c Consulting Site of Michael E. Sikes, Ph.D.

Consulting

Michael E. Sikes, Ph.D.

Return to Home Page

Toward a Distributed Model of Evaluation

Periodically, some major organizing pattern of society changes fundamentally. For over a decade, we have been going through such a change in communications with the World Wide Web, which supplanted top-down structures (television networks and telephone companies) with more interactive processes and user-generated content. Now it appears we are headed into a similar change in how we generate and use energy, as massive power grids give way to interactive webs in which buildings generate their own green energy and sell it backward to utilities. In both of these changes is a transformation from hierarchical structures to more dynamic systems in which many people participate.

It seems to me that the field of evaluation is moving through such a change process, one which the World Wide Web and grid-interactive power systems can help us to understand metaphorically. The changing face of evaluation has much in common with these other two change systems in terms of distribution, and for this reason I propose to apply the term distributed evaluation—which already exists in some more specialized contexts—to this broad phenomena.

In what follows, I explore the possibility of such a systemic change across evaluation, describe what I think it means for the work of society, and then invite readers—through the medium of some future questions—to consider why we should care. My method is inductive rather than didactic, in that I seek to explain much phenomena through this overarching framework of a distributed system.

The Aspects of Distributed Evaluation

While no definitive list characterizes every distributed system, the following are some preliminary indicators in the evaluation field:

  • Knowledge is no longer top down or hierarchical. Instead, it is shared in all directions. We move from the wisdom of experts to the wisdom of crowds.
  • Knowledge and method are provisional, not privileged; thus not everything that can be known, is.
  • There is no single best methodology. Segments or pieces of method are reused. Each method is local and is verified through local validity, acceptability, cultural competence, and usability. Criteria for judging past evaluations will move from best practices to usability tests.

Possible Implications of Systemic Change

Several overall consequences may result if evaluation is going through a systemic change, not just localized changes. For example:

  • An existing dominant methodology does not produce useful results.
  • The dominant paradigm is not replicable from one context to another.
  • An existing, overall organizing framework continuously fails to explain new phenomena.

It seems as if many of these conditions are already in play.

More to come

Contact Information

   email Michael Sikes