Area experts" images of African nations; a test of reputational measurement approach

Existing cross-national data are considered somewhat limited in their scope and often uncertain in their reliability. It has been suggested that these data might be supplemented by having a sample of African area experts code 39 African nations on several scales tapping attributes central to comparative political analysis and especially relevant to democratic theory. Expert agreement was good on highly familiar items such as regime type and role of the military. However, on more abstract scales and for more "exotic" nations, both dissensus and nonresponse increased considerably. There are several specific theory and measurement problems which must be confronted to adapt the judgmental measurement approach satisfactorily. The major difficulty is that of nonresponse. Another problem area is the specification of individual scales and the meaning of individual scale positions. At the same time, even in this preliminary test, the results allow a useful general characterization of regime types, role of the military, democratic development, corruption, and other political traits in Africa.