European Policy Analysis
in only six applications . One application ( Elliot and Schlaepfer 2001a ; 2001b ) proposed a new hypothesis related to determinants of membership defection between coalitions .
In the 16 applications that focused on coalition identification , only ten directly discussed the belief systems of the coalitions . Eight of these identified deep core beliefs , nine identified policy core beliefs , and seven identified secondary beliefs . Out of these applications , seven identified policy-oriented beliefs at all three levels . It should be noted , however , that the majority of Swedish applications ( n = 16 , 64 percent ) did not empirically document belief systems . Six applications tested hypotheses related to the structure of advocacy coalitions and their associated belief systems . Strong support was found for the assumption that shared beliefs are a driving force in coalition structure ( Matti and Sandström 2011 ; 2013 ; Nohrstedt 2007 ; 2009 ). However , questions still remain regarding the nuances of the belief system , particularly which tier of the three-tiered structure is more influential in stimulating coalition formation and behavior . The ACF also hypothesizes that coalitions will be stable over a period of a decade or more ( Sabatier and Jenkins- Smith 1993 ). Through empirical testing of this hypothesis , Elliot and Schlaepfer ( 2001a ; 2001b ) found that there may be changes in the membership of coalitions but that these rare changes are often preceded by policy-oriented learning and / or significant external perturbations to the subsystem .
Beliefs are not the only unifying force within advocacy coalitions ; coordination and cooperation are also important mechanisms by which groups of like-minded actors work toward a common goal ( Schlager 1995 ). However , very few applications in Sweden explicitly discussed coordination or cooperation among policy subsystem actors . Only two applications identified strategies of collaboration ( Matti and Sandström 2011 ; Nohrstedt and Olofsson 2016 ). These same applications plus one additional application ( Matti and Sandström 2013 ) also identified cooperation strategies among advocacy coalitions . As was concluded by Weible , Sabatier , and McQueen ( 2009 ) and demonstrated again in this review , the role and influence of coordination and cooperation are areas in which the ACF has unexploited potential , particularly given the growth of networks as a form of political organization in Sweden .
Policy Change
In our sample of Swedish ACF applications , we found 16 applications that focused solely on policy change , with an additional three applications that addressed policy change tangentially . Eleven of those 19 applications identified instances of major policy change , defined in the ACF as a change that is related to the policy core aspects of a government program ( Sabatier and Jenkins-Smith 1999 ). The remaining applications described policy change as minor , which refers to changes in the secondary aspects of a government program . The explanations for the policy change , whether major or minor , were largely attributed to external shocks ( n = 14 ), sometimes to learning ( n = 11 ), and seldom to internal shocks ( n = 2 ). The relatively small number of applications
31