Religion_4ER – Exploring Further the Relationship between Religion and Attitudes toward Environmental Laws and Regulations

Note to the Instructor: This is the second in a series of two exercises that focus on the relationship between religion and attitudes about environmental laws and regulations.  In these exercises we're going to analyze data from the Pew 2014 Religious Landscape Survey conducted by the Pew Research Center.  We're going to use SPSS to analyze the data.  This exercise uses frequency distributions, two and three-variable tables, Chi Square, and measures of association as our statistical tools.  A weight variable is automatically applied to the data set so it better represents the population from which the sample was selected.  You have permission to use this exercise and to revise it to fit your needs.  Please send a copy of any revision to the author so I can see how people are using the exercises. Please contact the author for additional

Goal of Exercise

The goal of this exercise is to introduce three-variable (i.e., multivariate) data analysis.  In the previous exercise (Religion_3ER) we explored the relationship between religion and how people felt about environmental laws and regulations.  We discovered that Christians who were more literal in their reading of the Bible were more likely to think that stricter environmental laws and regulations hurt the economy.  In this exercise we're going to elaborate this relationship by controlling for other variables.  We'll use three-variable crosstabulations, percentages, Chi Square, and measures of association as our statistical tools. 

Part I – The Data Set We'll be Using

The Pew Research Center has conducted a number of surveys that deal with religion.  Two of these surveys are the Religious Landscape Surveys conducted in 2007 and then repeated in 2014.  They were very large telephone surveys of about 35,000 adults in the United States.   For more information about the surveys, go to their website

We'll be using a subset of the 2014 survey in this exercise which I have named Pew_2014_Religious_Landscape_ Survey_subset_for_classes.sav.  For the purposes of these exercises I selected a subset of variables from the complete data set.  I recoded some of the variables, created a few new variables, and renamed the variables to make them easier for students to use.  There is a weight variable which should always be used so that the sample will better represent the population from which the sample was selected.  To open the data set in SPSS, just double click on the file name. [1]  Your instructor will tell you where the file is located.

Part II – Environmental Laws and Regulations

The Pew survey asked respondents to choose between two claims.  The first claim is that "stricter environmental laws and regulations cost too many jobs and hurt the economy" while the second claim is that "stricter environmental laws and regulations are worth the cost."  Respondents were asked to choose the claim that comes closest to their own opinion.  Let's start by finding out how respondents answered this question.  If you haven't opened the data set yet, open it now.  Run a frequency distribution for the variable ENV1 which is the name of the variable.  The variable name starts with the letters ENV which tells you that this variable describes how people feel environmental laws and regulations.  Some of you have used SPSS, the statistical package we're using, and know how to get a frequency distribution.  Others of you are new to SPSS.  There is a tutorial that you can use to learn how to get a frequency distribution.  The tutorial is freely available on the Social Science Research and Instructional Center's website.  Chapter 1 of the tutorial gives you a basic overview of SPSS and frequency distributions are covered in Chapter 4. 

It's easy to get frequency distributions.  Once you have opened the data set in SPSS, look on the menu bar at the top and click on "Analyze."  This will open a drop-down menu.  Click on "Descriptive Statistics" and then on "Frequencies."  Notice that the list of all variables is in the pane on the left.  Select ENV1 by clicking on it and then click on the arrow pointing to the right.  This will move ENV1 into the "Variable(s)" box.  Now all you have to do is click on "OK" to get your frequency distribution.[2] 

The frequency distribution tells you how respondents answered this question.  The difference between the percents and valid percents in the table is important.  Percents are based on everyone in the sample while valid percents are based on only those who gave a valid answer.  Notice that some respondents said they didn't know or refused to answer this question and others volunteered that they agreed with both or neither of these claims.  These are called missing data because we don't know how they feel about same-sex marriage. These respondents are given missing codes which for this variable are the value "3" and "9".  Valid percents are computed by removing these respondents from the base for the percent.  To make sure you understand the difference between the percents and the valid percents, answer the following questions.

  • What is the percent for those who think that stricter environmental laws and regulations cost too many jobs and harm the economy?  What does this mean?
  • What is the valid percent for this category?  What does this mean?

Why aren't the percent and valid percents the same?

