It was suggested above that cross-national differences in selection method use could not be explained solely as a result of systematic differences in task environments. Rather, at least some cross-national differences in the adoption of organisational forms and practices may be more adequately explained in terms of the extent to which these forms and practices are close to wider differences in societal regimes, norms and practices and are articulated through institutions in the organisational environment. A 'societal effects' approach then predicts that those methods that are closer to societal norms and practices and which are articulated within prevailing institutions may be more widely diffused regardless of their 'fit' with local task contingencies (i.e. the job being recruited into) (Sorge, 1991). In the next section we look at some of these societal effects, focusing on differences in institutional networks, in the regulatory environment, in economic factors, and in national cultures. These are considered in more detail next.
Individuals within companies are unlikely to invent entirely new methods of selection and assessment. Rather, when changes are introduced to the methods used to select personnel, we are looking at a process of innovation. That is, the company will be introducing selection methods that have been developed and used elsewhere. There is a whole literature on the diffusion and adoption of new ideas that is relevant to understanding this process of innovation (Rogers, 1995). Exploring national differences in innovation processes (that is in the diffusion and adoption of new ideas) can, in turn, help to explain the national differences in selection method use that we have observed.
The innovation process must begin with an individual becoming aware of a new idea (in this case about a particular selection method). They must then be persuaded that this new idea is relevant and useful within their own context and persuade others within their company of this. A decision to adopt can then be made, which will lead to the implementation and use of the new method. Ideally, there will also be an evaluation of the success of the new method and changes introduced if the evaluation is not positive. Thus, in order to adopt a new selection method, an individual within a company has first to become aware that the new method exists, and then they must be convinced or persuaded that this method will improve the current selection process. This individual must then persuade others in the company so that a decision is made to adopt and implement the new method. In implementing the new selection method, it is likely to be modified to fit the particular organisational context of use, since as we have seen, selection methods (like other technologies) have a high degree of interpretative flexibility. This is why the literature on the adoption of new technologies talks about appropriation rather than simply implementation. Appropriation implies that the technologies are modified to suit the particular organisational context of use, rather than simply 'plugged in' and 'switched on'.
Institutional networks are important as they influence the diffusion of knowledge across a community of users. These networks typically have national boundaries. For example, educational institutions, a source of new ideas for many, typically draw both students and staff from a particular national context. One such network that may play a central role in this process of knowledge diffusion is that provided by a professional association. A professional association is an obvious arena for dominant professional norms and ideas about best practice to be developed and communicated. Many practitioners join professional associations precisely to learn about the latest 'best practice' in their particular domain as well as to enhance their career status (Lynch, 1989). Professional associations in the HR and organisational psychology domain have essentially three types of members. Firstly, there are members who work as practitioners in HR departments. Secondly, there are academics who teach and research in the related areas. Finally, there are members who work as suppliers of HR products and services, including selection methods (consultants and specialist selection advisers). Professional associations thus provide links between central suppliers of new ideas about 'best practice' in selection methods and their users. These networks will be important in influencing the diffusion and adoption of HR tools and practices (including selection methods) (Rogers, 1995; Swan & Newell, 1995). Differences between countries in the development and activities of professional associations may therefore account for some of the observed differences in selection method use across countries.
For example, Eleftheriou and Robertson (1999) report that human resource management in Greece is very underdeveloped with the main professional association (the Greek Personnel Managers' Association) having only 300 members in 1992 (Ball, 1992), In conjunction with this, the evidence shows that Greek firms tend to use subjective and intuitive methods (e.g. interviews, CV and personnel recommendations) and believe that these methods are more effective than, for instance, psychological tests (Kantas, Kalogera & Nikolaou, 1997). Even in North America, where arguably there is the most active research in the area, Rowe, Williams and Day (1994) concluded that many practitioners were not using 'state-of-the art' selection methods that were high in validity and reliability. One reason they gave for this was that there was a lack of awareness about valid assessment tools and technology due to a 'general dearth of specialised education and training in personnel departments' (p. 77). They quote Thacker and Cattaneo (1993) who found only a very small proportion of employees in personnel departments were graduates (28%). They argue that this, and other evidence they considered, suggests 'questionable professional competence' (p. 77).
In a similar vein, in attempting to understand differences in use of selection methods across Europe, Shackleton and Newell (1994) speculate that this might be related to 'the long-standing differences between countries in their intellectual and scientific traditions' (p. 101). This, they suggest, manifests itself in differential approaches to and teaching of psychology and management studies across European countries. For example, in Britain and
Germany psychology is more heavily influenced by the natural scientific tradition than in France and Italy. Moreover, in general in Germany and Britain the influence of occupational psychologists may be more significant than in France and Italy. This, they suggest, may help to explain the use of less 'scientific' methods in the latter countries compared to the former. Together this evidence suggests that individuals in different countries will be exposed to rather different ideas about 'best practice' in selection and that this in turn will influence the use of particular selection methods.