European Policy Analysis Volume 2, Number 2, Winter 2016 | Page 33

European Policy Analysis
applications generally do not explicitly test any of the hypotheses ( Weible , Sabatier , and McQueen 2009 ).
The ACF was the only framework used in nine of the 25 applications . In an additional nine applications , the ACF was the main application used in conjunction with other frameworks or theories . In total , the ACF was the primary theoretical framework in 18 of the 25 applications . There was a wide spread of supplementary frameworks ; policy network theory was the most common but it was referenced in only four applications ( Elliot and Schlaepfer 2001a ; 2001b ; Olsson 2009 ; Olsson and Hysing 2012 ). Many supplementary frameworks appeared to be utilized based on their theory regarding actor behavior — such as resource dependency theory , public choice theory , and institutionalism — while other policy process theories — such as multiple streams , stages heuristic , punctuated equilibrium , and bottom-up implementation — were relatively common choices to complement the ACF .
Generalizability and Revisions
Our results show that eight applications explicitly discussed the generalizability of the ACF in Sweden . All of these eight applications addressed issues concerning descriptive validity , discussing whether the notion of advocacy coalitions is useful to describe actor constellations in the Swedish policy process and whether the ACF demonstrates an accurate description of the reality of the policy process . However , most applications addressed generalizability at a fairly broad level with reference to
H
As noted earlier , nine additional applications described groups of individuals as “ coalitions ”; however , these applications were excluded from the final sample as they did not acknowledge coalitions as a concept attributable to the ACF by referring to any of the foundational ACF works ( second inclusion criterion ). fundamental differences between politics in Sweden compared to , for example , the United States . These studies did not present empirical observations confirming or disconfirming applicability of the ACF ’ s assumptions in Sweden . Other case studies were more specific in identifying examples where the ACF has limited applicability . One recurrent observation concerned the prevalence of party politics , particularly the role of party discipline , in contested policy subsystems . As issues pose a threat to partisan interests , the importance of advocacy coalitions and their interactions becomes more limited as an explanation for policy change ( Nohrstedt 2005 ; 2009 ; Rossegger and Ramin 2013 ).
To address some of the shortcomings of the ACF in Sweden , authors suggested possible modifications . Overall , the modifications focused on developing a more specified model of the policy process . For example , Berggren ( 1998 ) recommended refining the concept of policy-oriented learning by co-opting insights from Heclo ’ s model . Elliot and Schlaepfer ( 2001b ) advocated developing the concept of coalitions by taking insights from policy networks and epistemic communities . Matti and Sandström ( 2013 ) and Olsson ( 2009 ) argued that the framework would be enhanced with more empirical work and theoretical distinction on the tripartite belief system . Finally , several authors offered suggestions for incorporating contextual factors , including a stronger emphasis on socioeconomic and historical developments ( Hysing and Olsson 2008 ) and partisan cleavages and political elite responsiveness to subsystem politics ( Nohrstedt 2005 ; 2009 ).
33