Religion_1SR – Exploring the Relationship between Religion and Attitudes toward Same-Sex Marriage

Note to the Instructor: This is the first in a series of two exercises that focus on the relationship between religion and attitudes about same-sex marriage. 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-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 information.

Goal of Exercise

All research starts with a question.  The question we're going to explore in this exercise is why people have different opinions about same-sex marriage. Some people favor same-sex marriage while others oppose it.  We'll use two-variable crosstabulations, percentages, Chi Square, and measures of association as our statistical tools.  In the next exercise (Religion_2SR) we'll expand our analysis and use three-variable tables.

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 – Same-Sex Marriage

The Pew survey asked respondents "do you strongly favor, favor, oppose, or strongly oppose allowing gays and lesbians to marry legally?"  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 SS1 which is the name of the variable.  The variable name starts with the letters SS which tells you that this variable describes how people feel about same-sex marriage.  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 very 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."  You screen should look like Figure 1.

Title: Figure 1 - Description: This is the SPSS dialog box for Frequencies.

Figure 1

Notice that the list of all variables is in the pane on the left.  I scrolled down to the variable that starts with SS.  Select SS1 by clicking on it and then click on the arrow pointing to the right.  This will move SS1 into the "Variable(s)" box.  Your screen should look like Figure 2.

Title: Figure 2 - Description: This is the SPSS dialog box for Frequencies with SS1 selected.

Figure 2

Now all you have to do is click on "OK" to get your frequency distribution.  Your screen should look like Figure 3.[2]

Title: Figure 3 - Description: This is the SPSS output showing the frequency distribution for SS1.

Figure 3

Take a few minutes to familiarize yourself with the information in the table.

  • The first column is the value and the value label.  The value "1" refers to all people who said they strongly favored allowing gays and lesbians to marry legally.
  • The second column is the number of respondents who said they were strongly in favor of same-sex marriage (8,584).
  • The third column converts the frequencies to percents.  Notice that some respondents said they didn't know or refused to answer this question.  This is called missing data because we don't know how they feel about same-sex marriage. These respondents are given a missing code which for this variable is the value "9".  The percent column converts the frequency to a percent by dividing the frequency (8,584) by the total number of cases including those with missing values (35,071).  Carry out the computation for yourself and convince yourself that it is 24.5%.
  • The fourth column converts the frequencies to valid percents by dividing the frequency (8,584) by the number of cases with valid information (32,405).  In other words, it excludes the cases with missing information (2,666) from the denominator when computing the percent.  Carry out the computation for yourself and convince yourself that it is 26.5%.  This is called the valid percent.  The more missing information there is in the distribution, the greater could be the difference between the percent and the valid percent.  Normally we want to use the valid percents when describing the frequency distribution.
  • The fifth column is the cumulative percent.  The second category (2) is for those who said they favored same-sex marriage.  Suppose we wanted to know what percent either strongly favored or just favored same sex-marriage.  The cumulative percent for the second category is 57.7%.  In other words, 57.7% of the cases with valid information selected either categories 1 or 2.  You can see where this comes from if you add up the valid percents for the first two categories.

Now it's your turn.  The third category in the distribution is for those that opposed same-sex marriage.

  • What is the value for this category?
  • How many respondents said they opposed same-sex marriage?
  • What is the percent for this category?  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?
  • What is the cumulative percent for this category?  What does this mean?

Part III – Religious Preference and Attitudes towards Same-Sex Marriage

Now let's turn to the question of why some people favor and others oppose same-sex marriage.  We're going to focus on various dimensions of religion in our exploration of this question.

Our starting point will be religious preference.  The Pew survey asked "What is your present religion, if any?  Are you Protestant, Roman Catholic, Mormon, Orthodox such as Greek or Russian Orthodox, Jewish, Muslim, Buddhist, Hindu, atheist, agnostic, something else, or nothing in particular?"  This is variable R1 in the data set.

