‘Personalized’ Instruction Using Tech Yields Uneven Results, Study Says

Associate Editor


Using technology to educate students isn’t always effective, or cost-effective, and doesn’t necessarily translate into what is often called “personalized learning,” according to a report released Monday.

The review of various studies by Noel Enyedy, an associate professor of education and information studies at the University of California-Los Angeles, concludes that education technology deployed in the name of personalized instruction yields modest improvements in educational outcomes, at best, in some cases, and none at all in others. 

Enyedy summarized his findings in a brief with a long title—”Personalized Instruction: New Interest, Old Rhetoric, Limited Results, and the Need for a New Direction for Computer-Mediated Learning”which was released by the National Education Policy Center, housed at the University of Colorado Boulder’s school of education. 

In a phone interview, Enyedy said he does not object to the use of technology. In fact, he calls himself “a critical friend” of education technology, evaluating his own use of it in the classroom and for research, and critiquing where educational technology is effective in general.

Educators who use digital tools bear the responsibility of finding out what works, he said, and policymakers would be well advised to make sure schools test technology on a limited basis before buying on a large scale. It is, he wrote, “unrealistic and irresponsible not to figure out how to use technology well.” 

He also says education technology an expensive option for districts, especially considering that it produces uneven results. 

Personalized Instruction vs. Personalized Learning

Enyedy draws a distinction between what he describes as personalized instruction and personalized learning.

Personalized instruction, he explains, “focuses on tailoring the pace, order, location, and content of a lesson” for each student. Most technology systems on the market today fall into this category, which is why he focused his study on this aspect of computer-based education.

By contrast, the broader term “personalized learning” centers on the process of learning, rather than honing in on the delivery of content, Enyedy says. This approach refers to the way “teachers or learning environments can vary the resources, activities, and teaching techniques to effectively engage as many students as possible,” his report states.

Based on his findings, Enyedy advises district officials to take several steps when weighing the merits of personalized instruction: 

  • Invest incrementally in technology. Policymakers should take a skeptical view of claims about what can be accomplished with computerized learning, unless it is supported by research-based evidence.
  • Conduct more research on the effects of personalized learning in K-12: Much of the evidence supporting technology’s potentially positive impact is based on research with undergraduates, who may be at different stages of development and motivation than pre-college students.
  • Set clearer definitions about the features of technology, and expectations for personalized instruction in the classroom. Shared definitions and ideas will help researchers more concretely define best practices in personalized instruction.
  • Test and validate software and hardware tools. Ed-tech developers should be encouraged to work with researchers and teachers in this effort, because market forces alone cannot be trusted to sort out which systems are effective.
  • Make professional development part of the implementation. A substantial investment in helping teachers learn how to use technology is a key component of working toward desired educational outcomes, Enyedy argues.

Showing Promise

Of all the approaches to using personalized instruction, those that rely on “blended learning”—which combines tech-based and person-to-person instruction—show the greatest potential academic benefits, he found. But blended learning strategies also tend to carry the highest costs, while producing only “moderate to mixed results.”

A RAND study, for instance, found that Cognitive Tutor Algebra I—which showed positive effects for students in the second year of implementation—was substantially more expensive than business-as-usual with standard textbooks, but that this cost premium was lower if the district already had strong technological infrastructure. 

Blended learning environments require investing in the technological infrastructure, licensing fees, and professional development for teachers and administrators on how to use the features, and how to integrate them into daily classroom practice.

The full brief is available here.

<a href=”https://www.scribd.com/doc/247016414/pb-personalized-instruction-pdf” style=”margin: 12px auto 6px auto; font-family: Helvetica,Arial,Sans-serif; font-style: normal; font-variant: normal; font-weight: normal; font-size: 14px; line-height: normal; font-size-adjust: none; font-stretch: normal; -x-system-font: none; display: block; text-decoration: underline;” title=”View pb-personalized-instruction.pdf on Scribd”>pb-personalized-instruction.pdf</a> by <a href=”https://www.scribd.com/nepc_publications” style=”margin: 12px auto 6px auto; font-family: Helvetica,Arial,Sans-serif; font-style: normal; font-variant: normal; font-weight: normal; font-size: 14px; line-height: normal; font-size-adjust: none; font-stretch: normal; -x-system-font: none; display: block; text-decoration: underline;” title=”View National Education Policy Center’s profile on Scribd”>National Education Policy Center</a><iframe class=”scribd_iframe_embed” frameborder=”0″ height=”500″ id=”247016414″ scrolling=”no” src=”http://www.scribd.com/embeds/247016414/content” width=”100%”></iframe><script type=”text/javascript”> (function() { var scribd = document.createElement(“script”); scribd.type = “text/javascript”; scribd.async = true; scribd.src = “http://www.scribd.com/javascripts/embed_code/inject.js”; var s = document.getElementsByTagName(“script”)[0]; s.parentNode.insertBefore(scribd, s); })() </script>

Related stories:

UPDATE: This blog post was updated with additional information about the RAND study.

2 thoughts on “‘Personalized’ Instruction Using Tech Yields Uneven Results, Study Says

  1. The fact that personalized instruction technology yields uneven results should be expected, but this is not a reason to discredit personalized instruction. We all have different learning styles. Computer instruction isn’t going to be effective for kinesthetic learners or extremely social children. Children who are visual learners, faster or slower than the speed of their classes, or who focus well are likely to benefit greatly from online personalized instruction. The key to making this technology beneficial to students is sorting children whose potential is optimized in face-to-face settings from those whose greatest success comes from working independently. In practice this likely wouldn’t be difficult. Even in our current one-size-fits-all system, most K-12 schools offer several periods of the same one class each semester. Schools could offer one class of 8th grade algebra online and one in a face to face version. Instead of randomly plopping students in classes, parents, students, and current teachers could decide together which method is most beneficial for this particular child. After trying both, students will eventually learn which method works best for them, or which methods they prefer for various subjects. We don’t need to worry about the overall uneven results, we need to teach children according to what will allow them to achieve their own best results.

    This article also speaks to the lack of data on the effectiveness of personalized instruction for K-12 students. I agree, this is a problem. There are many crucial differences between K-12 and college students. Probably most important is that college students have decided they value their education more than they value thousands of dollars. Given what an education is worth to them, it would be irrational to pay this price and then not use the resources they have paid for. Since K-12 schooling is free, we can’t say how K-12 children’s families value their education. Perhaps some see their education as holding very low value. They might see drawing, or socializing or staring off as a better use of their time. Since not all students continue on to higher education, it’s safe to say that college students come from a subset of K-12 students who value their education especially highly. Using college students’ success with online learning programs to make assumptions about K-12 students is an apples to oranges comparison. The only way we’ll be able to analyze relevant data, however, is if we start creating some. Collecting this data doesn’t even require dramatically changing schools or the school system. It simply means offering a great opportunity to children, some of whom will benefit and some who will decide face-to-face is more effective for them. Personalized online instruction is a great first step towards a school system that works for nearly all students rather than just the lucky few who happen to be a good fit with our conventional school system.

Leave a Reply