Part III – Biblical Literalism and Environmental Laws and Regulations

In the previous exercise (Religion_3ER) we explored the relationship between religion and how people felt about environmental laws and regulations.  One of the ways we can classify individuals is the way they view the sacred texts of their religion.  Do they see these texts as the literal world of God or do they feel that they should not be taken literally?  Additionally, some may view them as books written by humans and not by God.  The Pew survey asked "Which comes closest to your view?  Sacred text is the word of God OR sacred text is a book written by men and is not the word of God?"  This is variable RBL6.  A follow up question asked, "And would you say?  Sacred text is to be taken literally, word for word OR not everything in the sacred text should be taken literally, word for word?"  This is variable RBL7.  The term "sacred text" was replaced with the name of the sacred text for the respondent's religion (i.e., the Bible or the Torah or the Koran or the Holy Scripture).

Run frequency distributions for RBL6 and RBL7.  I created a variable that is a composite of these two variables and called it RBL7R1.[3]  Run a frequency distribution for RBL7R1.  Now let's look carefully at how that composite variable was created.  Value 1 in RBL7R1 is for respondents who said their sacred text is the literal word of God (see value 1 in RBL7).  Value 2 is for respondents who said their sacred text is the non-literal word of God (see value 2 in RBL7).  And value 3 is for those who said their sacred text is not the word of God (see value 2 in RBL6).  The missing value of 9 is for those who had missing information on RBL6 and RBL7. 

What we're interested in is whether respondents think their sacred text should be taken literally word for word.  Often literalists say the word of God is inerrant.  That means that it is without error.  So I combined values 2 and 3 in RBL7R1 and created a new variable called RBL7R2.  Run a frequency distribution for RBL7R2 and make sure you understand how it was created out of RBL7R1.

Let's limit our analysis to respondents who identified their religious preference as Christian.  We need to select out those respondents and limit our analysis to them.  To do this, click on "Data" in the menu bar at the top of the screen and then click on "Select Cases" in the drop-down menu.  (See Chapter 3, Transforming Data in the online SPSS book cited on page 1 of this exercise.)   Select "if condition is satisfied" by clicking on its circle and then click on the "if" button below.  Enter the specification for the cases you want to select.  Your specification should read "R5 < 50000".  Don't enter the quotation marks.  Now click on "Continue" and then on "OK".  To make sure you did this correctly, run a frequency distribution for R5.  This time you should only see categories for Christians and values less than 50,000.  If you made a mistake, you'll need to do it again.  You should also notice that some of the cases (i.e., the non-Christians) in the data window have been lined out. 

Run a crosstabulation showing the relationship between RBL7R2 and ENV1 being sure to get the correct percents, Chi Square, and Kendall's tau-b. (See Chapter 5, Cross Tabulations in the online SPSS book cited on page 1 of this exercise.)  You're going to put your two variables (i.e., RBL7R2 and ENV1) in the "Row(s)" and "Column(s)" boxes by clicking on the variable in the left-hand pane to select it and then clicking on the arrow that points to the right.  When you do that, the arrow will change so it points left.  If you click on it again, it will move the variable back to the left-hand pane.  That way you can correct errors you might make when you select the wrong variable.

But which variable goes in which box?  Typically, we put the independent variable in the column box and the dependent variable in the row box.  So we're going to put ENV1 in the row box andRBL7R2in the column box.  We're also going to click on the "Cells" box and check the box for the "Column" percents.  If your independent variable is in the columns, then you want to use the column percents.  If it is in the rows, then you want to use the row percents.  To get the table, click on "Continue" and then on "OK." 

There are two numbers in each cell of the table.  The top number is the number of cases in each cell and the bottom number is the column percent.  Notice that the column percents add down by column to 100.  Since the percents sum down to 100, you want to compare the percents straight across.  Always compare the percents in the direction opposite to the way they sum to 100. 

We're going to use Chi Square to help us interpret the table.  Chi Square is a test of significance that tests the null hypothesis that the two variables are unrelated to each other.  In statistical speak, we would say that the null hypothesis is that the variables are statistically independent.  Chi Square tests this null hypothesis and tells you whether you should reject or not reject it.  If you can reject it, then you have evidence that the two variables are related to each other.  If you can't reject it, then you don't have any evidence of a relationship.  

We're also going to use a measure of association.  A measure of association is a statistic that measures the strength of the relationship.   The Chi Square test doesn't tell you anything about the strength of the relationship.  You need a measure of association to do that.  There are many different measures of association.  Kendall's tau-b is used when both of your variables are ordinal.  Ordinal means that the categories have an inherent order to them.  In other words, they are ordered from high to low or from low to high.

