European Policy Analysis
Methods and Data
To compile a list of empirical applications of the ACF in Sweden , we searched three databases : Web of Science , Google Scholar , and Diva Portal ( a repository of research publications written at 37 universities and research institutions in Sweden ). Multiple databases were used to account for the limitations of the search capacity in each database as well as regional specificity . Each database was searched systematically using a combination of “ advocacy coalition framework ”, “ advocacy coalitions ”, and “ Sweden ” as key words in both Swedish and English . Explicit use of any of these keywords constituted the first criterion for inclusion in the dataset . Peer-reviewed journal articles , book chapters , and doctoral dissertations were included . Unpublished manuscripts , conference papers , and undergraduate- and master ’ s-level theses were all excluded . We included all applicable publications regardless of the date of publication . After the initial database search , we had identified 50 potential applications for coding . The final sample contains 25 publications in the period from 1998 ( first publication found ) to 2015 ( final search conducted in August , 2015 ). E Rules for inclusion in the final sample were ( i ) the title and / or abstract should explicitly mention “ advocacy ”, “ coalition ”, or “ learning ” and ( ii ) at least two foundational ACF works should be cited . F All applications that met this criterion were manually reviewed by authors to confirm inclusion . The majority of applications were published in peer-reviewed journals ( n = 18 ); five were published doctoral dissertations , with one book chapter and one report comprising the remainder of the sample .
Informed by prior efforts to review the state of the field of ACF scholarly work ( Jang , Weible , and Park 2016 ; Weible , Sabatier , and McQueen 2009 ), we applied a coding framework containing a total of 60 elements divided into eight main parts : ( 1 ) application attributes ( publication descriptives , e . g . author ( s ), year of publication ), ( 2 ) application scope ( e . g . purpose of using the ACF , empirical domain , level of analysis ), ( 3 ) research objective ( main areas of theoretical emphasis , e . g . coalitions , policy change , learning ), ( 4 ) methods ( data collection and analysis ), ( 5 ) coalitions ( number of coalitions , belief system , membership , strategies used ), ( 6 ) learning ( among or between coalitions , brokers ), ( 7 ) policy change ( minor or major change , role of learning , and / or external events ), and ( 8 ) overall evaluation ( generalizability and / or modifications discussed ). Based on these parts , we report results below focusing on substantive topic , application methods , level of government , and theoretical emphases . We also examine the extent to which the ACF has been used as the main framework or supplementary theory . Finally , we summarize scholars ’
E
Nine additional applications used “ coalitions ” as a description of groups of individuals or organizations sharing common beliefs or goals towards a particular issue or policy ; however , these applications included no other mention of the ACF nor did they cite foundational works . Therefore , they were excluded from the sample . This approach to identifying applications has been used in previous reviews ( see , e . g ., Jones et al . 2015 ; Weible , Sabatier , and McQueen 2009 ).
F
For this study and in accordance with previous meta reviews ( Weible , Sabatier , and McQueen 2009 ), foundational ACF publications included the following : Sabatier ( 1986 ; 1988 ; 1998 ), Sabatier and Jenkins-Smith ( 1999 ), Sabatier and Weible ( 2007 ).
25