Can Effective Schools Make Every Child Academically Proficient?

Dr. Russell T. Warne

Jun 20, 2025

Each State shall establish a timeline for adequate yearly progress. The timeline shall ensure that not later than 12 years after the end of the 2001–2002 school year, all students ... will meet or exceed the State’s proficient level of academic achievement … (No Child Left Behind Act, 2002, 20 U.S.C. § 6311)

For over a decade, the No Child Left Behind (NCLB) Act was a federal law in the United States that mandated that every student in public schools must be proficient in core academic subjects (math, science, and language arts) by the summer of 2014. If schools failed in this goal, then there would be consequences. Schools could be taken over by the state or closed down, personnel could lose their jobs (or be forced to reapply with the new regime at a school), and funding could be withheld.

As 2014 drew near, it became apparent to educational officials that it was impossible to make every child academically proficient by the deadline. In 2011, the Obama administration announced that they would grant waivers (of dubious legality) to exempt states from this requirement under certain conditions. The announcement said that this was because there were “specific NCLB mandates that were stifling reform” (“Obama Administration Sets High Bar for Flexibility,” paragraph 2). Behind the political spin was the reality that the goal enshrined in the law was not achievable. To prevent nearly every school in the country from breaking the law, the federal government gave these waivers to states that met certain requirements established by the US Department of Education (Kamenetz, 2014).

All of this was foreseeable. Educational psychologists and intelligence researchers had said for decades that it was impossible for every student to master a curriculum (e.g., Biemiller, 1993; Jensen, 1969; Kauffman & Konold, 2007). Yet, because the rhetoric of making every child proficient in core school subjects was appealing, the US Congress passed a law that mandated the impossible. A law banning gravity by 2014 would have been equally effective. In 2015, Congress replaced NCLB with a new law, the Every Student Succeeds Act. While the name is as optimistic as the previous NCLB, it did eliminate the legal mandate that every student reach an arbitrary level of educational competence.

The very names of these laws show unbounded optimism for the ability of the education system to produce successful outcomes for every student. Elsewhere in education, people claim that their favored curriculum or policy can “close the achievement gap” between students (e.g., Burris & Welner, 2005). The belief that everyone can be brought to a high standard of academic performance is so common that it “is a virtual article of faith in educational circles” (Gottfredson, 2009, p. 36; see also Frisby, 2013, pp. 212–215). The phrase “article of faith” is appropriate, because no country or state has ever created a school system that was successful in educating every student to a high level. Yet policy makers believe that this is possible anyway.

Why Johnny Can’t Read (Sometimes)

Not only has no education system ever made 100% of students competent, there are good theoretical reasons rooted in intelligence research to expect that such a utopian outcome is not possible. The first reason is fundamental: intelligence is positively correlated with educational achievement (C. M. Calvin, Fernandes, Smith, Visscher, & Deary, 2010; Cucina, Peyton, Su, & Byle, 2016; Damian et al., 2015; B. Roth et al., 2015). Depending on study characteristics, intelligence correlates with academic achievement at a level of r = .40 to .70. That correlation is so strong that – in most studies – intelligence is a better predictor of success in school than any other variable. This means that wherever there are individual differences in intelligence, there are individual differences in school competency, with smarter students usually performing better than low-g students. Higher-IQ students learn more rapidly, learn more efficiently, organize and generalize information more spontaneously, and make fewer errors than their average or below-average classmates. With these skills and a broader fund of knowledge available to them, high-IQ students perform better on standardized achievement tests than their peers, even if a teacher is highly effective at test preparation (Frisby, 2013). Individual differences in school success still develop because they arise from individual intelligence differences. No law or educational policy will change this. Frisby (2013) explained this succinctly: “Slow learners will always lag behind their brighter peers in academic work, and they will never catch up” (p. 211).

The second reason schools cannot make every child master the curriculum is that the causes of intelligence differences and the causes of academic achievement are both partially genetic. While it is true that keeping people in school longer raises IQ (see Chapter 14), differences in intelligence are one of the causes of differences in academic achievement. Therefore, the causes of intelligence differences will also be some of the causes of differences in school performance. Chapters 11 and 13 showed that – in typical environments in wealthy countries – genetic influences are about 20% to 80% of the cause of IQ differences. As a result, genes are an indirect cause of school performance differences because of how they act on intelligence. This chain reaction can be diagrammed as:

Genetic differences ➔ Intelligence differences ➔ School performance differences

This model of cause and effect is oversimplified, but still useful. There are other influences on intelligence and school performance (e.g., environmental influences, personality characteristics, school quality) that influence IQ and school performance, which makes the flow of causality from genes to IQ to academic achievement imperfect. Still, the connection between genes and school performance (via intelligence) is a major reason why no school system can make every child competent in every school subject.

