Book Reviews: Revisiting Realistic Evaluation
Book Reviews: Revisiting Realistic Evaluation
Book Reviews: Revisiting Realistic Evaluation
Book Reviews
Chris L. S. Coryn
Albeit some might argue that this review is a little late in coming, it is worth
revisiting Pawson and Tilley’s 1997 book, Realistic Evaluation (reprinted in 1998,
2000, 2001, and 2002) as the debate about causation and evidence-based research
and evaluation continues to be a topic of debate and concern in the evaluation and
research communities (see A Call to Action: The First International Congress of
Qualitative Inquiry and The Claremont Debate, in this issue of JMDE). Realistic
Evaluation is rooted in the tradition of scientific realism, which is said to be one of
the “dominant axes in modern European thinking” (p. 55). In the most general of
terms scientific realism concerns “the nature and operation of causal forces” (p.
55). The essential ingredients for assessing these causal forces are C-M-O
configurations—where C represents context, M represents mechanisms, and O
represents outcomes. Context refers to the “spatial and institutional locations of
social situations, together, crucially, with the norms, values, and interrelationships
found in them” (p. 216). Mechanisms are the “choices and capacities which lead to
regular patterns of social behavior” and the causal mechanisms which generate
these patterns of behavior are “deemed ‘social problems’ and which are the
rationale for a program” (p. 216). Outcomes “provide the key evidence for the
In Chapter 3, In With the New: Scientific Realism, the authors present the
principles and practice of scientific realism. As previously mentioned, the realist
view (generative) of causation can be described thusly (as illustrated by the
explosion of gunpowder):
Our basic concern is still, of course, the outcome (the spark causing the
explosion). But what does the explanatory work is first of all the mechanism (the
chemical composition of the substance which allows the reaction), and secondly
the context (the physical conditions which allow the mechanism to come into
operation). This proposition—causal outcomes follow mechanisms acting in
contexts—is the axiomatic base upon which all realist explanations build.
context
mechanism
outcome
Source: Pawson, R. and Tilley, N. (1997). Realistic evaluation. Thousand Oaks, CA: Sage.
Mechanisms (M)
Contexts (C)
Outcomes (O)
Theory
Observations
Multi-method data
collection and analysis
on M, C, O
Source: Pawson, R. and Tilley, N. (1997). Realistic evaluation. Thousand Oaks, CA: Sage.
The first case study presented is an evaluation of property marking and described
by the authors as testing theory, the second is an evaluation of a housing project
and described as theory formation and development, and the third is an evaluation
of a prison-delivered higher education program also described as theory formation
and development. These case studies are described in some detail and are intended
to represent exemplars of realistic evaluation.
I have opted to exclude a review of the remaining chapters (5-9) as these merely
focus on collecting realist data and the methodological procedures involved in
conducting evaluation as prescribed by Pawson and Tilley.
Despite the book’s title, the true underlying premise of Pawson and Tilley’s
Realistic Evaluation is not merely a proposition of how to conduct evaluation, but
rather a treatise on the nature of causation and science. While the author’s notion
of causation (scientific realism) is compelling, I am not entirely convinced that it is
the “final solution” to the causation debate. Neither is it a dramatic improvement
over either successionist or other traditions. In their haste to prescribe generative
explanations they fail to recognize or acknowledge that numerous experimentalists
(and non-experimentalists) give considerable attention to context in their accounts
of causation (e.g., moderators, mediators, interaction effects), often to a greater
degree than the examples provided throughout the book suggest. Moreover, these
causal accounts (i.e., realist accounts) seem little more than explanations of
program effectiveness for different groups or consumers, which can be
accomplished without the use of realist principles.
Prior reviews (Patton, 1999; Rogers, 1999) of Realistic Evaluation have been
mixed. For example, Rogers (1999) stated that “this is one of those rare books that
has the potential to permanently change one’s perspective on program evaluation”
(p. 381). Patton (1999), on the other hand, was not entirely convinced of the
credibility of Pawson and Tilley’s contribution and responded to their criticisms of
utilization-focused evaluation (p. 14) thusly:
While Realistic Evaluation has spurred serious interest and debate, and even
spawned an issue of New Directions for Evaluation (Henry, Julnes, & Mark,
1998), the approach has not quite received the attention in North America that it
has in the United Kingdom and Europe. A search of the American and Canadian
evaluation journals did not turn-up any publications related to the approach (with
the exception of Patton and Roger’s reviews of the book). While a search of the
major European evaluation journal (Evaluation: The International Journal of
Theory, Research, and Practice) returned 56 articles which focused on, or
emphasized, the realistic evaluation approach.
References
Pawson, R. and Tilley, N. (1997). Realistic evaluation. Thousand Oaks, CA: Sage.