Frame the Problem

The first step in understanding a problem is to frame it, a process that here involves (a) providing a description of the problem, and (b) identifying relevant context moderators. An effective frame should provide sufficient detail to encompass the scope of the problem, yet still be concise. It should describe the problem, its scope, and its impact. Consistent with the hermeneutic orientation described in the previous section, direct, tangible experiences are called upon here. Even someone with no direct experience with the problem begins with transferable knowledge from similar problems and preconceptions that shape initial understanding, as well as assumptions and biases that mediate interpretations of prior knowledge \citep{Maxwell}. A well articulated problem frame provides transparency into the researcher’s decision path when determining how to analyze and ultimately address the problem \citep{Munn-Giddings}. This can be critical when evaluating the viability of interventions for use in another context.
The act of describing involves the researcher’s initial schema in meaning construction \citep{Langer}, and begins the process of determining what is important about the problem within its situated context. A problem in one place may contain elements that are of less importance in another; thus, an effective framing lays down a marker for what the researcher believes to be important. Context is central to this. An intervention may prove more or less effective based on context \citep*{VanBavel}; therefore, explicating the relevant moderators allows the researcher to be intentional about incorporating them into the problem analysis.
Problems addressed during action research are wide ranging; and constraining them may involve subjective decisions about what is in and what is out of scope. A well-articulated description of the problem provides a rationale for these decisions, and a foundation for the remaining analysis that helps avoid scope creep during the research phase \citep*{Putman}. A description should also constrain the problem to elements that can be realistically addressed with the resources available to the researcher.

Deconstruct the Problem

With the frame in place, analysis now moves to deconstructing the problem. During this phase, the problem is broken into smaller elements that reveal its perceived root causes. This occurs iteratively across two nested levels. The first is the identification of factors. Factors are broad categories that define and delineate the main causes of the problem. These can be thought of as the trunk of the metaphorical tree. Identified factors are further deconstructed into drivers. If factors are the trunk, then drivers are the roots below the trunk. More precisely, drivers are constituent elements of the factors that specify the conditions, actions, or events that are believed to cause the problem. Drivers are the central focus of the analysis because they reveal and interpret the root causes of the problem, thus providing areas of focus for corrective action through later research.
Identified drivers are usually unique to each factor and do not overlap. Overlapping drivers are an indication that their associated factors are too similar to one another. When considering the drivers, it is important to remember that this analysis is a heuristic of the problem, and that the characteristics of the drivers identified in the analysis will be highly sensitive to contextual factors and researcher interpretation. Given this understanding, an analysis of the same problem will likely generate different drivers if done by another researcher, or even the same researcher across multiple sites. A nominal analysis will likely generate two factors with three drivers each, assuming the problem has been appropriately framed. An off-nominal number of elements indicates that the frame of the problem should be revisited.