One problem with this variable is that over 15,000 respondents said they were Protestant.  We know there are many different types of Protestants so we might want to break Protestants down more finely.  To do this the Pew survey asked another question – "As far as your present religion, what denomination or church, if any, do you identify with most closely?  Just stop me when I get to the right one. Are you Baptist, Methodist, Lutheran, Presbyterian, Pentecostal, Episcopalian or Anglican, Church of Christ or Disciples of Christ, Congregational or United Church of Christ, Holiness, Reformed, Church of God, nondenominational or independent church, something else, or none in particular?" This is variable R2. 

One of the problems with R2 is that there are so many categories.  The Pew survey dealt with this problem by classifying Protestants into the following three religious traditions.  This is variable R5.

  • Evangelical Protestant tradition
  • Mainline Protestant tradition
  • Historically Black Protestant tradition
  • 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."

Now it's your turn again.  Run a frequency distribution for R5 and use the information in this table to answer the following questions.  Be sure to use the valid percents. 

  • What are the five largest religious groups in R5?  Note that this table includes the religiously unaffiliated as a group.  What are the percents for each of these groups?
  • What percent of adults are Christian?  Non-Christian?  For this question be sure to also include Orthodox Christians, Jehovah's Witness, Mormon, and Other Christian as Christian when you compute the percent of adults who are Christian.
  • Which non-Christian group is the largest?  What is the percent for that group?

Now that we have figured out how to measure religious preference, let's see if religious preference helps us explain why some people favor and others oppose same-sex marriage.  This represents a shift from what we typically call univariate (i.e., one-variable) analysis to bivariate (i.e., two-variable) analysis.  Frequency distributions look at variables one at a time.  Now we are going to look at the relationship between two variables – religious preference which is R5 and how people feel about same sex marriage which is SS1.  Crosstabulation looks at variables two at a time.

The statistical tools that we're going to use to explore the relationship between R5 and SS1 are crosstabulation, Chi Square, and measures of association.  We're not going to go into a lot of detail about these tools.  Your instructor will provide that information.   We will talk briefly about how to get SPSS to compute them and how to interpret them. 

Before we look at the relationship between variables, we need to talk about independent and dependent variables.  The dependent variable is whatever you are trying to explain.  In our case, that would be how people feel about same-sex marriage (SS1).  The independent variable is some variable that you think might help explain why some people favor and others oppose same-sex marriage.  In our case, that would be religious preference (R5). 

To run a crosstabulation in SPSS click on "Analyze" in the menu bar at the top of the screen.  Now click on "Descriptive Statistics" in the drop-down menu and then on "Crosstabs."  (See Chapter 5, Cross Tabulations, in the online SPSS book cited on page 1 of this exercise.)   Your screen should look like Figure 4. 

Title: Figure 4 - Description: This is the SPSS dialog box for Crosstabs.

Figure 4

You're going to put your two variables (i.e., R5 and SS1) 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 would 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.  But since there are so many categories in our independent variable, that's going to create a table that has so many columns that it's difficult to read and copy into your report.  So in this case, we're going to put R5 in the row box and SS1 in the column box.  We're also going to click on the "Cells" box and check the box for the "Row" percents.  If your independent variable is in the rows, then you want to use the row percents.  If it is in the columns, then you want to use the column percents.  Your screens should look like Figure 5 and 6.

 

Title: Figure 5 - Description: This is the Crosstabs dialog box for Crosstabs with R5 in the rows and SS1 in the columns.

Figure 5

Title: Figure 6 - Description: This is the Crosstabs: Cell Display dialog box with Row percents selected.  The Observed counts is selected by default.

Figure 6

 

To get the table, click on "Continue" and then on "OK."  Your screen should look like Figure 7.  This is a big table so I'm only showing the top part of the output.

Title: Figure 7 - Description: This is the output showing the crosstab of SS1 and R5.

