Establishing qualitative geographic sample size in the presence of spatial autocorrelation

Auteur(s) et Affiliation(s)

School of Economic, Political and Policy Sciences, Univ. of Texas, Dallas, Etats-Unis

Description :
This article analyses the qualitative sample size determination for many purposeful sampling designs. It demonstrates that spatial autocorrelation can have a noticeable effect on sampling, whether it is quantitative or qualitative in nature. The results show firstly, that many sample sizes reported in the literature to dateappear to be inadequate to ensure that saturation was achieved. Secondly, that determining the level of positive spatial autocorrelation for a selection variable allows aggregate areal units to be targeted that will better suit purposeful sampling designs. And thirdly, that because because snowball sampling tends to be guided by positive spatial autocorrelation, communities should be targeted that display very high levels of such autocorrelation.

Type de document :
Article de périodique

Source :
Annals of the Association of American Geographers, issn : 0004-5608, 2013, vol. 103, n°. 5, p. 1107-1122, nombre de pages : 16, Références bibliographiques : 2 p.

Date :

Editeur :
Pays édition : Etats-Unis, Washington, DC, Association of American Geographers

Langue :
Droits :
Tous droits réservés © Prodig - Bibliographie Géographique Internationale (BGI)