Goal & Concept

Goal

The goal of this exercise is to examine the relationship between demographic and socio-cultural factors and religiosity among American high school students. Crosstabulations, comparison of means, and graphs will be used.

Concept

Religiosity is the term used to refer to the importance of religion in a person's life. It includes religious identity, behavior, attitudes, perceptions, and practices.

Research has shown that religiosity has a positive influence on adolescents' lives and is associated with healthier diets, exercise, sleep habits, and self-esteem, as well as lower rates of alcohol and drug use, early sexual behavior, delinquency, depression, and suicide.

Less is known about how religiosity develops in children and adolescents however, though research suggests that socio-cultural factors such as racial identity, gender, parental marital situation and geographical location of residence influence the extent of teenagers' religiosity, as do parents' attendance of religious services, parents' level of education, household income, and parental presence in the household.

Examples of possible research questions about adolescent religiosity :

• What are the mechanisms by which religious beliefs develop during childhood and how are they related to teenagers' behaviors and attitudes?

• How are family structure, gender, race, and ethnicity related to adolescents' religiosity?

• How does the contextual environment (family, peers, school, and community) influence religious beliefs and behaviors in youth and through these religious beliefs and behaviors influence health risk behaviors in youth?

• How do parents' religious beliefs and practices relate to their parenting styles?

• How are religious beliefs transmitted across generations?

• Does exposure to multiple religions affect an individual's sense of religiosity?

• How are youths' perceptions of religion's proscriptions related to their health risk behaviors?

Dataset

Data for this exercise come from the Monitoring the Future (12th Grade Survey), 2006 (MTF). The 2006 MTF is part of an ongoing series of cross-sectional data collection designed to explore changes in important values, behaviors, and lifestyle orientations of contemporary American youth. Each year, large, distinct, nationally representative samples of 8th, 10th, and 12th-grade students in the United States are asked to respond to drug use and demographic questions, as well as to additional questions on a variety of subjects, including attitudes toward religion, parental influences, changing roles of women, educational aspirations, self-esteem, exposure to sex and drug education, and violence and crime both in and out of school. Data used for this exercise are restricted to the 2006 12th Grade Survey, Core Data. Funding for MTF is provided by the United States department of Health and Human Services, National Institute on Drug Abuse. The principal investigators are Lloyd D. Johnston, Jerald G. Bachman, Patrick M. O'Malley, and John E. Schulenberg of the University of Michigan Institute for Social Research, Survey Research Center.

The MTF 12th Grade Survey is a school-based sample designed to represent high school seniors in the contiguous United States. In this exercise, data from the Core dataset (DS1) are used.

This exercise will use the following variables:

• Respondent's sex (V150)
• Respondent's race (V151)
• Respondent's father living in household (V155)
• Respondent's mother living in household (V156)
• Father's education (V163)
• Mother's education (V164)
• Respondent's attendance at religious services (V169)
• Religion importance to respondent's life (V170)

Application

For this exercise, you will examine some of the factors that relate to the importance of religion in the lives of American teenagers. You will use crosstabulations and graphs for your analyses.

Importance of Religion among American Teens

For this analysis, you will use two different measures of the importance of religion in adolescents' lives. The first is v169, which we have renamed "ATTEND." ATTEND measures respondents' self-reported attendance of religious services. The response options are: "Never" (1); "Rarely" (2); "1-2 times per month" (3); "once per week or more" (4). You will also use v170, which we have renamed "RELIMPORT." This variable asks "how important is religion in your life?" with response options: "not important" (1); "a little important" (2); "pretty important" (3); "very important" (4).

Demographic Characteristics and Religiosity

Look at the crosstab of "ATTEND by v150 (sex)" . Do males or females seem to report more frequent attendance of religious services? What percentage of males reports never attending religious services?

Next use the second measure of religious importance. In the crosstab of RELIMPORT by v150 (sex) does religion appear to be more important to male or female teens? What percentage of females reports that religion is not important to them?

MTF's measure of race (V151) contains three possible response categories: "Black"; "White"; "Hispanic." Find out if religious service attendance varies by race . Compared to whites, are Hispanics more or less likely to report never attending religious services?

Next, run a crosstab of. RELIMPORT by V151 . Which of the three racial groups is most likely to say religion is not important to them? What percentage of whites reports that religion is either pretty important or very important to them?

