Throwing out the Baby with the Bathwater

Back in 1997, I worked on an evaluation of the Technology Literacy Challenge Fund (TLCF). The goal of TLCF was to ensure that every classroom had a computer and to integrate technology into the curriculum. At the same time, E-Rate was established by the Telecommunications Act of 1996 to provide affordable internet access to schools and libraries. In trying to reduce the digital divide, we learned that the students who had the best insights on using technology were not the typical straight-A students. Teachers who were willing to learn from the students were the most successful, as the biggest obstacle was ensuring teachers had sufficient training to use the technology effectively.

It did not take long for US education to move from a computer in every classroom to a device for every student. Funding was provided to ensure all students had access to Wi-Fi, and by 2015, schools were working with Chromebooks and iPads regularly, in addition to the traditional desktop computers. The pandemic accelerated the trend further as students needed computers and Wi-Fi to attend remote classes.

Now, in 2026, we are seeing a reversal in the trend. States are calling for a reduction in screen time, and some are even legislating removing computers from the classroom. So, what happened? Was this an experiment that went too far?

The problem

Testing trends show that students have been scoring lower on national tests and have been since about 2012. The 2026 NAEP release  showed students are reading less, and we have been hearing for years that attention span is lower and depression is higher among pre-teen and teenage children. The Anxious Generation by Jonathan Haidt[1], argued that the rise of smartphones, social media, and screen time are linked to depression, anxiety, and self-harm among adolescents. The book was a best seller and called for concrete changes such as banning cell phones in schools.

Source: NAEP Long-Term Trend Assessment Results: Reading and Mathematics

This book followed the launch of ChatGPT 3.5, bringing artificial intelligence to the masses. It had been publicly available since its launch in November 2022, but this version of OpenAI software made it mainstream, followed closely by Google’s Gemini and Anthropic’s Claude (and a multitude of others, but you know who they are). This groundswell of usage also brought the fear of technology taking over and signaling to our children that learning was no longer important, as you could ask anything of a Chatbot.

Additionally, research shows that when technology is misused, it can distract students from learning. Even notetaking turns from an active to a passive activity when moved from writing to keyboarding (Allen, et al., 2020).

The Current Solution

In response to concerns about screen time and decreasing achievement scores, many districts and states are passing cell phone bans and even full technology bans. For example, Utah recently passed legislation setting grade-level limits on screen time and technology use, with almost no screen time in grades K–3 and limiting use in grades 4–6. The argument is that students need to return to traditional learning and focusing on skills like collaboration and communication that are considerably narrowed through the overuse of screen time.

Kansas recently proposed one of the nation’s strongest screen-time law, eliminating digital instructional devices entirely from K–5 classrooms. Oklahoma has also made moves to limit instructional screen time in K–5 to roughly one hour a day. Iowa and Los Angeles both adapted a similar cap on screen time.  All of these laws have exceptions for students with disabilities needing technology for accessibility; however, students with disabilities achieve more when their learning is integrated with their non-disabled peers. Singling them out for technology is yet another distinction that separates them from others.

The Research

Research demonstrates that when computers are used effectively, they can boost achievement, although the effect size is moderate. A metanalysis conducted in 2016 examined research on the effects of one-to-one devices used in teaching and learning between 2001 and 2015 and found positive effects in English, writing, mathematics and science.[2]

Other research shows the importance of technology supporting personalized instruction, and particularly technology that adapts to the learner. A meta-analysis demonstrated that a personalized approach that adapts to students’ knowledge and skills had a positive impact with an effect size of 0.35, considered moderate.

It is important to recognize that “personalized instruction” can mean many things, including:

1. Adaptive learning systems, that change the complexity of a task based on performance,

2. Intelligent tutoring systems (ITS) that provide targeted feedback,

3. Personalized “move on when ready” pathways that allow students to move at their own pace,

4. Recommendation systems that provide lists of recommended resources.

The research evidence is much stronger for the first two types than for the last two. For example, A meta-analysis on research comparing the outcomes from students learning from ITS to those learning from non-ITS learning environments found that the use of ITS was associated with greater achievement compared to large-group instruction (effect size of .42), non-ITS computer-based instruction (effect size of .57), and textbooks or workbooks (effect size of .35). There was no significant difference between learning from ITS and learning from individualized human tutoring or small-group instruction. These effects were found across grades in almost all subjects, although the majority of the research  is conducted in middle and high schools.

In addition to adaptive tutoring, technology has significant benefits for students with disabilities. Technology allows students to incorporate necessary accommodations, such as text-to-speech, large-print, and alternative response formats, seamlessly. Without technology, students needing these accommodations could be isolated from the classroom to receive the support (such as reading out loud) without disrupting others.

Conclusion

Although I agree that too much screen time is detrimental to our children, I worry that policymakers will throw the baby out with the bathwater. Research shows that technology used appropriately can have benefits to students, particularly students with atypical learning patterns. Students with disabilities will suffer most from a technology blackout, as will students who benefit from personalized lessons. Of course, teachers can provide personalization, but with large class sizes, technology provides the support for teachers to provide more individualized instruction.

Technology can support students who are easily distracted by providing noise-cancelling headphones or those who need print or graphics enlarged for reading without requiring separate printouts. Most states now have computer-based assessments in grade 3 through high school, with substantial research showing students need time to familiarize themselves with the computer devices prior to the assessment to do their best work.

Problems arise when technology is used as an award. I observed a classroom for students with disabilities, and when students finished an assignment early, they were allowed to watch movie trailers on their laptops. There is no educational value in movie trailers, and the lesson was to work quickly, not to do the best job possible. Unmonitored technology time also causes problems when students are on social media or surfing the web rather than focused on a class assignment. Additionally, young children do benefit from reading print material and writing by hand. Technology should not replace those activities.

However, technology can be used effectively to help students analyze data, create graphics, or research a topic. These uses have been in place for more than three decades. Additionally, new approaches to instruction provide students with the opportunity to take short quizzes, receive instant feedback and target learning to their exact misconception or gap in knowledge. In short, technology works well when integrated into a full lesson plan.

[1] Haidt, Jonathan. (2024). The anxious generation: How the great rewiring of childhood is causing an epidemic of mental illness. New York, NY: Penguin Press.

[2] Zheng, B, Warschauer, M, & Chang, C. (2016). Learning in one-to-one laptop environments: A meta-analysis and research synthesis. Review of Educational Research, 18(4). pp. 1052–1084.