On this website you can learn about the NCS, review the survey methodology and questionnaires, read about the research findings, and contact the research team. You also can find out basic results from almost any question asked on the survey. You can examine responses to one question, or you can pick two survey questions and create your own cross-tabulations of responses to those questions. These searches can provide information that reflects the average congregation in the U.S. or the experience of the average participant in U.S. congregations. Explore the site and the NCS survey information and let us know if you have questions.

The results from this study allow religious leaders across the country to compare their own congregations to a national sample of other congregations.

> They will be able to see what other congregations like theirs are doing when it comes to worship, religious education, social ministries, and other activities.

> They will be able to learn how their congregations compare to others on characteristics like size and staffing.

> They will be able to understand better which of the issues and challenges they face also are faced by many other congregations.

> And they will be able to see which changes experienced by their own congregations are broad changes within and across religious groups and regions.

Religious leaders and people in congregations often find that knowing more about the ways in which their own congregation compares with others enables them to understand better their own experiences, situations, and practices. The context provided by the National Congregations Study helps congregations to better understand and assess where they are and how they are doing as a congregation.

You can select almost any question asked in the survey and get the number and percentage of answers for each response category for that question. This is called the “Frequency Distribution.” This will also give you the total number of responses and the number of congregations who did not answer it. Those who did not answer are missing cases for that question. You can get these findings to represent the number of congregations or the number of attendees. For example, you could find out the percent of congregations that livestream their worship services, or you could find the percentage of attendees in congregations that livestream their worship services.

Explore the data to create frequency tables of results

You can select two survey questions (also called “variables”) and cross-tabulate answers to them. By choosing a variable to display along the left side of the table and another to display across the top of the table, you will be able to see how the two variables relate. This will produce a “cross-tabulation” table. You can get these findings to represent the number of congregations or the number of attendees. For instance, by cross-tabulating applause with denominational tradition, you could see the proportion of congregations in different traditions that have applause in their services. From a different perspective, you could also find out the number of people (attendees) in different traditions who experience applause in their congregations’ services.

Explore the data to create a cross-tabulation table and compare two variables

Do you want to know about the characteristics of the average congregation or the characteristics of the congregation attended by the average worship service attendee? Do you want your findings to reflect the number of congregations of a certain type or the number of persons in congregations of a certain type? These questions point to subtly different perspectives on the same information. In 2018-19, for example, the average congregation had 70 regular participants; the average attendee, however, was in a congregation with 360 regular participants. In 2018, to give another example, 48 percent of congregations were able to accept donations electronically, but 73 percent of attendees were in congregations that could accept donations electronically. The difference, of course, is that large congregations are more likely than small congregations to have this capacity.

If you want to know about the characteristics of the average congregation or if you want your results to reflect the percentage of congregations of a certain type, then use the “Tables Reflecting the Number of Congregations.” If you want to know about the characteristics of the congregation attended by the average worshipper or if you want your results to reflect the percentage of people in congregations of a certain type, then use the “Tables Reflecting the Number of Attendees.”

For instance, in exploring the facts about the gender of the senior leader, you may want to see what percent of congregations are led by males and by females. To get this, you would use the “Tables Reflecting the Number of Congregations.” And you would see that 86.5% of congregations have male leadership while 13.5% have female senior leadership, according to the 2018-19 survey.

On the other hand, you might be interested in the percent of people who attend U.S. congregations that are led by male or female senior clergy. Then you would use the “Tables Reflecting the Number of Attendees.” This would show that in 2018-19, 92.6% of U.S. worshippers were in congregations led by a male and 7.4% were in congregations with a female senior leader.

The same two perspectives are presented when cross-tabulating the responses to two questions. For instance, you may ask what percentages of male- and female-led congregations have a particular characteristic, such as having handed out voter guides. To get this, you would look at the “Tables Reflecting the Number of Congregations.”

Or you might want to look at the “Tables Reflecting the Number of Attendees” if you were more interested in the percentage of people attending female-led congregations who experience guitars in worship compared to the percentage of people in male-led congregations who hear guitars in worship.

Using the online tool, you can create tables that summarize information from one question (or variable) or tables that compare responses from two variables. Exact wording of questions and notes about the variables in any tables that you create using the online tool can be found below the tables. This information and more is also available in the NCS cumulative codebook, which also includes questionnaires for all waves of the survey. Definitions of key terms can be found in the glossary below.

Frequency Tables that Reflect the Number of Congregations provide basic information about the percentage of U.S. congregations that have any of the characteristics about which we have information. Frequency distributions list all of the ways a congregation was classified with respect to a given topic, and then add up all of the congregations that fall into each category.

Here’s an example ⇢

In this example, you can see what percentage of U.S. congregations have each of three types of theological orientation: More Conservative, Right in the Middle, and More Liberal. The key numbers are in the “valid percent” column. This table tells us, for example, that 33.9% of congregations (in 2018-19) were described as theologically “right in the middle.”

Frequency Tables that Reflect the Number of Attendees provide basic information about the percentage of U.S. worshippers that attend congregations with any of the characteristics about which we collected information.

Here’s an example of this type of table ⇢

In this example, you can see what percentage of people attending U.S. congregations are in congregations that have each of three types of theological orientation: More Conservative, Right in the Middle, and More Liberal. This table tells us that 35.2% of people who attend worship services (in 2018-19) were in congregations that were described as theologically “right in the middle.”

Cross-Tabulation Tables that Reflect the Number of Congregations put congregations into the categories of the two variables you have selected. These cross-tabulations are set up so that you can see the number (or percent) of congregations in each left-hand-side category that have the characteristics identified along the top of the table. They should be read “across” each line of the table.