Parents and Religiosity

Parent's education may be related to the importance of religion in one's life. Although it is possible to look at mother's education and father's education separately, if parent education is related to importance of religion, it is likely that it is a combined effect of the education of both parents rather than the education of just the mother or just the father. Creating a useful combined parent education variable requires a few steps. First, we recoded the existing father's education variable (v163) into a three category variable called "PEDU3" which groups those with less than high school education (1), those who have completed high school, but not a college degree (2), and finally those fathers with a college degree (3). We created a new variable for mother's education (MEDU3) using the same categories.

Using these new variables, we created a combined parent education variable called "PAREDUCOMBO5" by simply adding PEDU3 to MEDU3. The values of this variable range from "2" to "6" with a "2" meaning that both parents have less than a high school education, "3" meaning that one parent has less than a high school education, "4" could be that both parents have completed high school or that one parent has a college degree and the other less than high school, "5" meaning that both parents have completed high school and one parent has a college degree, and "6" meaning that both parents have college degrees.

Look at the crosstab of ATTEND by PAREDUCOMBO5 . Looking at the table as a whole, do those whose parents have higher levels of education seem to be more or less likely to attend religious services frequently?

Next, look at the crosstab of RELIMPORT by PAREDUCOMBO5 . Looking at the table as a whole again, do you see any pattern of either steady increase or steady decrease in religious importance as parents' education level increases?

It is also possible that parent's presence in the household may be related to teenagers' religiosity. We recoded variables V155 and V156, which ask whether the father (or male guardian) and mother (or female guardian) live in the same household as the respondent, into the new variables DADPRES and MOMPRES. Both are coded as binary variables where "0" means that parent is not present in the household and "1" means that the parent is present in the household.

For ease of analysis, we also created two new religiosity variables (recoded from ATTEND and RELIMPORT). The new attendance variable is coded as "0" for all respondents who do not attend religious services weekly and "1" if they do. The new importance variable is coded as "0" if they do not say that religion is a very important part of their life and "1" if they do. The new variables are called "RELATTEND2" and "RELIMP2," respectively.

Look at the comparison of means test with RELATTEND2 as the dependent variable and MOMPRES and DADPRES as the independent variables. What percentage of respondents attends religious service weekly in households where a mother is present, but a father is not? How about when both parents are present?

Now look at the comparison of means of RELIMP2 by MOMPRES and DADPRES. How do you interpret the .20 value in the cell at the intersection of the "no mother present" row and the "father present" column?

Interpretation & Summary

Think about your answers to the application questions before you click through to the interpretation guide for help in answering them.

Demographic Characteristics and Religiosity

• Do males or females report more frequent attendance of religious services?
• Is religion more important to male or female teens?
• Are Hispanics more or less likely to report never attending religious services than whites?
• Which of the three racial groups is most likely to say religion is not important to them?

Parents and Religiosity

• Are respondents whose parents have higher levels of education more or less likely to attend religious services frequently?
• Are higher levels of parent education associated with higher importance of religion?
• What percentage of respondents attends religious service weekly in households where a mother is present, but a father is not? How about when both parents are present?
• How do you interpret the .20 value in the cell at the intersection of the "no mother present" row and the "father present" column?

Interpretation

Things to think about in interpreting the results:

• When a survey or poll is administered, researchers go to great lengths to try to ensure that the sample is representative of a target population. If this is done well, then researchers can have a higher degree of confidence in saying that the results of the survey can be generalized to the population from which the sample was drawn. Since this survey was drawn from American high school 12th graders, the hope is that the results can be generalized to all American 12th graders and not just be used as a description of the sample.

• One potential problem when using inference to generalize the results to all American 12th graders is the amount of missing data in each relationship. If there seems to be a large amount of missing data, it is important to try to understand why this might be the case. Sometimes questions are sensitive and respondents choose not to answer. Other times, questions might not be asked to all respondents. One particular quirk in this dataset is that the researchers note that they have "obliterated" responses from the western region for the questions about religion, thus increasing the amount of missing data. This does not necessarily mean that the results cannot be generalized, but it does mean that we have some reason to be cautious about making generalizations to all American 12th graders.

• Reading the results: The numbers in each cell of the crosstabulation tables show the percentage of the respondents who fall into the overlapping categories, followed by the actual number of people that that represents in this sample. The coloring in the tables demonstrates how the observed values in each cell compares to the expected value if there was no association between the two variables. More specifically, if there is no association between variables, then the percentages across each cell in the row would all be equal. The accompanying bar charts display the patterns visually as well.

