Concepts & Mesurement

Typologies explicate the meaning of a concept by mapping out its dimensions (Problematic case: Polity IV has 5 dimensions, but are collapsed into one dimension: It remains unclear how these different dimensions relate to each other. how do you decide whether the overarching concept is still met, and if so, if there needs to be a limitation of it).

Typologies guide our empirical analysis by clarifying concepts and pointing out good instances of the phenomenon (eg vote buying)

Typologies help us to sort cases into categories (parlamentary versus presidential democracies etc.) - with analytical "value added" (implications of parlamentarism).

The dimension and cell types of typologies may also help to identify and describe the phenomena under analysis (by means of complexity reduction).

Eg parliamentarism vs. presidentialism; Analytical implications: The relationship between parliament and government are very different. Classificatory implications: Why is Switzerland a quasi-presidentialist system (because they can not be voted out of office).

→ both the term and the typology provide analytical meaning - description has an analytical function (always!)

→ descriptive and explanatory typologies cannot be strictly separated. (eg Welfare state typology)

• "Pure" types (theoretical/hypothetical) and not a real world phenomena;
• Formed from characteristics and elements of given phenomena, but by stressing certain elements we can better understand the phenomena
  • most famously: Max Weber and the Protestant Ethic
• Hypothetical (ideational) construct: idea constructs that help us understand the (chaotic) world (very hight level of concept intension, no concept extension)
• Strong link to analyticism (who does not share the idea of world-mind dualism)
• Theoretically guide our thinking about the (real) world
  • example: Sweden as the prototypical social-democratic welfare state but not the ideal type.

• Multi-Dimensional conceptual classifications based on an explicitly stated theory;
• Cell types (outcomes) are to be explained by the rows and colums of the matrix;
• Rows and Colums of the matrix selected on the basis of theory;
• Tow crucial ontological assumptions underly (work with) explanatory typologies:
  • Configurational thinking (multidimensionality of empirical phenomena);
  • limited diversity: not all types exist in the real world
• Cases versus variables;
• Early stages of research project (but not necessarily).

Limited diversity means that not all "cases" exist in reality. It occurs because the empirical variation in which the social world presents itself tends to be highly limited in its diversity:

• logical impossibilities: Run counter to common sense (e.g. pregnant man, tropical village in Himalaya);
• clustered cases: Occurs because social reality is prestructured by historical, social etc. processes (path dependent) (eg Female US President, former German colony in Latin America);
• Mathematically imposed limited diversity: Number of types exceeds number of cases (eg five conditions (32types) in Study EU28).

→ you cannot avoid making assumptions about limited diversity! they might explain logic in the way reality is organized. A case might be exceptional but it has to be defined with regards to what it is exceptional. If something never appears without some other attributes, these attributes might not be independant and we might measure the same thing twice wich gives it more weight than other attributes have.

• Limited diversity: some "things" do not seem to exist - cell number increases esponentially;
• Some outcomes may be observed only in presence of a configuration of conditions;
• Some outcomes may be the result of different configurations of conditions (equifinality);
• Typologies help us detect patterns in the data, which we might struggl to see otherwise (imagine more dimensions!);

→ The procedure of analyzing typological data can be formalized: QCA (quantitative comparative analysis).