The data collection involved a survey questionnaire
directed at 3500 managers of German companies. All companies
had more than 200 employees, and their selection
provided some degree of control over the size of the
companies in our sample. In surveying business unit and
functional managers, we were consistent with previous
studies on the controllability principle and focused on similarly
high-ranking managers (Giraud et al., 2008; Merchant,
1989; Simons, 2007). Following Giraud et al. (2008), we
excluded support function managers because their performance
is difficult to define and measure, which makes
the controllability principle less relevant. The managers
were randomly chosen from the Hoppenstedt database,
the market leader in Germany for providing professional
contact details of business managers for direct marketing
purposes.
Our survey instrument was web-based. We sent an individualized
e-mail to each manager with the request to
participate, a unique and anonymous link to the online
questionnaire, and a link to the web page of one of the sponsoring
universities. In designing the survey instrument, we
followed the guidelines provided by Dillman (2007), taking
several measures to increase both the quality of the
instrument and the response rate. First, we made the website
visually appealing. Second, whenever possible, we used
scales from the extant literature to measure our constructs.
Third, we drafted our questionnaire in English, translated it