a decade ago. The debate between these two tests is not simply a comparison of two products competing for market share but a reflection of deeper questions about the nature of language proficiency itself, the meaning of testing validity and reliability in an era of sophisticated adaptive assessment technology, and the responsibilities that institutions bear toward the diverse populations of test-takers who must navigate these systems to access educational and professional opportunities. Both tests have genuine strengths and genuine limitations, and a clear-eyed assessment of both is essential for anyone seeking to make an informed decision about which to pursue.
TOEFL Established Its Legacy
The TOEFL examination was developed in the early 1960s through a collaborative effort between major educational organizations in the United States that recognized the need for a standardized mechanism for evaluating the English proficiency of international students seeking admission to American universities. When the test was first administered in 1964, it represented a significant advance in the systematic assessment of English language ability, providing admissions offices with a common metric that could be used to compare applicants from vastly different educational backgrounds and native language contexts. Over the subsequent six decades, the TOEFL grew from a North American assessment tool into a globally recognized credential accepted by thousands of institutions in more than 150 countries across every inhabited continent.
The evolution of the TOEFL from its paper-based origins through the computer-based format introduced in the 1990s and into the current internet-based iBT format reflects both the technological changes of the past several decades and the ongoing effort by Educational Testing Service, the organization responsible for administering the TOEFL, to maintain the test’s relevance and validity in a changing assessment landscape. The iBT format introduced in 2005 represented a particularly significant shift because it integrated all four language skills, reading, listening, speaking, and writing, into a single integrated assessment that attempted to measure English proficiency in ways that more closely approximated the actual language demands of academic study. This integrated approach has remained the defining structural characteristic of the TOEFL iBT and continues to distinguish it from assessments that evaluate language skills in relative isolation from one another.
Duolingo Disrupted Old Assumptions
The Duolingo English Test entered the language proficiency testing market in 2016 as a product of Duolingo, the Pittsburgh-based language learning company best known for its gamified language learning application that had attracted hundreds of millions of users worldwide. The test was designed from its inception to address what its creators identified as fundamental accessibility problems with existing high-stakes English language tests, particularly the cost barriers and logistical challenges associated with testing center-based examinations that required test-takers to travel to designated facilities, pay substantial examination fees, and wait weeks for results. By delivering the test entirely online through a standard computer with a webcam, at a cost dramatically lower than competing assessments, and with results available within 48 hours, the Duolingo English Test aimed to democratize access to language proficiency certification in ways that existing tests had failed to achieve.
The initial reception of the Duolingo English Test within the higher education community was cautious, with many admissions offices citing concerns about test security, validity research, and the unconventional assessment approach as reasons to maintain their existing requirements for TOEFL or IELTS scores. The COVID-19 pandemic, which forced the closure of testing centers worldwide in early 2020 and created an urgent need for remote assessment alternatives, proved to be a pivotal moment for the Duolingo English Test. As institutions scrambled to find ways to continue evaluating international applicants during a period when traditional testing was impossible, many accepted the Duolingo English Test on an emergency basis and discovered that it provided useful and reliable information about applicant English proficiency. Many of these institutions retained their acceptance of the test after testing centers reopened, contributing to a rapid expansion of the test’s acceptance network that continues to this day.
Test Format Differences Matter
The structural differences between the TOEFL iBT and the Duolingo English Test reflect the distinct assessment philosophies and technical approaches that underlie each examination, and these differences have practical implications for test-takers in terms of preparation strategies, test-taking experience, and the kinds of English ability that each test most effectively measures. The TOEFL iBT is a lengthy examination that requires approximately three hours to complete and consists of discrete sections assessing reading, listening, speaking, and writing ability through a combination of passage-based questions, audio recordings, integrated tasks that require synthesis of multiple inputs, and extended writing responses. The length and format of the TOEFL are designed to provide comprehensive coverage of the academic English skills that research suggests are most predictive of success in university-level study.
The Duolingo English Test takes a fundamentally different structural approach, using an adaptive algorithm to select questions dynamically during the examination in response to the test-taker’s performance, and compressing the full assessment into approximately one hour. The test includes a mix of question types that assess reading, listening, writing, and speaking through relatively brief tasks, followed by an unscored video interview and writing sample that are sent to institutions alongside the official score. The adaptive nature of the Duolingo English Test means that test-takers encounter questions calibrated to their demonstrated ability level rather than working through a fixed sequence of items, which the test’s designers argue produces more precise ability estimates in less time than fixed-form tests can achieve. This efficiency is one of the Duolingo English Test’s most frequently cited advantages and is a central part of its appeal for test-takers who find the time and logistical demands of the TOEFL burdensome.
