This paper demonstrates how to construct reliable scales for measuring attitudes to immigration by using a diversity methodology. It discusses underway the validity of the two most established presumptive criteria for creating quantitative scales: A dominant assumption in measurement theory is that there are common dimensions underlying multiple interrelated items. This paper asserts that this assumption may disable the resultant scales to represent certain "statistically less important" but ontologically interesting dimensions characterizing the very nature of a small number of cases in a sample, thus making the diversity in the data material invisible. This may have the consequence that one fails to uncover such dimensions, and treats them as if they were noise instead of regarding them as cases misrepresented by the respective scale. In order to solve the problem of making diversity invisible in comparative analysis, this paper proposes using a diversity methodology based on deployment of multiple ideal types.
Downloads: (external link) http://eurospheres.org/wp-content/themes/EuroSpher ... aper_19_Sicakkan.pdf Full text (application/pdf) Our link check indicates that this URL is bad, the error code is: 404 Not Found (http://eurospheres.org/wp-content/themes/EuroSphere/erpa//files/2010/08/Eurosphere_Working_Paper_19_Sicakkan.pdf [301 Moved Permanently]--> http://eurospheres.org/wp-content/themes/EuroSphere/erpa/files/2010/08/Eurosphere_Working_Paper_19_Sicakkan.pdf)
Related works: This item may be available elsewhere in EconPapers: Search for items with the same title.