My quantitative study of the Ghost Dance movements and their effects involved several methodological procedures, as discussed in this appendix.

Sources of data

First, measuring the analytical variables to be used required the attainment of various types of data.

Participation

In studying participation in the Ghost Dance movements, I had first to determine which groups of American Indians knew of them, and when. Nonparticipation had little, if any, meaning if a group did not know about the movements, as I have indicated. Therefore I needed data about Ghost Dance knowledge initially. Data were then required about which tribes did or did not participate in either movement.

The information I used to establish which of hundreds of nineteenth-century American Indian peoples were aware of either movement and which participated in either or both was obtained from published, primarily scholarly, accounts of the Ghost Dances. Such accounts typically discuss the spread of the movements from tribe to tribe. They also typically contain information about tribes knowing of the movements but not participating. For each of the two Ghost Dances, there is a more or less standard scholarly reference: Du Bois's The 1870 Ghost Dance (1939) and Mooney's The Ghost-Dance Religion and the Sioux Outbreak of 1890 (1896). These works were my starting points for the respective movements; however, they were supplemented by other accounts.