Institutional Acceptance Comparison Remains
The question of institutional acceptance is often the most practically decisive factor in a test-taker’s choice between the TOEFL and the Duolingo English Test, because regardless of a test’s technical quality or convenience, it serves no purpose for a specific applicant if the institutions they are applying to do not accept it as evidence of English proficiency. The TOEFL maintains an acceptance network that includes thousands of universities, professional certification bodies, and government agencies worldwide that have used it as a standard for decades and whose policies, score benchmarks, and evaluation frameworks are built around familiarity with TOEFL score interpretation. For test-takers applying to institutions in countries where TOEFL acceptance is essentially universal, such as the United States, Canada, Australia, and the United Kingdom, the practical risk of the TOEFL not being accepted by a desired institution is extremely low.
The Duolingo English Test’s acceptance network has grown dramatically in recent years and now includes several thousand institutions in more than 150 countries, including many of the world’s most prestigious universities. However, acceptance is not uniform across all institution types or all countries, and test-takers should carefully verify the specific acceptance policies of each institution on their application list before choosing the Duolingo English Test over a more universally accepted alternative. Some institutions accept the Duolingo English Test for undergraduate admissions but not for graduate programs, some accept it conditionally with additional requirements, and some accept it only up to a certain score threshold above which they require TOEFL or IELTS scores for additional validation. Careful individual research into institutional policies is therefore an essential step that cannot be replaced by general guidance about acceptance trends.
Cost and Accessibility Comparison
The cost differential between the TOEFL iBT and the Duolingo English Test is substantial and represents one of the most concrete and tangible differences between the two assessments from the perspective of test-takers who must budget for the examination as part of the already considerable financial burden of international education applications. The TOEFL iBT is priced at approximately 220 dollars in the United States, with prices varying by country and region, and test-takers who need to send scores to multiple institutions must pay additional fees for each score report beyond the four included with the initial registration. The cost of multiple sittings for test-takers who do not achieve their target score on the first attempt adds further to the total financial commitment associated with the TOEFL.
The Duolingo English Test is significantly more affordable, priced at 65 dollars per examination as of recent pricing, with unlimited free score sends included in that fee, meaning that test-takers pay a single flat fee regardless of how many institutions they send their scores to. This pricing structure has particularly significant implications for test-takers who are applying to large numbers of institutions, as is common among students pursuing competitive programs in fields where acceptance rates are low and broad application strategies are advisable. Beyond the direct cost difference, the Duolingo English Test’s at-home delivery eliminates the travel costs and time costs associated with reaching a testing center, which can be substantial for test-takers in regions where testing centers are sparse and may require long-distance travel to access.
Scoring Systems Explained Simply
The TOEFL iBT uses a total score scale ranging from zero to 120, derived from four section scores of zero to 30 each covering reading, listening, speaking, and writing. This four-component structure provides institutions with diagnostic information about a test-taker’s relative strengths and weaknesses across different skill areas, which can be valuable for admissions decisions in programs where specific language skills are particularly important. Many institutions publish minimum score requirements for each section separately rather than only for the total score, reflecting the value they place on skill-specific information in evaluating whether an applicant’s English proficiency profile aligns with the demands of their particular program.
The Duolingo English Test reports a single overall score on a scale from 10 to 160, along with four subscores covering literacy, comprehension, conversation, and production that are intended to provide some skill-area diagnostic information alongside the overall score. The Duolingo English Test also includes percentile information and the unscored video interview and writing sample that institutions can use as qualitative supplements to the numerical score. While the overall score reporting approach of the Duolingo English Test is simpler to interpret than the multi-section TOEFL score report, some admissions professionals have noted that the translation between Duolingo scores and the TOEFL or IELTS scores they have years of experience interpreting requires a period of familiarization that introduces some uncertainty into the evaluation process. Duolingo provides score concordance tables that map its scores to TOEFL and IELTS equivalents to assist institutions in this interpretation process.
Validity Research Ongoing Debate
The validity of a language proficiency test refers to the extent to which it actually measures what it claims to measure and the extent to which scores on the test predict the real-world outcomes, such as academic success in English-medium programs, that institutions care about when using test scores as admissions criteria. The TOEFL has accumulated a substantial validity research base over its decades of use, with numerous independent studies examining the relationship between TOEFL scores and academic performance, the test’s construct validity, and its fairness across different demographic groups. This research base provides institutions with an empirical foundation for confidence in the TOEFL’s ability to provide meaningful and useful information about applicant English proficiency.