The theory sounds plausible, but is there evidence to support it? The answer is a resounding yes, especially in the form of genome-wide association studies (GWASs). In a GWAS of intelligence, researchers collect DNA samples from thousands of people and identify variations in people’s DNA that are correlated with IQ. If a DNA variation is more common in high-IQ people than among people with low IQ (or vice versa), then it indicates that this DNA segment is associated with intelligence. These DNA variants may then be used to calculate a score (called a polygenic score) that can be used to predict a person’s IQ (Plomin & von Stumm, 2018). GWASs are not unique to intelligence; they can be conducted to identify DNA portions that are associated with any trait. Research has shown that many of the same genes that are associated with high intelligence are also associated with high educational performance (W. D. Hill et al., 2019; Lee et al., 2018; Okbay et al., 2016; Plomin & von Stumm, 2018). Therefore, the differences in school performance are partially genetic, which means that there will always be some differences among students in their academic performance. These differences will mean that some people will perform so poorly in school that they will not meet the standards of competence that teachers, school personnel, or lawmakers expect children to meet.

Although GWASs are a major breakthrough in understanding how genes relate to traits, there are some unanswered questions about these genetic influences. One problem is that the polygenic scores created from DNA segments that are associated with intelligence explain only about 4% of all influences on IQ variability. On the other hand, studies of IQ similarities and differences among family members show that the combined influence of all genes on intelligence variability is 20% to 80%. The gap between known DNA variants (identified through GWASs) and total heritability (from studies of family members) is called “missing heritability” (Hunt, 2011; Plomin & von Stumm, 2018). It represents the strength of genetic influences that have an unknown location in the genome. Experts predict missing heritability to drop in the future as behavioral geneticists conduct GWASs using better technology and larger, more diverse samples (Plomin & Deary, 2015; Plomin & von Stumm, 2018).

A more serious problem with GWAS results is that it is not clear how genes cause people to differ on their traits. This is especially true for behaviors or psychological traits, such as intelligence. A gene is just a segment of DNA. It cannot think, and it does not see what is happening outside the body. Genes don’t give people hints about how to answer intelligence tests or excel in school. All genes do is make strands of RNA that the body then uses to make amino acids for cells to create proteins (Plomin, 2018). If genes do affect intelligence levels, then it is not through a direct impact, but rather through the biological consequences of those genes. Biologists have not fully solved the mystery of how a segment of DNA results in a psychological trait, but it must be a biologically-based influence because genes are biological in nature, and they can only exert a direct influence on biology. Work on understanding the connections between genes, biology, and psychological traits shows that most genes are probably associated with multiple biological and psychological traits (Belsky & Harden, 2019).

Educational Consequences of g Denialism

Despite the importance of intelligence in determining educational outcomes, teachers and other educational leaders rarely acknowledge its importance. Education departments at universities often minimize the value of intelligence, with some choosing to emphasize Howard Gardner’s theory of multiple intelligences instead (Burton & Warne, 2020), even though the theory is not supported by empirical evidence (see Chapter 5). As a result, educators often deny or do not understand the implications of individual differences in intelligence. For example, few teachers understand that most educational tests measure g to some extent. As I discussed in Chapter 18, some educators believe that every child is gifted and that individual differences in intelligence are not important because every child is smart.

This denial of g has serious negative consequences in the education system. One result of g denialism is the blame game that often ensues when children’s educational performance fails to meet the expectations of adults because people refuse to admit that some children are always going to struggle in school. As Gottfredson (2005a, p. 546) explained, “Frustrated expectations devolve into blame. Test critics blame the tests, test companies blame the schools, [and] educators blame already angry parents ... But flagellating one group or another for lack of will or commitment has no constructive effect.” Playing the blame game discourages adults – who all have the same goal of helping children learn – from partnering with one another to improve the education system.

Another negative consequence of denying intelligence is that it causes teachers to assume that all of their students are approximately the same in their readiness to learn new material. This incorrect belief causes a teacher to assume that one lesson serves every student well. However, a typical group of students displays a wide span of cognitive abilities:

A not uncommon finding is that the children in an ordinary third-grade class span a range of competence in reading comprehension equivalent to the norms for the second through the eighth grades, or that those in the fifth span the range from the third through the tenth. (Herrnstein, 1973, p. 112; see also Frisby, 2013, pp. 229–231)

Unless a class is created to reduce these differences – such as by selecting children for a gifted or a special education program – then the class is likely to have children whose abilities span several grade levels (Biemiller, 1993). And – as Herrnstein implied – the variability in educational readiness increases as children age (Burt, 1917; H. D. Hoover et al., 2003). This usually is not apparent because typical grade-level tests are not designed to measure the abilities that far exceed (or are far below) the nominal grade level. Fully measuring these children’s abilities often requires additional testing, such as in an individual testing session or by giving a more difficult test to bright children – or an easier test to struggling students (Rambo-Hernandez & Warne, 2015; Warne, 2012, 2014).