• In the comparison of means tables, the numbers in the cells represent essentially the same thing - the top number is the mean of the dependent variable for each category of the independent variable(s) and the bottom number is the actual number of people in that category in the sample. In these analyses, since the dependent variables (our measures of religiosity) are all coded "0" and "1," the mean in each cell also represents the proportion of people in that category who are coded "1" on the dependent measure. To translate a proportion into a percentage, simply multiply it by 100.

The analyses show the following:

• Females report a slightly higher level of religious service attendance than males. 34.1% of females attend once per week or more, compared to 30.2% of males. There is also a larger proportion of males who never attend religious services (20.5%) compared to females (16.8%).

• Females also seem to place a higher value on religion with 33% claiming that religion is very important in their lives compared with 26% of males. This is also evident when looking at those who claim that religion is not important; 14.3% of females versus 21.1% of males.

• Hispanics and Whites are almost equally likely (with 20% and 20.4% respectively) to report that they never attend religious services. Blacks (44.4%) are much more likely to report that they attend religious services than either Whites (30.7%) or Hispanics (25.3%).

• Similarly, if we look at importance of religion, Blacks (52.9%) are about twice as likely to report that religion is very important than either Whites (25.3%) or Hispanics (27.8%). Still, the majority (52.4%) of Whites responded that religion was either pretty important or very important. We found this value by adding the value in the "pretty important" cell (27.1%) to the value in the "very important" cell (25.3%).

• Respondents who have more highly educated parents seem to be more likely to attend religious services frequently. As we look across the row of those who attend religious services once per week or more, we find that the percentages increase as our combined parent education variable increases. If we look across the row of those who report that they never attend religious services, the pattern is not quite as prominent. Still, respondents whose parents were less educated ("2" and "3") are more likely to report never attending religious services than respondents whose parents are more highly educated ("5" and "6").

• Using importance of religion instead of religious service attendance to measure religiosity yields a very different picture. In fact, it does not appear as if there is any pattern, as the difference in between low and high levels of parent education is very small across each of the response options. This suggests that parent's education is not associated with how individuals value religion, but from the previous paragraph, parent education does seem associated with frequency of attending religious services.

• In general, when the mother is present in the household, we do see a higher percentage of respondents who attend religious services weekly and who value religion as a very important part of their life. Looking at the row totals of both tables, we see that attendance increases from 21% with no mother present to 34% when the mother is present. We also see the percentage of respondents that feel that religion is very important to them increases from 26% to 30% when the mother is present.

  Father's presence is more difficult to interpret with regard to religiosity. On the one hand, we do note that respondents from households where the father is present are 10% more likely to say that they attend religious services weekly than respondents from households without a father. However, when the mother is not present, father's presence does not seem to make any difference as 21% of respondents say that they attend religious services weekly both when the father is present and when the father is not present. When we turn our attention to importance of religion, 30% of respondents say that religion is very important to them when the father is not present as well as when present. However, when the mother is not present, respondents are 10% less likely (30% versus 20%) to say that religion is very important in their lives when the father is present in the household compared to households in which the father is not present. For both our attendance and our importance measures, the highest percentages are observed when both parents are present in the household (36% attend weekly religious services and 30% consider religion a very important part of their life).

Summary

The goal of this exercise was to examine potential influences on the religiosity of American high school students. We considered differences between males and females as well as across racial groups (defined as "Black," "White," and "Hispanic") on two different measures of religiosity, namely how often religious services are attended and how important religion is in the respondent's life. We also considered potential influences from the respondents' parents including amount of education the parents received and the presence/absence of mother/father in the household. The differences across demographic categories are easy to see and interpret and are relatively consistent across our two measures of religiosity. The results using the parent education and presence variables are a bit more difficult to interpret, but we can make the following statements:

• The amount of parent's education does seem to have an influence upon attendance of religious services, but not any significant influence upon the importance of religion.

• Although the absence of either parent appears to be associated with lesser religiosity, mother's presence in the household appears to be more strongly associated than father's presence with regard to both attending religious services and the importance of religion in the respondent's life.

Religion among Teens: A Data-Driven Learning Guide

Bibliography

We have compiled a list of references (http://www.icpsr.umich.edu/cgi/CITATIONS/search?study={id}&method=study&path=OLC) that might be useful to instructors and students wishing to further explore this topic. All were chosen because they relate to the topic of study, whether or not they use the specific dataset that was used in this exercise. Some relate directly to the concepts as defined by the exercise, others explore the topic more broadly either conceptually or empirically. For even more resources, try a key word search in the ICPSR Bibliography (http://www.icpsr.umich.edu/ICPSR/citations/)!

---

Goal & Concept - Dataset - Application - Interpretation & Summary - Bibliography