Modeling church services supply and performance, using geographically weighted regression

Abstract: The objective of this study is to develop a multiple linear regression model that measures the relationship between the church services supply and the attendance to the services in the Uppsala diocese, Church of Sweden. By reviewing previous models and examining the nature of data available, two research questions were introduced, namely, the problem of omitted variables and the problem of spatial autocorrelation. For the first question, two methods were compared, namely, the Y-lag method and the first-differenced equation. Statistical tests then showed that the latter was more preferable for this study. For the second question, geographically weighted regression was used to examine the spatial variations in relationships estimated by above modeling strategies. However, no significant spatial variation was found for them. In conclusion, by using the ordinary least square estimation for the first-differenced equation the most suitable regression model was obtained. The data showed no need to consider the issue of spatial non-stationarity.