Figure 7

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 row percent.  Notice that the row percents add across by row to 100.  Since the percent sum across to 100, you want to compare the percents down.  Always compare the percents in the direction opposite to the way they sum to 100.  This part of the table shows you that 26% of Mainline Protestants and 22% of Roman Catholics strongly favor same-sex marriage but only 10% of Evangelical Protestants and 16% of Historically Black Protestants strongly favor it.  If we add the percent that strongly favor and favor same-sex marriage, we see that 62% of Mainline Protestants and 62% of Roman Catholics favor (i.e., either strongly favor or favor) same-sex marriage but only 30% of Evangelical Protestants and 43% of Historically Black Protestants favor it.  That's quite a difference.

Write a paragraph that summarizes the relationship between religious preference and attitudes toward same-sex marriage in the full table.  Be sure to answer the following questions.

  • Which religions are most likely to favor same-sex marriage?
  • Which religions are least likely to favor same-sex marriage?
  • Write two sentences that summarize this pattern.  The first sentence should describe the pattern in words without using the percents.  The second sentence should use the percents to illustrate the pattern.  Don't just read back each percent.  Rather summarize the pattern using the percents to illustrate that pattern.

Earlier we said we were also going to use Chi Square and a measure of association in our exploration of this relationship.  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 such a relationship. 

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.  Cramer's V is a measure that you can use when one or both of your variables are nominal variables.  A nominal variable is one in which the categories have no inherent order.  R5 is a nominal variable since the different religious groups could be listed in any order.

To get Chi Square and Cramer's V click on the "Statistics" button and then click on the boxes for both Chi Square and Cramer's V.  You screen should look like Figure 8.

Title: Figure 8 - Description: This the Statistics dialog box with Chi Square and Cramer's V selected.

Figure 8

Now the question is how to interpret Chi Square and Cramer's V.  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 Cramer's V look at the value which should be .269 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. 

Part IV – Born-Again Christians and Attitudes toward Same-Sex Marriage

You probably have heard the term "Evangelical Christians" and perhaps have wondered what it meant.  In Part 1 of this exercise I asked you read an article by John Green on the difference between the Evangelical and the Mainline Protestant traditions.  In this article Green lists four defining beliefs of an Evangelical Christian.

  • The belief that the Bible is inerrant (i.e., without error).
  • The belief that the "only way to salvation is through belief in Jesus Christ."
  • The belief that one must have had a "born-again experience."
  • The belief in proselytization or spreading the word.

The Pew survey includes a question on the third belief.  It asked, "would you describe yourself as a 'born-again' or evangelical Christian, or not?"  This is variable R3.  Clearly this question only makes sense for respondents who view themselves as Christians.  Consequently, Pew only asked this question of those who said they were Christian.  Run a frequency distribution for R3 and write a sentence or two describing what the distribution tells you.  Note that there are 10,294 cases listed as system missing.  Those are the non-Christians that were not asked this question.

Now we're ready to look at the crosstabulation of R3 and SS1 to see if born-again or Evangelical Christians are more or less likely to favor or oppose same-sex marriage.  This time put the dependent variable (SS1) in the row and the independent variable (R3) in the column.  This is the more traditional of setting up a crosstabulation.  This means you will want to get the columns percents this time.  Remember the rule -- if your independent variable is in the rows, then you want to use the row percents; if it is in the columns, then you want to use the column percents.  Tell SPSS to compute Chi Square.  For our measure of association, this time use Kendall's tau-b.  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.

Write a paragraph describing the relationship between having had a born-again experience and attitudes toward same-sex marriage.  Use the column percents in your answer.  Remember that since the column percents sum down to 100, you will compare the percents straight across.  Use Chi Square and Kendall's tau-b to help you interpret the table.  Like Cramer's V, tau-b varies from 0 (no relationship) to 1 (strongest possible relationship).  Unlike V, tau-b can be either positive or negative.  However, for this exercise ignore the sign when you interpret tau-b.