To get Chi Square and Kendall's tau-b click on the "Statistics" button and then click on the boxes for both Chi Square and Kendall's tau-b.  Click on "Continue" and then on "OK" to get the table.

Now the question is how to interpret Chi Square and Kendall's tau-b.  To interpret Chi Square look at the first row for "Pearson Chi Square" and the column for "Asymptotic Significance."  In your output, it should read ".000".  This is the probability that you would be wrong if you rejected the null hypothesis.  It's actually not 0, but rather it is less than (<) .0005 since it's a rounded value.  That tells you that it's very unlikely that this is a chance relationship.  There probably is some relationship between these two variables.  Our rule is to reject the null hypothesis when the significance value is < .05.  In other words, when the probability of being wrong is less than five out of one hundred.

To interpret Kendall's tau-b look at the value in your table.  Think of a continuum from 0 (no relationship) to 1 (strongest possible relationship).  Measures of association are useful when comparing tables to see which table has the stronger or weaker relationship.  Tau-b varies from 0 (no relationship) to 1 (strongest possible relationship). Tau-b can be either positive or negative.  However, for this exercise ignore the sign when you interpret tau-b.

Write a paragraph that summarizes the relationship between biblical literalism and attitudes toward environmental laws and regulations.  Write one or two paragraphs explaining what the relationship is between these two variables.  The first part of your answer should explain in words (without using the percents) what the relationship is and the second part should use the percents to illustrate the relationship.  Be sure to also use Chi Square and Kendall's tau-b in your answer.  Make sure that you answer the following questions and use the valid percents.

Were people who are biblical literalists more or less likely to think that stricter environmental laws and regulations harm the economy?  Use the column percents to illustrate your answer.

What does the Chi Square test tell you about this relationship?

What does Kendall's tau-b tell you about the relationship?

Part IV – Spuriousness due to Age

At this point we have only considered two variables.  We need to consider other variables that might be related to both biblical literalism and attitudes toward environmental laws and regulations.  For example, age may be related to both these variables.  Older respondents may be more likely to be biblical literalists and to think that stricter environmental laws and regulations harm the economy.  This raises the possibility that the relationship between biblical literalism and how one feels about environmental laws and regulations might be due to age.  In other words, it may be spurious due to age.

Let’s check to see if age is related to both our independent and dependent variables.  This is important because the relationship can only be spurious if the third variable (age) is related to both your independent and dependent variables.  Use CROSSTABS to get two tables – one table should cross tabulate D6R2 (recoded age) and RBL7R2 (biblical literalism) and the other table should cross tabulate D6R2 and ENV1.  Limit your analysis to Christians.  Be sure to get the percents, Chi Square, and Kendall's tau-b.  If age is related to both variables, then we need to check further to see if the original relationship between biblical literalism and how people feel about environmental laws and regulations is spurious as a result of age.

Write a paragraph describing the relationship between age and your independent and dependent variables.  Remember to use the percents, Chi Square, and Kendall's tau-b in your answer.

Since age is related to both variables we need to check on the possibility that the relationship between biblical literalism and how people feel about environmental laws and regulations is due to the effect of age on that relationship?  What we're going to do is separate respondents into different age categories and then look at the relationship between biblical literalism and attitudes toward environmental laws and regulations separately for each age category.  In effect, we're going to see what happens then we hold age constant.  We can do that in SPSS by running a crosstab with RBL7R2 in the column (our independent variable), ENV1 in the row (our dependent variable), and D6R2 in the third box down in SPSS.  (See Chapter 8, Multivariate Analysis, in the online SPSS book mentioned on page 1 of this exercise.)  In this case, age is the variable we are holding constant and is often called the control variable. 

If the original relationship is spurious, then it either ought to go away or decrease substantially for all four age categories.  Look carefully at the four tables (i.e., one table for each age category).  How can we tell if the relationship goes away or decreases for each age category?  One clue will be the percent differences.  Compare the percent differences between those who are biblical literalists and those who are not for each of the four age categories with the percent difference in the original two-variable table.[4]  Did the percent differences stay about the same or did they decrease substantially?  Another clue is your measure of association.   Did Kendall's tau-b stay about the same or did they decrease substantially from that in the original two-variable table?

