• 1. The following members of the 'Comparative Public Organization Data Base for Research and Analysis' (COBRA) network were involved in the collection of survey data used for this chapter Belgium (Flanders): Koen Verhoest, Bram Verschuere and Geert Bouckaert (University of Leuven);Portugal: Silvia Mendes (University of Minho);and the Netherlands: Sandra van Thiel (University of Nijmegen) and Kutsal Yesilkagit (University of Utrecht). We owe many thanks to these colleagues.
  • 2. Ragin, Charles and Sean Davey. 2009. fs/QCA [Computer Programme], Version 2.5. (Tucson: University of Arizona).
  • 3. This variable included also the composition of the board. The variable was constructed as follows: 0 = no board or a board entirely composed of government representatives (fully upward accountable board);1 = hybrid board, with government and civil society representatives; 2 = fully downward accountable board, with only civil society representatives (and without government representatives). Regulatory agencies with a hybrid board did not exist in our data sample, so we calibrated the index on the basis of a score 0 = no board or a board entirely composed of government representatives and 1 = fully downward accountable board.
  • 4. In FsQCA, including this category is not problematic as it refers to the most extreme form of agencification, which entails theoretically the largest degree of formal independence (calibrated to one). In order to check the impact of this 'outlier' we performed a second set of robustness tests with the sample excluding this single private law agency. Also, these tests were taken into account when writing the conclusions, but their results were not substantially different.
  • 5. An upper-case letter indicates the presence of a condition; a lower-case letter indicates the absence of a condition.
  • 6. The dataset and the truth tables are available upon request: astrid.molenveld@
  • 7. We refrained ourselves from making assumptions about any logical remainder. So, the intermediate solution is in fact the conservative solution.
  • 8. While 'high use of individual upward accountability mechanisms' is a necessary condition for 'high perceived accountability', we checked in the analysis of sufficient conditions for logical remainders that contradict the statement of necessity. The only logical remainder present did contradict the statement of necessity: absence of high use of individual upward accountability mechanisms AND presence of high use of organizational upward accountability mechanisms AND absence of high use of downward accountability mechanisms.
  • 9. The robustness tests with the stricter calibration of the outcome revealed 'high use of individual upward accountability mechanisms' as necessary condition, but no paths with sufficient conditions, which does not contradict our main conclusion that the presence of horizontal accountability mechanisms do not really foster a high level of self-perceived accountability towards society.
< Prev   CONTENTS   Source   Next >