Part V – Biblical Literalism and Attitudes toward Same-Sex Marriage

Another way we can classify individuals is the way they view the sacred texts of their religion.  Do they see these texts as the literal word 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 and see 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.  Look back at the article by John Green in the PBS Frontline article on "Evangelicals v. Mainline Protestants."  Notice that this is one of the beliefs that Green says is central to Evangelical Protestants.  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.

Now we're ready to see whether biblical literalism is related to how respondents feel about same-sex marriage.  Let's limit our analysis to Christians.  Click on "Data" in the menu bar at the top of the SPSS screen. This will be the second row at the top of the screen.  Now click on "Select Cases" in the drop-down menu.  Your screen should look like Figure 9.

Title: Figure 9 - Description: This is the SPSS dialog box for selecting cases.

Figure 9

Select "If condition is specified" in the option on the right by clicking on the circle.  (See Chapter 3, Transforming Data, in the online SPSS book cited on page 1 of this exercise.)   Now click on the blue button just below this option.  We want to select those cases for which the variable R5 is less than 50000.  If you run a frequency distribution for R5 you'll see that Christians have codes less than 50000.  Copy the following statement and paste it into the box to the right of the arrow.

            R5 < 50000

Your screen should look like Figure 10. 

Title: Figure 10 - Description: This is the SPSS dialog box for selecting only the Christians (e.g., values less than 50000 for R5).

Figure 10

If you rerun the frequency distribution for R5 you should see only the Christians in the output.

Now run the crosstab for RBL7R2 and SS1 being sure to get the correct percents, Chi Square, and Kendall's tau-b.  Think carefully about which variable should be your independent and dependent variables and put the independent variable in the columns.  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.

Part VI – Using a Typology of Christians to Further Explore Attitudes toward Same-Sex Marriage

In the previous two parts of this exercise we looked at two defining beliefs of Evangelical Christians – the belief that one must have had a born-again experience (R3) and the belief that the Bible is the literal word of God (REL7R2).  Now let's combine these two beliefs.  Run a frequency distribution for RBL21.  This variable is a typology of Christian beliefs combining R3 and RBL7R2.  Your analysis will automatically be limited to Christians since non-Christians are defined as missing data. 

Run the crosstab for RBL21 and SS1 being sure to get the correct percents, Chi Square, and Kendall's tau-b.  Think carefully about which variable should be your independent and dependent variables and put the independent variable in the columns.  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.

Part VII – Religiosity and Attitudes toward Same-Sex Marriage

Still another dimension of religion is religiosity which refers to the strength of a person's attachment to their religious preference.  This describes how religious a person is.  There are three commonly used measures of religiosity – how often a person attends religious services, how important they say religion is to them, and how often they pray.

The Pew survey asked, "Aside from weddings and funerals, how often do you attend religious services … more than once a week, once a week, once or twice a month, a few times a year, seldom, or never?"  This is REL1.

They also asked, "How important is religion in your life – very important, somewhat important, not too important, or not at all important?"  This is REL2.

Finally, Pew asked "People practice their religion in different ways.  Outside of attending religious services, do you pray – several times a day, once a day, a few times a week, once a week, seldom, or never?"  This is REL3.

Run three crosstabs to show the relationship between each of these variables and how people feel about same-sex marriage.  Don't limit your analysis to Christians for part 7.  To make sure you are using the full data set, click on "Data" and then on "Select Cases."  Select the "All cases" option and then click on "OK." 

Think carefully about which variable should be your independent and dependent variables and put the independent variable in the columns.  For each crosstab, 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.

Now reread your answers and write another paragraph comparing the relationships you just described.  Did you find the same relationship for all three measures of religiosity or were they different?  What does this tell you about the relationship between religiosity and how respondents feel about same-sex marriage?

Part VIII – Conclusions

Write one or two paragraphs summarizing what you learned about religion and attitudes toward same-sex marriage.  Be sure to consider what you discovered in each of the first seven parts of this exercise.

 

 


 

[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".