Blogging Disrupting Class #2: Chapter 1 -- Why Schools Struggle to Customize

The previous post summarizes how the introduction to Disrupting Class discusses common hopes for schools, why schools struggle to improve, and how this book proposes to describe its solution to this problem. The main point of Chapter 1, "Why Schools Struggle to Teach Differently When Each Student Learns Differently," is that students learn in different ways, so they need customized learning; but schools are designed to deliver standardized teaching, and their structure makes it very hard and expensive for schools to deliver customized or modularized learning.

Preceding Chapter 1, there is an "opening vignette" to a fictional story designed to illustrate the key issues and solutions discussed in the book. This story continues at the start of Chapter 1 (pp.21-23). It does a good job of anecdotally capturing some long-standing issues in the current system, for instance how teachers perform "triage" out of necessity because they simply don't have the time to reach all of their students. The story which illustrates how different students learn differently is a little strained but does its job of setting up the main insight of this chapter:
students learn differently, so they need customized learning; but schools are designed to deliver standardized learning, and their structure makes it very hard for schools to deliver customized or modularized learning.

Taken separately, these ideas are hardly new or earthshaking. Individualized instruction has been around now for what, 30, 40, 50 years? Or longer if we count Mark Hopkins at one end of a log, how Thomas Jefferson learned the law, Alexander the Great's tutor, etc. And the standardization of schooling has a long and well-documented history. Still, the assertion that we learn differently from each other is a good place to start, and Disrupting Class does a good job of walking us through this argument -- we know this anecdotally and through our personal experience (pp. 23-24); cognitive psychologists and neuroscientists are increasingly supporting this (p. 24); Howard Gardner's multiple intelligences framework shows us how this is so (pp.25-29) and how within intelligence types there are also different learning styles (p.28). This last example is particularly important, for as we'll see later, the book relies heavily on the multiple intelligences framework for its proposed solutions.

Up until now, this book hasn't been much different from numerous other books on education. But starting with the "Interdependence and Modularity" section (p. 29), the book sharply changes course and becomes quite unlike most other education books I've ever read, as Christensen et al. begin to apply their knowledge of disruptive innovation to the field of education. The section starts with a description of the concepts of interdependence and modularity as defined in the world of product design. The key points essentially are these:

• A product or service design is interdependent if the design and manufacture of each of its components is dependent on how other components are designed and made.
• When these interdependencies are unpredictable (i.e., you can't know ahead of time how to build one part until you also build the other part(s) on which it depends), proprietary architectures result because an organization has to develop its own interdependent design to figure out how the parts will work together.
• Modular architectures, by contrast, consist of modular components which can be built independently of one another, so they can be developed by independent organizations or work groups according to pre-established specifications.
• Customizing products or services with interdependent architectures is complicated and expensive because changing one part requires making changes in other parts as well.
• Customizing products or services with Modular architectures is much easier because changing one part can be done independently.
• Products and services tend to become more modular as they become more mature (as examples, compare Apple's early computer with current PCs; compare Windows OS with Linux [p.32])

What does all this have to do with education? The basic point is that US public schools (still primarily talking K-12 here) have a highly interdependent architecture in several important ways:

• temporal, i.e, you can't study topic y in one grade if you didn't study topic x in an earlier grade (to paraphrase their example [p.33]; other examples are easy to spot, for example you can't study topic y at all because you're in grade x, and topic y is scheduled for study some time before or after grade x).
• lateral = the example Christensen et al. use here is rather weak (you can't teach foreign languages in other more efficient ways because you'd have to change the way English grammar is taught, which is thus dependent on changing the English curriculum). I don't think the first assumption is correct; the reason that foreign languages are generally taught so poorly is dependent on other factors which have the usual inertial resistance to change IMO. But this doesn't negate their essential point that there are lateral interdependencies.
• physical = for example, schools' physical layouts are not well designed to support project-based learning (no surprise there since classroom design is still heavily influenced by its factory model legacy).
• hierarchical = Christensen et al. cite many examples here, from union-negotiated work contracts to governmental mandates at various levels to centralized curriculum and textbook decisions to standardized tests. This seems to be where the bulk of the issues lie.

Their conclusion: because of all these interdependencies, there are "powerful economic models" (one might also add political and cultural models as well) which support standardizing instruction and assessment even though we know that these should be customized because "students learn in different ways." But they focus on the economic side of it because "customization within interdependent systems is expensive" (p.34). They'll be dealing with this issue more in Chapter 5, but for now they briefly cover the example of IEPs (individualized education plans) to illustrate how current attempts at customization are expensive.

The next question Disrupting Class poses is, "Can we customize economically within the present factory model schools?" (p.35). Their answer, of course, is no, but they arrive there in a rather curious fashion, thanks to their emphasis (fixation?) on multiple intelligences. The answer is no because teaching to multiple intelligences in the current model is problematic, creating a "reverse magnetic attraction" where teachers favor one intelligence over another because of their personal preferences and the nature of the subject (pp. 36-37). The process creates "intellectual cliques" which favor students who use that preferred intelligence. For example, language arts teachers/curriculum developers favor linguistic intelligence, etc., and they are "trapped by their own strengths" by favoring methods which exclude students not endowed with strong linguistic intelligence. Instead, Disrupting Class argues, learning should be customized by catering to multiple intelligences (a recommendation which foreshadows lots of recommendations in forthcoming chapters).

At the end of the chapter, interspersed with now-commonplace calls for "student-centric" learning where teachers become "guides on the side", Disrupting Class introduces one of its other key recommendations: the use of computer-based learning as a "disruptive force" used to "modularize the system" and customize learning in the process (p.38). As we will see, their recommendations for how to use computer-based learning are truly far-reaching, radical, and transformative.