Building Composite Measures of Religion and State
With increasing amounts of cross-country data on religion available, it is important to assess the optimal design of the variables in the datasets. This study assesses the strengths and weaknesses of existing cross-country data on religion. It then presents the Religion and State round 2 (RAS2) data...
| Published in: | Interdisciplinary journal of research on religion |
|---|---|
| Main Author: | |
| Format: | Electronic Article |
| Language: | English |
| Check availability: | HBZ Gateway |
| Journals Online & Print: | Drawer...
|
| Fernleihe: | Fernleihe für die Fachinformationsdienste |
| Published: |
[publisher not identified]
[2011]
|
| In: |
Interdisciplinary journal of research on religion
Year: 2011, Volume: 7, Pages: 1-39 |
| Online Access: |
Volltext (kostenfrei registrierungspflichtig) |
| Summary: | With increasing amounts of cross-country data on religion available, it is important to assess the optimal design of the variables in the datasets. This study assesses the strengths and weaknesses of existing cross-country data on religion. It then presents the Religion and State round 2 (RAS2) dataset as a potentially superior alternative to existing data collections on government religion policy. The comparison shows that the RAS2 dataset has a number of advantages over other data collections. It contains more variables covering a longer time period, and the data are based on a wider variety of sources than is the case with other data collections. The codings include scaling for the severity of a policy. RAS2 avoids methodological problems that are present in several other data collections. Finally, it is the only data collection on religion that can claim to include codings for all relevant government actions. Although previous datasets have some of these advantages, several are unique to RAS2, and the others are not present in all cross-country data collections of state religion policy that include indexes. The study also compares the RAS2 indexes, which weight all components equally, to indexes that are weighted on the basis of expert assessment and factor analysis and finds that the nonweighted indexes are statistically nearly identical to the weighted indexes. |
|---|---|
| ISSN: | 1556-3723 |
| Contains: | Enthalten in: Interdisciplinary journal of research on religion
|