**3. Theoretical background**

The study of religious activity and education is mainly empirical. However, the empirical analysis is based on a sound theoretical background of economic literature of religion. This part states about the theoretical research related with the paper. Fan (2008) argues that people's religious participation is determined by the concern for their children's human capital accumulation as well as their religious beliefs. This part of the paper introduces the economics theoretical research of religion. As pointed out in Fan's paper, in recent decades, some important contributions have been made in modeling religion and religious behaviors. Azzi and Ehrenberg (1975) analyze a model of church attendance and contributions in which individuals allocate their time and money among religious and secular commodities to maximize lifetime and afterlife utility. Iannaccone (1990, 1998) applies Stigler and Becker's (1977) idea of "consumption capital" to explain rational habit formation in religious activities. In this framework, current religious participation increases an individual's stock of "religious human capital" and thereby increases the individual's utility from future participation. Iannaccone (1992) presents a model that accounts for the continuing success of

Religious Participation and Educational Attainment: An Empirical Investigation 325

Table 1 is the descriptive statistics about people's marital status, age, family income, and education. People on average have married for twice, the standard error and variance are quite low. The average family income is 30,813 and the median income is 23,310, the standard error for family income is big while the variance is small. The highest year of education for an individual is 14.64 years; the standard error and variance are both small. The highest year of education for their mother is a little lower: 11.55 years, the median is about 12 years; the standard error and variance are both very small. The highest year of education for the spouse is 13.73 years, the standard error and variance are both small. This means that there is no big differences between the education of the parents and their children, neither there is a big difference between the education of the spouse and the

In this research, two research questions are addressed. The first research question is: "Whether a person's education is related with his or her religious activities." The analysis of

> Education Variable: One's Highest Degree Important Religious Variables Religious Facts Religious Opinions

> > Fundamental Is He or She?

Table 2 demonstrate the logic of our analysis: we analyse the correlation between one's highest degree and his or her religious activities. The religious activities are divided into two categories: the first category is the fact concerning one's religious belief such as how often one prays, takes part in religious activities, and in which religion is one raised. The second category is one's opinions considering how religious one person is. The questions include how fundamental one considers him or her to be, the strength of affiliation, etc. The

Religious Preference Whether One Considers Him or Her Religious?

Strength of Affiliation

How

**4.2.1 The cross tab analysis of religious and educational variables** 

the first question could be expressed in the following table:

The Religion One Is Raised

corresponding results of the analysis will be shown in the result part.

Variable Name Mean Std. Error Median Variance Marital Status 2.41 0.036 2 2.557 Age 47.97 0.391 47 312.5 Family Income 30813.3 690.787 23310 0.0008 Highest Year of Education 14.64 0,588 11.33 0.00008 Highest Year Mother Completed 11.55 0.088 12 13.8 Highest Year Spouse Completed 13.73 0.102 14 0.0008

Table 1. Descriptive Table of Important Variables

respondent.

How often Pray?

**4.2 Empirical estimation** 

How Often Take Part in Religious Activities?

Table 2. The Structure Tree of Analysis

groups with strict requirements. In particular, the model shows that efficient religions with perfectly rational members may benefit from stigma, self-sacrifice and bizarre behavioral restrictions because deviant norms mitigate the free-rider problems faced by religious groups. Bisin and Verdier (2000) extend the study of religion into an intergenerational framework. Assuming that parents get more utility if the children adopt their religion, Bisin and Verdier (2000) present an economic analysis of the intergenerational transmission of religious traits through family socialization and marital segregation. Barros and Garoupa (2002) introduce spatial location models into the economics of religion. Dehejia et al. (2005) show that involvement with religious organizations insures an individual's stream of consumption and of happiness.

This study is also related to the human capital models based on social interactions. In analysing the relationship between religion and social factors, there are several important theoretical papers. For example, Borjas (1992) and Lundberg and Startz (1998) analyze models in which an individual's human capital is determined by the average level of human capital of the ethnic group to which she belongs as well as her own parental human capital. Benabou (1993) suggests that the neighborhoods with a high level of average parental human capital facilitate one's human capital formation. He shows that neighborhoods are formed endogenously, higher income people live in the communities whose rent and average level of human capital are both higher. Epple and Romano (1998) and Brock and Durlauf (2001b) posit that a student's academic achievement is determined by both her own ability/effort and mean ability/effort of her classmates. Epple and Romano (1998) show that in equilibrium, parents who have high income and high-ability children pay high tuition to send their children to private schools in which there is better peer-group externality than public schools. In summary, the existing theoretical literature implies that education is related with religious activities; besides, religious activities are related with other social factors. The following part of the empirical research aims to detect the relationship of education and religious activities, as well as the relationship between religious activities and other social factors.
