Heterogeneity of the effect of religion on income inspected

In this master thesis large micro datasets are used for the Netherlands, Germany, the United Kingdom and the United States of America to test for heterogeneity among these countries with respect to the effect of religion on income. Two types of measures of religion are used: membership and participation. Membership is defined as being a member of a Roman Catholic, Protestant or other Christian religion and membership as the number of times visiting a religious event. Income is defined as net household income and is only given in deciles. Therefore ordered probit estimations are applied. For both measures of religion heterogeneity is proved, since models in which heterogeneity is assumed are more efficient in estimating than models with homogeneity. The effect of membership is most negative in Germany, less negative in the Netherlands and insignificant in the United Kingdom. For the United States of America, the membership variable is not comparable with the other countries. The effect of participation (visiting church at least once a week) is positive for Germany and the Netherlands, insignificant in the United Kingdom and positive in the United States of America. Making a distinction between Roman Catholics and Protestants / Christians does not result in more efficient estimations so there is no significant difference between these denominations.