**4. Methodology**

This part of the research describes the data used for analysis and the methodology applied for estimation of the results.

### **4.1 Data**

We draw data from the General Social Survey (GSS) of the USA in our empirical analysis. The General Social Survey is conducted by the National Opinion Research Center of the University of Chicago. The dataset is collected through a detailed personal interview survey of U.S. The GSS contains social and demographic characteristics of families. Most importantly, GSS is a rich dataset that contains rich information about people's religious activities of the sample. The religious variables include "how often the person attend religious services", "how many times a person prays every day", "allow anti-religionist to teach", etc. Besides, the GSS data also includes information about people's education achievement such as one's the highest degree. The GSS is not a panel data, which means that it does not follow the same objectives every year. The data from the year 2010, which is the most recent available dataset used for the analysis of this paper.


Table 1. Descriptive Table of Important Variables

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 respondent.