Creating one lesson that meets the needs of all students in a typical classroom is nearly impossible. The best alternative is to create multiple lessons that are targeted at groups of students within a classroom (e.g., struggling readers, typical readers, advanced readers). This practice – called ability grouping – happens regularly, but because most teachers are unaware of the span of abilities in a typical classroom, it is only partially effective in most situations. Another problem with ability grouping within a classroom is the basic fact that it is more work and effort for a teacher to prepare multiple lessons than to prepare one (Hertberg-Davis, 2009). Educators who understand intelligence recognize that the best option is to perform the ability grouping at the classroom level (i.e., create a class full of advanced learners, a class or two of typical learners, and a class of struggling learners). In this way, the variability in educational readiness is reduced and teachers are better able to serve the educational needs of the students in their classroom.

Another negative consequence of intelligence denialism is that it leads to policies that are ineffective. An example of this occurred in California when a judge in a high-profile court case (Larry P. v. Riles, 1979) issued a statewide injunction against using intelligence tests for African American students because he believed that intelligence tests were a tool to confine struggling students to “dead end” special education classes. Even when the injunction was lifted several years later, the California State Department of Education still maintained the ban in public schools unilaterally because of a politically motivated denial of the importance of intelligence for determining which children should be placed in special education classes. But banning the intelligence tests did not eliminate intelligence differences or fix the problem of selecting children for special education. Instead, it forced school psychologists to use lower-quality tests and more subjective methods of making this important decision (Frisby & Henry, 2016).

Another consequence of g denialism occurs when educators or policy makers misdiagnose the source of educational differences. To someone who does not understand that g differences are – partially – a product of genetic differences, then the reason some children perform better in school must be completely environmental. As a result, some educators see the expanding variability in educational achievement and the inability of schools to equalize educational outcomes as signs that the educational system is failing (Gottfredson, 2000a). They advocate for more resources (e.g., by reducing class sizes, increasing funding) and mandate that all children be taught the same material in the same way in an attempt to eliminate differences in educational outcomes. But differences in educational performance are inevitable because the differences in intelligence (caused partially by genetic differences) among children still remain. The end result is a cycle of frustration and failure to achieve policy goals.

Some policies that deny intelligence actually harm students. One of these policies is the idea that every child should attend college. A college education – like education at the K-12 level – requires intelligence for success. However, because of intelligence differences, some students will struggle in college and not graduate. In 2018, 61.3% of Americans who were 25 or older had attended at least some college. Of these, 35.7% did not have any post-high school degree, and another 7.2% had only an associate’s degree (US Census Bureau, 2019, Table 1). Thus, almost half of college students do not earn a four-year degree. Many of these students require remedial classes to compensate for a lack of readiness for college-level classes or accumulate debt to pay for degrees they never earn. Some of these students drop out because of academic difficulties and feel like failures. If a majority of Americans attend college, then some of them will inevitably have an IQ below average, and these students will struggle mightily to earn a degree.

A Needed Dose of Realism

Political slogans like “Every Student Succeeds” and “No Child Left Behind” may feel good, but they are not grounded in reality (Frisby, 2013). I do not deny the good intentions of lawmakers and educational staff who claim that their favored policy will eliminate educational failure. However, reality does not care about good intentions. Intelligence differences are “real, stubborn, and important” (Gottfredson, 2000a, p. 76), especially in education. These differences are rooted – partially – in genetic differences. As a result, basing policy on high-minded platitudes and promising the impossible will only result in disappointment and disillusionment. Some policies may even harm students.

Scientific research cannot determine social or policy goals because those goals are inherently value-laden, while scientific facts are value-neutral. However, scientific research can inform which policies are infeasible and which have a chance of succeeding. Chapter 33 will give practical suggestions for policies based on intelligence research in education and other areas. I hope readers find that these suggestions are realistic and serve the needs of individuals with a wide variety of intelligence levels.

From Chapter 19 of "In the Know: Debunking 35 Myths About Human Intelligence" by Dr. Russell Warne (2020)

We hope you found this information useful. For further questions, please join our Discord server to ask a Riot IQ team member or email us at support@riotiq.com. If you are interested in IQ and Intelligence, we co-moderate a related subreddit forum and have started a Youtube channel. Please feel free to join us.