Elite research mobilizes a wide range of methodologies, both qualitative and quantitative techniques (Hoffmann-Lange 2007), especially as the volume of available data sources has increased significantly over time (Genieys 2011). Elite research may be broadly classified into four substantive areas (Hoffmann-Lange, 2007). Social background studies collect data on family background (socioeconomic status of parents), regional background, religious affiliation, and education. This data allows us to compare the social backgrounds of elites to those of the general population, and comprehends a considerable theoretical significance, since they show to what degree the advancement to positions of power and influence is determined by the economic, social, and cultural capital of one's family. Social background studies are the most common type of elite research because of the easy availability of social background data and because they may also be the only data available, i.e. for historical elites or for elites which are otherwise not accessible for interviews (Hoffmann-Lange, 2001).
Second, another substantive area is elite careers: the more or less structured patterns of professional advancement that eventually lead into elite positions. Career patterns vary between sectors and organizations, depending on the qualifications that are considered important by selectorates at crucial career stages. The degree of professional specialization is a fundamental variable when comparing the elites of different sectors and organizations that allows us to assess if more emphasis is placed on specialized knowledge or rather on generalist qualifications acquired in different organizational contexts (Hoffmann-Lange, 2007).
The development of survey research has opened up two additional fields of elite research. Elite research makes it possible to study the activities, beliefs, values, and attitudes and reveal patterns of conflict and consensus among different elite groups and also makes it possible to study elite interactions and networks, which provide crucial information on the access of various elite groups as well as of non‐elites to central political decision makers (Hoffmann-Lange, 2007).
Researching political elites is not an easy task. Social scientist know a lot about political institutions, but much less who are those that lead these institutions, their cognitive framework, their beliefs, opinions and attitudes. And yet, these are relevant matters insofar politicians, understood in a broad sense as Higley and Burton (2006:7) do, decide about matters that are crucial for citizens’ lives and the quality of democracies.
There are three paths to identify politicians for research purposes (Putnam 1976:15-16). The first one is based on the reputation (Hunter 1959), the second relies on the analysis of decisions taken by/in institutions (Dahl 1961), and the third one is based on the positions occupied by politicians. Most recent studies of political elites use the positional analysis. It consists of identifying a particular political elite according to the positions defined previously in a representation or government institution (parliament, municipality). Thus, MPs or local councilors, members of government, will be part of the political elite.
When adopting the positional method, researchers face basically two choices—analyze the whole population or just a sample of it. Obviously, the choice will be determined by the nature of the research. But both options pose problems for researchers. This paper will delineate pros and cons of both choices by analyzing some of the most relevant studies carried out in Greece, Portugal and Spain.
Analyzing the whole population of political elites