Constant Comparative Method

The constant comparative method is an inductive data coding process used for categorizing and comparing qualitative data for analysis purposes. Developed by sociologists Barney Glaser and Anselm Strauss in 1965, it is usually associated with the methodology of grounded theory, although it is widely used with other research and evaluation frameworks as well. Theory developed using the constant comparison method is considered “grounded” because it is derived from everyday experience as constituted by the data. True to its roots in symbolic interactionism, inductive analysis enables the investigator to build an understanding of the phenomena under investigation through the lives, relations, actions, and words of the participants themselves.

The constant comparative method is an ideal analytic tool for evaluators using qualitative or mixed methods. Because different stakeholders often support different values, understandings, and perceptions of programs and policies, this approach provides a basis for systematically organizing, comparing, and understanding the similarities and differences between those perceptions. A unit of data (e.g., interview transcript, observation, document) is analyzed and broken into codes based on emerging themes and concepts, which are then organized into categories that reflect an analytic understanding of the coded entities, not the entities themselves. The essential feature of this method is that each unit of data is analyzed and systematically compared with previously collected and analyzed data prior to any further data collection. Purposeful sampling is consistently employed in this iterative process to solicit data variations that exhaust all angles of a topic.

Although this method focuses on the subjective experiences of individuals and is considered to be qualitative, it is not bound to any one paradigmatic orientation. Glaser and Strauss assumed a correspondence between generated explanatory propositions and real-life incidents, although others who use this method do not. For example, Fourth Generation Evaluation, developed by Egon Guba and Yvonna [Page 81]Lincoln, uses the constant comparative method within a radical constructivist framework. In their model, the constant comparative method is reenvisioned as a hermeneutic dialectic process: a dynamic tool for gathering and generating participants' constructions of the practice being evaluated. Much as in the constant comparative method, a respondent is interviewed and asked to describe the nature of her or his perceptions of the evaluand. Immediate analysis follows, for the purpose of grasping the respondent's point of view and to acquire material to include in subsequent interviews. The second respondent is interviewed first for his or her personal viewpoint and then for comments on the first respondent's characterizations. This process is repeated as long as new perspectives are brought into the dialogic process. The aim is not to develop explanatory theories but to enable a group of divergent stakeholders to jointly construct a deeper level of understanding of the practice or program being evaluated.