Types Of Measurement Validation : Content, Convergent And Discriminant, And Nomological Esp Construct Validation

1321 Words Dec 15th, 2016 6 Pages
There are “three types of measurement validation”: content, convergent/discriminant, and nomological/construct validation. Content Validation is preoccupied with the questions whether or not key elements are “omitted from the indicator” and whether or not “inappropriate elements are included in the indicator.” Although content validation is an “important contribution” to measurement validity, its chief limitation is that “alone it is incomplete, for two reasons”: “first, although a necessary condition, the findings of content validation are not a sufficient for establishing validity” and second “the trade-off between parsimony and completeness that arises because indicators routinely fail to capture the full content of a systemized concept.” (539). Then there is nomological/construct: “In a domain of research in which a given causal hypothesis is reasonably well established, we ask: Is this hypothesis again confirmed when the cases are scored (level 4) with the proposed indicator (level 3) for a systemized concept (level 2) that is one of the variables in the hypothesis?” (542) Skepticism about nomological validation gives rise to concenrs such as “circularity” and “presupposes the valid measurement of other systemized concept involved in the hypothesis.” (543)

However, the type of measurement validation discussed by Adcock and Collier that is pertinent to this papers examination of methodology in Making Democracy Work is convergent/ discriminant. Convergent/discriminant…

Related Documents