To capture both the technical aspects of the controllers’
tasks in providing accounting information for decisionmaking
and control purposes as well as the managerial user
perspective, we decided to adopt a dyadic research design.
This means that in each company we addressed both a
controller and a general manger (i.e., a member of upper
management like the CEO, managing director or division
manager) to fill out a functionally customized questionnaire.
Whereas the variable integration level of accounting
systems was surveyed with the controllers, the other three
variables consistency of financial language, controllership
output quality, and controllership impact on management
decisions were assessed by the managers’ answers. Each
dyad, i.e., controller and manager from the same company,
thus forms a unit of observation. Compared to a research
setting in which only controllers’ ratings are surveyed,
this enables us to draw valid conclusions with respect to
the management perspective on controllership effectiveness.
Hence, a possible common method bias (Podsakoff
et al., 2003) is reduced as the respective key informants
are addressed. Additionally, the managers’ questionnaires
were separate from the controllers’ questionnaires and did