The "ideal" in ideal type refers to its ideational character not its desirable character.

  1. Definitional issues
    1. First of all, they have nothing to do with moral ideals or values
    2. Not a statistical average
    3. An analytic concept representing the pure or extreme version of a phenomenon
    4. Definition: A mental construct used in social analysis that is composed of a set of characteristic elements of a class of phenomena. The elements are abstracted based on concrete observations of concrete instances of the phenomenon, but the resultant construct is not designed to correspond exactly to any single empirical observation. It is a heuristic device, that is, a tool for describing, comparing, and testing hypotheses about empirical reality. Ideal types are not meant as true pictures but as tentative models.
    5. There is a similarity between an ideal type and a stereotype, but stereotypes (1) are not constructed consciously for the purpose of scientifically abstracting reality, (2) often involve characterization of the whole based on one or a few characteristics, (3) are usually taken as substitutes for real empirical cases, and (4) are not constantly examined and refined.
    6. Examples: charismatic leader, bureaucracy, chapel, Calvinistic protestant, feudalism, economic man, marginal man, sect, church, Gemeinschaft, Gesellschaft, film, movie
  2. An attempt to get out of a methodological dilemma
    1. historicism which holds that everything is unique  no comparisons possible
    2. human sciences which gather too much under “religion” for example  too much detail lost
  3. Method – how to do it
    1. “An ideal type is formed by the one-sided accentuation of one or more points of view and by the synthesis of a great many diffuse, discrete, more or less present and occasionally absent concrete individual phenomena, which are arranged according to those one-sidedly emphasized viewpoints into a unified analytical construct.”
    2. Accentuate typical features or typical courses of conduct
      1. Examples: “the casual dresser” vs. “the style conscious dresser”
    3. Constructed out of elements of observed reality but formed into a logically precise and coherent whole
    4. Cf phenomenologist’s “eidetic variation” to arrive at “table” or “tree” or the scientist’s “maple tree”
  4. Three kinds/levels of ideal types
    1. Historical particularities (e.g., “the Western city,” “the Protestant ethic,” “modern capitalism”)
    2. Abstract elements of social reality (e.g, “bureaucracy,” “feudalism,” “we-relation”)
    3. Rationalizing reconstructions of a particular kind of behavior (e.g., propositions in economic theory, rational model of organizational behavior)
  5. Ideal types and meaningful action
    1. Cf. Schutz “construct of constructs”
      1. people typify to make sense of self and others
      2. scientist typifies typifications
  6. Utility – what are they good for?
    1. Construct hypotheses linking them to conditions that bring them into being or consequences that follow from them.

References

[http://en.wikipedia.org/wiki/Ideal_type]
("The Ideal Type" from Coser, 1977:223-224)