Education’s Broken Compass
Why Schools Ignore Science
I was recently – thanks Zach Groshell – alerted to this paper by Douglas Carnine – you know, from Project Follow Through – about something that has kept me busy a large part of the past 25 years namely why it’s so hard to get the education world to use techniques that have been shown to work.
Douglas Carnine’s paper – Why Education Experts Resist Effective Practices (And What It Would Take to Make Education More Like Medicine) – examines the persistent reluctance of education experts to embrace scientifically validated teaching methods. Despite their significant influence over curriculum design, teacher training, and educational policy, many of these experts – whom I often disparagingly refer to as ‘eduquacks’ – prioritise ideology, tradition, gut-feelings, and personal beliefs over empirical evidence. This resistance has resulted in the widespread use of ineffective instructional methods, particularly in the teaching of reading and mathematics, which has disproportionately harmed students who are most in need of effective education.
Carnine argues that education, unlike fields such as medicine, lacks a strong foundation in rigorous scientific research. He points out that while other professions have evolved by relying on experimental studies and measurable outcomes, education remains dominated by subjective theories and trends. A key example of this is the long-standing debate over reading instruction. Decades of research, including findings from the National Reading Panel, have consistently demonstrated that systematic phonics instruction—focusing on phonemic awareness and explicit teaching of letter-sound relationships—is essential for developing technical reading skills, particularly for struggling and disadvantaged students. However, during the 1990s, many American classrooms were dominated by the whole-language approach, which emphasised context-based guessing and de-emphasised phonics. Whole language was promoted by a close-knit group of education school professors, textbook publishers, and teacher trainers, despite the overwhelming evidence that phonics was the superior method. The adoption of whole language was not based on its effectiveness but rather on its philosophical appeal, demonstrating how educational trends often spread through ideology rather than scientific validation.
A similar pattern has occurred in mathematics education. In 1989, the National Council of Teachers of Mathematics (NCTM) introduced new math teaching standards that were rapidly adopted across the United States. These standards, which encouraged a shift away from direct instruction in favour of student-led discovery learning, were implemented without first undergoing rigorous experimental testing. The lack of empirical support did not stop policymakers from endorsing the approach, and it became the dominant framework for math instruction in thousands of schools. Carnine criticises this unscientific approach, noting that in no other serious profession would a major intervention be implemented without first proving its efficacy. He draws a comparison to the Food and Drug Administration (FDA), which requires rigorous double-blind studies before approving any new medication. If a medical treatment were approved before being tested and later found to be ineffective or harmful, there would be widespread outrage. Yet in education, untested methods are frequently introduced, and even when they fail, they often persist for decades or, as Richard Mayer argued, are reincarnated with a new name are implemented as something new.
One of the most compelling pieces of evidence Carnine presents is Project Follow Through, the largest educational experiment ever conducted in the United States. Beginning in 1967, this federally funded study followed over 70,000 disadvantaged students across 180 schools, comparing more than 20 different educational models organised as nine approaches (see the following figure). The goal was to determine which approaches were most effective in improving student outcomes.
The results, analysed by independent research organisations, were clear: Direct Instruction (DI), a highly structured, teacher-led method developed by Siegfried Engelmann and Wesley Becker, produced the best academic outcomes by far. Direct Instruction students performed at or above national norms in reading, math, and language skills, while those taught using student-centred, constructivist models often performed worse than the control group, even on measures of self-esteem! Despite these conclusive findings, Direct Instruction was largely ignored or dismissed by the education establishment. Instead of adopting the most effective method, policymakers and education experts continued to promote models that had failed to produce positive results.
Carnine attributes this resistance to several factors. First, he argues that education remains an immature profession, similar to how medicine was before the advent of modern scientific practices. Historically, medicine was dominated by subjective beliefs, ineffective treatments, and resistance to scientific innovation. It was also dominated by ‘experts’ in what can be called eminence-based practice.
Detail from The Extraction of the Stone of Madness
(Hieronymus Bosch 1488–1516) – Public Domain
It was only through external pressure—from the public, the government, and the insurance industry—that medicine evolved into an evidence-based profession. Similarly, fields such as accounting, pharmacology, and seafaring have matured by implementing standardised, scientifically validated practices. In contrast, education remains guided by individual experts’ opinions rather than by controlled research studies. Many education professors, influenced by philosophical commitments rather than empirical data, prefer qualitative research methods and resist experimental studies that might challenge their beliefs. Some argue that education is too complex to be studied using the kinds of randomised controlled trials that are standard in other fields, dismissing such research as irrelevant.
This resistance has real consequences. Carnine illustrates how untested and ineffective teaching methods not only fail to improve learning but also deepen educational inequality. Disadvantaged students suffer the most, as they are often placed in classrooms where ineffective instructional methods prevent them from acquiring fundamental skills. Instead of using evidence-based approaches that could help close achievement gaps, educators continue to implement methods that sound appealing but do not work in practice.
To change this, Carnine calls for a transformation in education similar to what occurred in medicine. He argues that external pressure—from policymakers, parents, and the public—is essential to force education to embrace scientific methods. He suggests that government agencies should require rigorous experimental validation before new teaching methods are implemented, much like how the FDA evaluates new drugs. Education schools should be held accountable for preparing teachers with scientifically supported methods rather than trendy but ineffective theories.
Ultimately, Carnine’s argument is not just about improving education but about professionalising it. He envisions a future in which teaching is guided by objective data, where ineffective practices are abandoned in favour of proven ones, and where the well-being of students takes precedence over ideology. Until that shift occurs, he warns, education will continue to be plagued by cycles of unproven fads, with students—especially the most vulnerable—paying the price.