The Duolingo English Test’s validity research base is newer and less extensive than the TOEFL’s, which reflects the reality that the test has existed for a much shorter period and has only recently achieved the scale of institutional acceptance that motivates independent validity research by institutions using the scores for high-stakes decisions. Duolingo has published validity research conducted by its own researchers and by independent academics demonstrating correlations between Duolingo English Test scores and TOEFL scores, demonstrating the internal consistency and reliability of the adaptive assessment algorithm, and beginning to examine predictive validity for academic outcomes. Critics of the Duolingo English Test have argued that the existing research is insufficient to support the confidence that some institutions are placing in it for high-stakes admissions decisions, while supporters contend that the research foundation is growing rapidly and already supports reasonable confidence in the test’s ability to identify genuine differences in English proficiency levels.
Test Security Approaches Differ
Test security is a concern of particular importance for high-stakes language proficiency assessments because the incentives for cheating are substantial, and the consequences of score inflation through cheating affect not only individual admissions decisions but the credibility and utility of the entire assessment for all legitimate test-takers. The TOEFL addresses test security primarily through its testing center infrastructure, which requires test-takers to present government-issued identification, submit to physical verification procedures, and complete the examination under direct human proctoring in a controlled physical environment with video monitoring. This approach provides a high level of security assurance that is difficult to replicate in an uncontrolled home environment.
The Duolingo English Test uses a combination of artificial intelligence-based proctoring, biometric verification, and post-examination review by human proctors to maintain security in the remote testing environment. The AI proctoring system monitors the test session through the computer’s webcam and microphone, flagging behaviors that may indicate attempts to receive assistance or consult unauthorized materials, and all flagged sessions are reviewed by human proctors before scores are released. Duolingo has been transparent about its security approach and has argued that the combination of adaptive testing, which makes it difficult to share specific questions in advance, and AI-based proctoring provides adequate security for the stakes involved. However, some security researchers and testing professionals have argued that remote proctoring technology has not yet achieved the reliability needed to prevent all forms of cheating with sufficient consistency, and that the absence of a controlled physical environment introduces security risks that cannot be fully mitigated through technological means alone.
Preparation Strategies Vary Significantly
Preparing effectively for the TOEFL iBT and the Duolingo English Test requires different approaches that reflect the structural and content differences between the two assessments, and test-takers who approach preparation for one test using strategies developed for the other may find themselves underprepared for the specific demands of their chosen examination. TOEFL preparation appropriately emphasizes building familiarity with academic English texts and audio content of the kind encountered in university lectures and readings, developing note-taking strategies for the listening section, practicing the integrated speaking and writing tasks that require synthesizing information from multiple sources under timed conditions, and building the stamina required to maintain focus and performance quality over a three-hour examination period.
Preparing for the Duolingo English Test requires familiarity with its distinctive question types, which include tasks like identifying real English words from lists that include invented non-words, typing sentences from dictation, describing images in written and spoken responses, and producing extended speech samples on provided topics. Because the test’s adaptive algorithm calibrates question difficulty to the test-taker’s performance in real time, there is less benefit to practicing specific difficulty levels of content and more benefit to developing broad and flexible English ability across the full range of skills the test assesses. Both tests reward genuine underlying English proficiency more reliably than test-specific gaming strategies, and preparation approaches that focus on building authentic language ability alongside format familiarity tend to produce better and more durable results than those focused primarily on test-taking tricks and format optimization.
Which Test Suits Whom
The choice between the TOEFL and the Duolingo English Test is ultimately a personal decision that should be made based on a careful assessment of individual circumstances, including the specific institutions being targeted, the test-taker’s financial situation, logistical constraints around testing center access, preferred testing format, and timeline for receiving and submitting results. Test-takers applying to institutions in countries where TOEFL acceptance is universal and where no financial hardship is imposed by the test fee, who are comfortable with the longer examination format and have access to a nearby testing center, may find the TOEFL to be the safer and more straightforward choice that eliminates any risk of encountering an institution that does not accept their scores.
Test-takers facing financial constraints, applying to many institutions simultaneously, located in regions without convenient testing center access, working within tight timelines that require rapid score delivery, or applying specifically to institutions that have clearly communicated acceptance of the Duolingo English Test may find that it serves their needs more effectively than the TOEFL. Some test-takers choose to take both examinations, using the Duolingo English Test for institutions that accept it while submitting TOEFL scores to those that require it, though this strategy adds cost and preparation burden that should be weighed against its benefits. Consulting directly with admissions offices at target institutions about their preferences and experiences with both tests can provide invaluable information that supplements general guidance and helps test-takers make the most informed possible decision for their specific situations.
