Outcome pattern matching and program theory - Evaluation and Program Planning, Vol. 12, Issue 4

Pattern matching is presented as a general framework which can guide the use of theory within program evaluation. Pattern matching minimally involves the specification of a theoretical pattern, the acquisition of an observed pattern, and an attempt to match these two. Pattern matching logic assumes that more complex theoretical patterns, if corroborated, provide a stronger basis for valid inference. Pattern matches in program evaluation can be divided into two types: process pattern matches which assess the construct validity of the program, participants, or measures, and outcome pattern matches which assess the causal hypothesis and address the traditional concerns of internal and external validity. Each of the three types of process pattern matches can be further divided into characteristic pattern matches (which examine the interrelationships between key characteristics across programs, participants, or measures) or object pattern matches (which view interrelationships between programs, participants, or measures based on their overall degree of similarity). Outcome pattern matching can be accomplished for any process pattern match by examining outcomes across programs, participants, or measures viewed either in terms of their characteristics or as molar objects. Hypothetical examples of pattern matching in program outcome evaluation contexts are presented along with consideration of the value of pattern matching for theory-based research.