Here’s an example ⇢

In this example, you can see how the length of a congregation’s service breaks down for each different type of theological orientation. For example, the first line of this table says that, in 2018-19, 13.6% of theologically conservative congregations had services lasting from 91 to 120 minutes, and 7.1% had services exceeding 2 hours.

You also can use these tables to examine differences between different types of congregations. This can be done by comparing a percentage from one line with the analogous percentage from a different line. In this example, 69.7 percent of the theologically conservative congregations have services longer than an hour (49% with 61–90 min, plus 13.6% with 91-120 minutes, plus 7.1% with more than 2 hours), while only 58 percent of theologically more liberal congregations have services longer than an hour. Below the table is a note indicating that the association between these two variables is statistically significant at the .01 level. That means it is correct to say with 99% confidence that a congregation’s theological orientation is related to the length of a typical service in that congregation.

Cross-Tabulation Tables that Reflect the Number of Attendees put congregations into the categories of the two variables you have selected. These cross-tabulations are set up so that you can see, of all people attending congregations in each category listed on the left of the table, what percentage are in congregations that also have the characteristics identified across the top of the table. Interpret these tables by reading “across” each line of the table.

Here’s an example of this type of table ⇢

In this example, you can see how the length of a congregation’s service breaks down for each different type of theological orientation. For example, the first line of this table says that, of all worshippers attending theologically conservative congregations in 2018-19, 10.8% attended congregations with services lasting from 91 to 120 minutes, and 5.7% attended congregations with services that exceeded 2 hours.

You also can use these tables to examine differences between different types of congregations. This can be done by comparing a percentage from one line with the analogous percentage from a different line. In this example, 63.6% (47.1% + 10.8% + 5.7%) of worshippers attending theologically conservative congregations attend congregations that have services longer than an hour, while only 53.3% of worshippers attending theologically more liberal congregations have services lasting that long. The association between these two variables is statistically significant at the .05 level, so it is correct to say with 95% confidence that a congregation’s theological orientation is related to the length of a typical service in that congregation.

Column Totals
Figures in this row (in a crosstab table) show the number and percentage of congregations in each column of the table, ignoring differences between the rows. There are two numbers in each column. The first number represents the weighted percentage of congregations in each column. The second number represents the weighted number of congregations in each column. Because of the sampling procedure employed in this study, the number of congregations is weighted, and it is difficult to interpret. You can find more information about the sampling technique and weighting used in this study in the codebook.

Cumulative Percent
The cumulative percent (in a frequency table) reflects the weighted percentage of congregations that fall into a response category plus the weighted percentage of congregations that fall into all of the categories above that category.

Frequency
Figures in this column (in a frequency table) tell you the weighted number of congregations in the sample that fall into each possible response category. Because of the sampling procedure employed in this study, these numbers have been weighted, and are therefore difficult to interpret. It is best to ignore this column in favor of the information presented in the Valid Percent column.

Missing Cases
Sometimes representatives from congregations were unable or unwilling to provide answers to questions. These instances represent blank spaces in the data, and are therefore called missing cases. This number (in a frequency table) tells you how many congregations have missing data on the topic that you have selected.

Percent
These numbers (in a frequency table) tell you the percentage of congregations that fall into each possible response category, as well as the percentage of congregations sampled that have missing information on this topic. The information in the Valid Percent column is preferable to this information because the missing cases are excluded from the calculation of the valid percentages.

Response
Labels in this column (in a frequency table) describe each of the possible response categories into which congregations can fall regarding the topic you’ve selected.

Row Totals
Figures in this column (in a crosstab table) show the number and percentage of congregations in each row of the table, ignoring differences between the columns. There are two numbers in each row. The first number represents the weighted percentage of congregations in each row. The second number represents the weighted number of congregations in each row. Because of the sampling procedure employed in this study, the number of congregations is weighted, and it is difficult to interpret. You can find more information about the sampling technique and weighting used in this study in the codebook.

Statistical Significance
This information (following a crosstab table) tells you if the two variables in your table are related. By simply looking at the percentages in the table you can get a sense if the two variables you are examining are related to each other. In the example described in the “how to read the results” section, you might notice from examining the percentages in the table that having a conservative theological orientation appears to be related to having longer services. However, since the numbers presented in the table are based on information from a random sample of 1,210 congregations in the United States in 2018 and 2019 — not from a census of all congregations — it is important to assess how confident we can be that the relationship observed in this sample of 1,210 congregations also is present among all congregations in the United States.

The test for statistical significance allows you to determine if the two variables are related to each other. The note at the bottom of each table indicates whether or not the association between two variables is statistically significant in the population of all U.S. congregations. If the association is significant at the .05 (.01) level, you may conclude with 95% (99%) confidence that the two variables in the table you are examining really are related to each other. If the association is not significant, there is better than a 5% (1%) chance that any differences in the table occur by chance and do not represent a real association between these two variables. In the last example described in the “how to read the results” section, the significance value is less than .05, so it is correct to say with 95% confidence that a congregation’s theological orientation is related to the length of a typical service in that congregation.

It is important to note that this “significance” test tells you only that the two congregational characteristics displayed in the table are related. It does not tell you anything about the nature of that relationship, nor does it tell you anything about how strongly the characteristics are related.

Valid Percent
These numbers (in a frequency table) tell you the weighted percentage of congregations that fall into each response category, ignoring the congregations with missing information on this question. These figures are probably the most informative in the table, because the missing values are excluded from the calculation of the percentages.