This report summarizes the results of a study of analysis reliability based
on current file production at the Human Relations Area Files. A total of
1,357 replicated pages were analyzed for this study, resulting in an average
level of absolute coding agreement of 52 percent between the marker (the
primary or original analyst) and the replicator (the secondary analyst) for
the fifty sources that constitute the replication unit. In addition, a new
method of measuring reliability is introduced, which indicates the level of
agreement for major or dominant categories. This method resulted in an 82
percent level of agreement.
Get full access to this article
View all access options for this article.
References
1.
Beierle, John M.1974 "Analysis reliability reports: replication unit number 1—Korea & Blackfoot," Behavior Science Research9: 327-29.
2.
Beierle, John M., and Stanley Witkowski1974 "HRAF coding reliability," Behavior Science Research9: 57-65.
3.
LeBar, Frank M.1970 "Coding ethnographic materials," in Raoul Naroll and Ronald Cohen, eds., A Handbook of Method in Cultural Anthropology, Garden City, N. Y., Natural History Press, 1970: 707-20 (reissued 1973 by Columbia University Press).
4.
Murdock, George Peter1975Outline of World Cultures, 5th ed., New Haven, Human Relations Area Files.
5.
Murdock, George Peter, et al. 1971Outline of Cultural Materials, 4th rev. ed., 5th printing, with modifications, New Haven, Human Relations Area Files.
