Research does not become dated

The Atlantic recently included an article criticizing the educational version of personalization promoted and supported by Mark and Chan Zuckerberg because it relies on dated research.

And for good reason: The results from the 1984 study underlying it have essentially never been seen in modern research on public schools.

The “dated” research was described by B. Bloom as the two-sigma challenge and it sets as a challenge the search for instructional methods that could come close to matching the achievement of students working with a tutor. Individualized instruction with a dedicated result is proposed as the benchmark against which other instructional tactics could be prepared. Then as now, this optimal experience would be financially impractical for nearly all students. The personnel costs involved in education is another variable that has not changed with time. I cite Bloom and the educational tactics I see as related to this position (mastery learning) myself so I take some offense at the characterization of the notion of “aging research”. Scholars criticize research mainly for issues of methodology and not the date of publication. The article does remark on the content used in the research and the number of schools/participants involved in the research. Let me say that the value of tutors has been substantiated repeatedly (see Hattie data on most effective educational tactics). The connection between tutoring and other approaches to personalization does add some additional complexity, but there have been multiple large-scale reviews of mastery learning as well.

Personalization is hot in education. Unfortunately, it is an ambiguous term and it is easy to interpret it based on personal experiences. The kind of personalization emphasized by mastery advocates involves recognizing that individuals will learn at different rates based on aptitude and background knowledge. Pushing a group ahead at a fixed pace often fails to take these differences into account boring the more advanced students and frustrating the weaker students. With the weaker students, moving ahead without mastery further contributes to the limitations of existing knowledge increasing these students problems in the future. It makes some sense to me that because certain content is more hierarchical than other content areas, mastery before progress matters more in some situations than others.

Zuckerberg’s version of personalization takes advantage of technology to address individual differences. Some have a gut level reaction to this idea without investigating any further. When I advocate for similar ideas, I suggest that educators do not have sufficient time to work with individual students and certain uses of technology free them from certain traditional duties allowing them greater flexibility in how they spend “teacher time”. This basic idea shares some overlap with the popular and somehow more accepted idea of flipped classrooms. Why use class time for presentations when class time could be more productively spent working with students. In the case of mastery learning, why not allow technology to handle certain instructional tasks (presentation, practice) and apply these tasks at the level required by individual students and allow the teacher to rotate from individual spending his or her time with those students most needing assistance?

How such approaches are used can easily be mischaracterized. There is nothing to say that these tactics must be used in all content areas all of the time. This is basically the idea of a flexed approach.

I write more about these ideas in a discussion of mastery learning and the Kahn Academy.

Larry Ferlazzo comments on the same Atlantic article. I doubt that Bloom or any of the rest of us who have since conducted research on mastery concepts would suggest that this is a silver bullet. Mastery learning addresses a specific problem in educational practice – different students are at different points and would best be served if met at these different points. If providing this form of individualization is not possible for a teacher, it makes sense to search for other means.



This entry was posted in Uncategorized and tagged . Bookmark the permalink.