If the relationship had been due to age, then the relationship between biblical literalism and opinion on environmental laws and regulations would have disappeared or decreased substantially for all age categories.  In other words, the relationship would be spurious.  Spurious means that there is a statistical relationship, but not a causal relationship. It's important to note that just because a relationship is not spurious due to sex doesn’t mean that it is not spurious at all.  It might be spurious due to some other variable.

Write a paragraph describing the relationship between biblical literalism and attitudes toward environmental laws and regulations for each age category.  Now write another paragraph discussing whether this relationship is spurious due to age.  Be sure to describe how you came to your conclusion.  Remember to use the percents, Chi Square, and Kendall's tau-b in your answer.

Part V – Another Use of Control Variables

This process of introducing other variables is often called elaboration.  We are elaborating our two-variable analysis by introducing another variable into the analysis.  This is also referred to as multivariate analysis.  One reason to introduce other variables into the analysis is to check for spuriousness as we did in Part IV. 

It's possible that the relationship between two variables is different for some categories of individuals than it is for other categories.  We know from Part 3 that biblical literalists are more likely to think that strict environmental laws and regulations harm the economy.  We also know that individuals belong to many different religions.  Run a frequency distribution for R5 which is religious preference.  The four categories with the most cases are Evangelical Protestants (value 1), Mainline Protestants (value 2), Historically Black Protestants (value 3), and Roman Catholics (value 4).  To find out what the Pew Center means by these traditions, read the following Pew reports:

  • Chapter 1 of the full report for the 2014 Religious Landscape Survey on "The Changing Religious Composition of the U.S. Population" and
  • Appendix B to this report on "Classification of Protestant Denominations."

For more information on the difference between the evangelical and the mainline Protestant traditions, read the article by John Green in the PBS Frontline article on "Evangelicals v. Mainline Protestants."  For a history of the black church, read Marilyn Mellowes' article on "The Black Church."

Let's see if the relationship between biblical literalism (RBL7R2) and opinion on environmental laws and regulations (ENV1) is the same for Evangelical Protestants, Mainline Protestants, Historically Black Protestants, and Roman Catholics.  To do this we're going to select out only these four groups.  Click on "Data" in the menu bar at the top of the screen and then click on "Select Cases" in the drop-down menu.  Select "if condition is satisfied" by clicking on its circle and then click on the "if" button below.  Enter the specification for the cases you want to select.  Your specification should read "R5 < 20000".  Don't enter the quotation marks.  Now click on "Continue" and then on "OK".  To make sure you did this correctly, run a frequency distribution for R5.  This time you should only see the four categories for Evangelical Protestants, Mainline Protestants, Historically Black Protestants, and Roman Catholics.  If you made a mistake, you'll need to do it again.  You should also notice that other religious groups in the data window have been lined out. 

Now run a three-variable table with biblical literalism (RBL7R2) as your independent variable, opinion on environmental laws and regulations (ENV1) as your dependent variable, and religious preference (R5) as your control variable.  Since you have selected out only those in the first four preference categories, you should only see these respondents in your table.

Compare the percent differences for each age category.  Look to see they changed drastically for any of the age categories.  You should find one of the age categories for which the percent difference decreased sharply.  This is called specification because you have specified the conditions under which the relationship is either much stronger or much weaker. 

Part VI—Conclusions

Summarize what you learned in this exercise.  Was the relationship between biblical literalism and how people felt about environmental laws and regulations spurious when you controlled for age?  What happened when you controlled for religious preference?  Why is it important to introduce other variables into the analysis?

 


[1] This assumes that the proper associations have been set up on your computer so the computer knows that .sav files are SPSS data files

[2] SPSS allows you to change the way your output is displayed.  You can change these preferences by clicking on "Edit" in the menu bar at the top of the screen and then clicking on "Options" and finally on the "Output" tab.  Under "Variables in item labels shown as" select "Names and Labels" and then under "Variable values in item labels shown as" select "Values and Labels."  Then click on "OK."  You can also try out other options.

[3] The "R" indicates that it is a recoded or composite variable and the "1" indicates that it is the first recoded or composite variable.  The second recoded or composite variable would be "R2".

[4] The percent difference refers to the difference between the percents for the biblical literalists and the non-literalists.  For example, subtract the percent of literalists who chose the first claim from the percent of non-literalists who chose that claim.  It doesn't matter which percent you subtract from the other percent as long as you are consistent.