Technology Reshapes Language Testing
The emergence of the Duolingo English Test as a serious challenger to the TOEFL reflects broader technological trends that are reshaping assessment across virtually every domain, as advances in artificial intelligence, adaptive testing algorithms, remote proctoring technology, and data analytics create possibilities for assessment design that were not technically feasible even a decade ago. The adaptive testing approach at the heart of the Duolingo English Test represents a genuine advance in assessment efficiency, and the research demonstrating that highly reliable ability estimates can be obtained in far less testing time than fixed-form tests require has implications for assessment design across language testing and beyond. As these technologies continue to mature and as the research base supporting their validity and security grows, the pressure on traditional testing center-based assessments to justify their continued reliance on costly and logistically demanding physical infrastructure will intensify.
Educational Testing Service has responded to the competitive pressure from the Duolingo English Test and other challengers by introducing its own at-home TOEFL option and by making various modifications to the examination format and logistics that improve convenience for test-takers. This response reflects the recognition that the testing landscape is changing and that even well-established assessments with strong institutional relationships must adapt to evolving test-taker expectations and technological possibilities to remain competitive. The long-term trajectory of the language proficiency testing market is likely toward greater use of remote delivery, adaptive assessment technology, and faster score reporting, with the remaining questions centering on how quickly these approaches can develop the security and validity research foundations needed to earn the confidence of the most demanding institutional users.
Global Implications for Test-Takers
The competition between the TOEFL and the Duolingo English Test has implications that extend well beyond the experiences of individual test-takers making a choice between two examination options, touching on fundamental questions about equity, access, and the role of standardized testing in mediating access to educational and professional opportunities in a globalized world. The historical dominance of expensive, testing center-based assessments has disadvantaged test-takers from lower-income backgrounds and from regions where testing infrastructure is sparse, creating barriers to demonstrating English proficiency that have nothing to do with actual language ability. The emergence of more affordable and accessible alternatives like the Duolingo English Test challenges these structural inequities and has the potential to expand the pool of qualified candidates who are able to demonstrate their proficiency to target institutions.
At the same time, the rapid expansion of assessment options introduces new complexities for institutions that must evaluate credentials from an increasingly diverse set of testing formats and providers, and the risk that acceptance of less rigorously validated assessments could reduce the accuracy of proficiency evaluation in ways that ultimately harm both institutions and the students who are admitted without the English ability needed to succeed in their programs. Navigating these competing considerations requires ongoing collaboration between test developers, institutional users, policymakers, and test-taker advocates to ensure that the evolution of language proficiency testing serves the genuine interests of all stakeholders rather than simply the commercial interests of testing providers or the administrative convenience of institutions.
Conclusion
The comparison between the Duolingo English Test and the TOEFL illuminates a pivotal moment in the history of language proficiency assessment, a moment when established traditions are being challenged by technological innovation and when the assumptions that have long governed what a credible high-stakes language test must look like are being productively questioned and reexamined. Neither test is straightforwardly superior to the other across all dimensions and for all test-takers, and the honest answer to the question of which test is better is that it depends on the specific circumstances, priorities, and goals of the individual test-taker making the decision. What is clear is that the competition between these two assessments has already produced benefits for test-takers in the form of reduced costs, improved accessibility, faster score delivery, and expanded institutional acceptance networks, and that this competitive dynamic is likely to continue driving improvements across the assessment landscape.
For the millions of international students and professionals who must navigate language proficiency requirements each year, the expansion of credible testing options represents a genuine improvement in a system that has historically imposed significant burdens on those who could least afford them. The young person in a rural region of a developing country who lacks access to a TOEFL testing center and cannot afford the examination fee, but who has genuine English proficiency earned through years of dedicated study, deserves a credible pathway to demonstrating that proficiency to the institutions that could transform their life trajectory. The degree to which the Duolingo English Test and future innovations in language assessment can reliably serve that person without compromising the quality of information that institutions need to make good admissions decisions is ultimately the most important question in this space, and it is one that deserves continued serious attention from everyone with a stake in the outcomes.
The future of language proficiency testing will almost certainly look more like the Duolingo English Test than the TOEFL in terms of delivery format, assessment technology, and accessibility, but the specific technical and institutional evolution through which that future is reached matters enormously for the quality and equity of outcomes produced along the way. Institutions bear a particular responsibility in this transition to maintain rigorous standards for the evidence they require before accepting new assessments for high-stakes purposes, while remaining genuinely open to the possibility that well-designed technology-based assessments can provide valid and reliable proficiency information in ways that expand access without compromising quality. Test developers bear a responsibility to invest in independent validity research, transparent security practices, and ongoing improvement of their assessments in response to evidence rather than commercial pressure alone. And policymakers bear a responsibility to ensure that the regulatory and accreditation frameworks governing language proficiency requirements evolve in ways that protect test-takers and institutions alike from the risks of both premature adoption and excessive conservatism as this consequential transformation unfolds across the global landscape of language assessment.
