“Scaling” educational innovations

This is an attempt to briefly (and possibly badly) express a disquiet I have with the idea of scaling educational innovations. It’s sparked by a post by Rick Hess titled “Why education innovation tends to crash and burn”.

The post suggests that there are two sets of obstacles that prevent educational innovations scaling

  1. Reliance on tough-to-replicate elements.
  2. Structural conditions that impede the growth of replicable models.

Structural conditions

This second set of obstacles resonates with me. It suggests that innovations are impeded by obstacles such as

  • True innovation challenges existing institutions and ways of doing things and hence likely fails within those existing institutions.
    This is Christensen’s disruptive innovation stuff.
  • The lack of price competition (e.g. many/most? parents don’t pay for education, at least not beyond taxes) leads to institutions not having to spend time building cost-effective models.
    There are some questions here, but it’s a factor.
  • The difficult of effectively comparing outcomes means you can’t objectively decide between alternatives.
  • A discomfort with for-profit companies – which are obviously better at innovation – being involved with education reduces risk etc.
    Not sure I agree entirely with some of the assumptions of this point, but there is an aspect of this which may reduce diversity. Not sure there is a lack of for-profits and also think some of the prior obstacles may contribute to this lack. Not just a discomfort.

Don’t think this list is complete, but it’s a start. It’s the first set of obstacles that trouble me.

The need for pilots – wrong set of “tough-to-replicate” elements

The underpinning assumption in the post is that the problem here is how to scale a successful pilot. i.e. some senior manager has identified a good innovation from somewhere, set up a pilot, it’s worked and now there is a need to scale this throughout the entire organisation. Based on this assumption the “tough-to-replicate elements” include: funding for the pilot, expertise of pilot staff, enthusiasm from leadership, accommodating policies.

The idea seems to be that the nature of a pilot makes it more likely to succeed, but that this nature is missing when you attempt to scale it more broadly.

I agree that these can be a problem, but I think the problems come from the assumptions that underpin this view of organisations. A view that results in the following solutions.

Hess’ solutions

The solutions Hess suggests essentially seek to minimise/negate some of the above obstacles and include

  • Pay more attention to innovations that scale easily.
    This is because you don’t have to pony up the resources to make it succeed.
  • Don’t innovation within existing institutions.
    More of the Christensen flavour. It’s to difficult to innovation in existing organisations with established cultures, do it elsewhere. This means leadership don’t have to be involved and even if they are, you don’t have to worry about reinvention.
  • Focus on cost and outcomes in allocating public dollars.
    Change public policy to avoid formula funding and limited measures of quality. Of course there’s no mention of how to do this, but it is a short article.

Change the type of system

All of the above seems heavily based on the assumption that an education institution is an ordered system. That the best way to scale an innovation is to pilot it, test if it succeeds and then scale it. It treats pilots as something separate, which is in part what the above argues against. But it goes further than this, if you assume an educational institution is a complex system.

Some of the implications of this view include

  • Limited knowledge of the existing system is a big problem.
    Pilots identified and supported by senior management are problematic because senior management – by the nature of their position – have little or now idea of the reality of organisational life. The “coal-face” is a mystery to them and so pilots often suffer from unexpected problems generated by clashes with the reality of organisational life, and this is why innovations are often worked around by coal-face workers. Pilot’s work around these clashes by having resources, expertise and leadership buy-in.
  • Organisational culture != the hierarchy.
    What senior management usually know about an organisation is based on the existing management hierarchy and the information flows it provides to management. The description of the value of a new innovative IT system given by the head of the IT division is going to markedly differ from the description offered by the people using it every day. Guess which one senior management are more likely to hear regularly?
  • The constraints of efficiency and purpose driven design.
    I don’t think it’s the established culture within the organisation (i.e. the recalcitrant workers) that provide the largest barrier to innovation. Instead it’s the “rational” organisational demand for efficiency which results in an increase in top-down, hierarchical policies and practices to ensure that resources aren’t being wasted. This is what constrains change and actively works against innovation from arising within.

    Within schools this can be seen in national curriculum, standardised computer equipment etc.

    This is part of the problem of organisations and their components being overly constrained by a particular purpose. “You are charged with teaching the Year 10 Core mathematics curriculum, any deviation from that curriculum is bad.” A great way to encourage innovation.

  • Simplified measurements.
    In order to measure effectiveness and/or learn more about the functioning of the institution simplified, standardised measures are adopted. For example, standardised testing, checklists, quality assurance processes, minimum defaults etc. All of which fail miserably at capturing the full diversity of the institution and instead end up driving the members of the institution toward the visible performance of some minimal default. i.e. it drives out diversity (or at least its visibility) and hence a source of innovation.
  • Decomposition.
    Schools as with most modern organisations are decomposed into smaller and smaller blocks. This decomposition tends to make the establishment of lines of communication between people in different blocks much harder. Driving out cross-silo communication limits capacity to innovate.
  • Innovation requires radical change and are predictable.
    An assumption behind “scaling an innovation” is that you are changing (possibly radically) the institution and that you can predict the outcome. If you view an educational institution as a complex system, you know you can’t predict the outcome because there will be unpredictable, non-linear effects.

    This means that you not only can’t predict what will happen, it also means that you don’t need radical change projects. Small changes within a complex system can have radical outcomes.

3 thoughts on ““Scaling” educational innovations

  1. How this is theorised is important. There is a story about shifting water pumps to the Sahara. Did not work. Turns out you have to move most of Western civilisation to the Sahara for the pumps to survive. Lots of stories like that. The thing about schools is that they need to be thought about as a technology, a way of doing things. That means you pay attention to all the bits and pieces that are in place that keep this ‘technology’ ticking over. So if you want to change that you have to attend to it all (i.e. like Western civilisation). This is an ANT-onformed account, i.e. it is not a diffusion of innov way of thinking about this stuff. I’ve scribbled about that long time back. At base it is how you think about/theorise stuff that is moving cf. stuff that is more or less stable. ANT works well for change that kind of thing. For the stable stuff older forms of social theory will work ok. Not sure this is helpful.

    1. I can see a lot of resonances between the ANT perspective and that of Complex Adaptive Systems, which was the “theoretical” source for much of the above.

      :) I’ve moved on a bit from the DoI stuff, but I still have a soft place in my heart for it. Perhaps because it might hold some potential positives when applied at a much finer grain….but that’s a story for another day.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s