"Early childhood education is important." There are few objections to this slogan. However, in the world of research, another "promise" has long been discussed. -- The gap that widens in preschool education fades by the time children enter kindergarten. This is the so-called "fade-out" effect. PMC

However, what ScienceDaily reported on January 1, 2026, was a result that directly challenged this common belief. The protagonist is the Montessori education, which has a history of over 100 years. Moreover, the setting is not a "luxurious private school" but a public Montessori preschool program in the United States. ScienceDaily

A "strong" design of a nationwide lottery-based RCT

The main reason this study is attracting attention is its strong design. The research team (University of Virginia, University of Pennsylvania, American Institutes for Research) tracked children who applied to 24 public Montessori schools where admission was determined by lottery. By comparing the group that won a spot in the lottery with the group that did not, they created a situation as close to randomization as possible within the context of real school choice. The subjects numbered 588. ScienceDaily

The paper was published in PNAS (Proceedings of the National Academy of Sciences), and the analysis was estimated using **intention-to-treat** and multiple models. To address the suspicion that "families choosing Montessori might already be highly conscious," they used the lottery system to approach fairness as much as possible. PMC

What improved? Reading, memory, executive function, social understanding

The results are simply strong. At the end of kindergarten, children who received a Montessori spot showed significantly higher scores in

• reading

• short-term memory

• executive function

• social understanding / theory of mind
  . The effect size is described as "large for field research (over 0.2 SD)." PMC

Interestingly, at the end of PK3 (age 3) or PK4 (age 4), noticeable differences were hard to see, but the differences appeared at the end of kindergarten. While many studies have shown "immediate improvement that later disappears," this time it seems to have "taken effect later." PMC

The researchers discuss the possibility that the unique learning in Montessori (spontaneous activities, peer teaching through mixed-age groups, phonics-based reading learning, etc.) crystallizes over time, contributing to this "delayed effect" pattern. PMC

Even more surprising: Good yet inexpensive (approximately $13,127 per person)

Education topics often end with "but the cost..." However, the researchers themselves were surprised by this aspect.

For the three years from ages 3 to 6, an estimate of $13,127 per person was considered "cheap" (presented as "about $13,000 less"). The main reason is said to be the relatively high child:teacher ratio in PK3 and PK4. While private Montessori is known for being expensive, the public model showed potential for structurally reducing costs. PMC

An article from the University of Virginia also explains that the cost analysis included teacher training and specialized teaching materials, showing a difference of about $13,000. UVA Today

"Public Montessori" is no longer rare

It is often misunderstood, but Montessori is not a "rare experiment." The ScienceDaily article states that over 600 public schools in the United States offer Montessori education. In other words, the research results are an evaluation of a system that already exists in society, not some "utopia." ScienceDaily

Additionally, while the effects were seen overall, the university's explanation also touches on the possibility that they might be stronger for children from low-income families or boys. UVA Today

However, it's not a panacea: The study itself lists its "limitations"

The sincerity of this study is also evident in its detailed listing of limitations. Notably --

• The target schools are those with "excess applications for PK3," and they may not represent all Montessori schools nationwide. PMC

• Applicants are families already interested in Montessori, and they do not represent the general population. PMC

• Consent to participate in the study was low, at about 20% of applicants, and there was a potential bias with higher consent rates in the treatment group. PMC

• There were missing evaluations at the kindergarten stage, with 32% missing in the treatment group and 42% in the control group, requiring assumptions to be made with multiple imputations. PMC

• The follow-up is "only until the end of kindergarten." What happens in subsequent grades is unknown. PMC

It's not a simple "effective/ineffective" dichotomy; it's necessary to read "under what conditions and to what extent it can be generalized."

How was it received on social media? (Summary of key points)

When research is published, "discussion points" inevitably emerge not only in academia but also on social media. This case was no exception.

1) LinkedIn: Researchers and educators hope it "reaches policy"

Lead author Angeline Lillard introduced the research on LinkedIn, succinctly posting key points such as "random assignment in 24 public programs" and "surprisingly lower costs." She also expressed enthusiasm for follow-up research. LinkedIn

In the comments,

• there was anticipation for "the next step," such as "Will it be tracked beyond elementary school?" LinkedIn

• There was also support for it to "reach policymakers." LinkedIn
  On the other hand, questions from a consumer perspective also emerged. For example, "Does 'cheap' refer to administrative operating costs? It doesn't match the parental burden, right?" This highlighted the issue of public cost efficiency and private tuition rates being easily confused, even on social media. LinkedIn
  
  

2) Reddit: Praise and skepticism coexist in the "same thread"

A more candid temperature was found in Reddit's parenting community (ScienceBasedParenting). While there was an evaluation that "educational research is often weak, but this is relatively good with RCT and pre-registration," there were also lengthy comments carefully pointing out the study's limitations, such as missing data, dropout, the wavering of randomness, and the possibility of overestimating effect sizes. Reddit

"It’s always nice to see an RCT… it’s preregistered, and it uses intention-to-treat analysis…" Reddit
"over a third … did not stay in the study through kindergarten… forcing the researchers to guess…" Reddit

There was also a different kind of "field perspective." As an experience of moving from Montessori to a typical public school, there were voices saying, "It was tough to move from a high degree of freedom to a structured classroom." This response highlights the importance of transition design, not just the discussion of effects. Reddit##HTML_TAG