Not long ago, I expressed concern about the gap between academia and the nation’s educational policymakers. This disconnect has significant implications. Recently, I witnessed a compelling episode. Paul T. von Hippel elaborated in a fascinating article for Education Next how inflated assertions about the effectiveness of tutoring can be traced back to a questionable “two-sigma” effect suggested by psychologist Benjamin Bloom four decades ago.

Since 2013, I have been advocating the potential of tutoring, so I am not dismissive of it. However, I’ve encountered many educators and officials who have embarked on major initiatives based on overly optimistic promises. It is clear that many educational decision-makers, advocates, leaders, and tech enthusiasts have latched onto the “two-sigma” concept, leading to questionable choices and inadequate focus on implementation.

Seeing von Hippel’s article as a chance to manage expectations better, I recently wrote a Forbes column stating that the post-pandemic surge in tutoring may be influenced by “suspect science.” I pointed out that rigorous research indicates that the actual benefits are likely much lower (less than one-fifth) than what many educators and officials had believed.

Several academics criticized the Forbes column. Their argument? The gist of it was that the Bloom “two-sigma” concept is outdated, newer research shows modest but real benefits from tutoring, and thus, it was unfair or ill-intentioned to focus on von Hippel’s criticism of Bloom. It seemed as though I was being accused of tarnishing tutoring, despite my efforts to explore its potential benefits and effective delivery methods in various articles. The assumption was that everyone was already aware of the latest credible research, including a well-regarded 2020 meta-analysis from the National Bureau of Economics Research that confirms the benefits of tutoring (as I mentioned in the column).

This perspective reflects a detached view. While academics may consider a four-year-old NBER paper common knowledge, it isn’t. The experimental studies incorporated in the meta-analysis are also not widely known. It’s safe to say that few educational leaders, advocates, or policymakers have familiarized themselves with the material that my critics presumed to be common knowledge.

Moreover, the “two-sigma” claim is still widely accepted, even if its origins are not well-known. Many assume it represents cutting-edge research. Sal Khan even used Bloom’s finding as the title of his 2023 TedX talk introducing his AI tutor-bot, Khanmigo.

In a podcast episode, tech investor Marc Andreessen confidently stated that one educational intervention, one-on-one tutoring, reliably produces two-sigma better outcomes. Influential organizations like the Bill and Melinda Gates Foundation and the Chan Zuckerberg Institute have aimed for significant improvements in mathematical proficiency based on Bloom’s work.

Where do these influencers and decision-makers get their faith in the two-sigma effect? Mostly from conferences, vendor discussions, and webinars with supporters—a pervasive belief.

This situation offers some valuable insights.

Firstly, academics should recognize that expertise in a subject like tutoring doesn’t equate to comprehensive knowledge of all related aspects. Assuming everyone is aware that the impact of tutoring is closer to 0.35 standard deviations than 2.0 is presumptuous. Engaging in discussions to bridge these gaps would require academics to acknowledge that their expertise in research design doesn’t automatically translate to understanding the perspectives of educators or policymakers.

Secondly, it would be beneficial if more individuals in education were familiar with valid and reliable research findings. However, it is important to understand that superintendents, tech developers, and legislators have demanding roles that may limit their time for engaging with scholarly content. They may lack extensive training in research or statistics, making complex terminology challenging to grasp. Therefore, researchers should strive to communicate their findings in accessible language repeatedly. It would be valuable for academics to redirect their energy toward educating the public on the realities of tutoring instead of critiquing the attention drawn to older research.

Lastly, academics aspiring to make a difference in education should focus on practical understanding rather than relying solely on published reports. While it’s tempting to assume that once information is disseminated, it becomes common knowledge, this is not always the case. Some scholars were offended by my emphasis on Bloom’s “suspect science” despite newer, more relevant research available. While they may be well-versed in the latest studies, the same cannot be said for the key decision-makers impacting children and schools.

This entire episode offered valuable insights into the dynamics at play.

Frederick Hess is an executive editor of Education Next and the author of the blog “Old School with Rick Hess.”