Establishing Cultural Relevance and Measurement Equivalence Using Emic and Etic Items (cross-Cultural, Philippines, Job Satisfaction)

149 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 1985.A basic concern of cross-cultural research is to be able to document cognitive, affective, and behavioral processes of individuals across a variety of cultural settings. This type of research involves gathering data in different cultural settings. In an effort to maintain standardization, measures and procedures that have been proven to be successful in previous undertakings are implemented in other cultural settings. The negative consequences of these procedures were described in de Vera and Park (1983), in Hulin and Mayer (in press), and are also described in the present thesis. The major focus of this thesis was to develop an alternative procedure that will fulfill the need of cultural relevance and standardization and, thereby, satisfy the goals of cross-cultural psychology. Responses from 1053 Filipino and 500 American employees provided the data base for the procedure. This procedure incorporates a stage that focuses on writing items that convey important emic concepts. This process provides richer data for a finer definition of the theory. An important contribution of this thesis is the development of a test construction procedure that appropriately addresses the accomplishment of the objectives of cross-cultural organizational research. Including both emic (culture-specific) and etic (culture-general) items in a measure generates a measure that is valid for both multi-cross-cultural comparisons and local cultural research. Item Response Theory (IRT) provides the procedure for equating two (or more) scales on to the same metric in order to make the appropriate multi-cross-cultural comparisons. IRT does not require that the scale contains exactly the same items for both versions. Therefore, comparisons can be made on scales that contain both etic and emic items.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD