Music remains a common part of worship in the United States. Congregational singing appeared in more than 96% of U.S. congregations’ most recent main services in every National Congregations Study wave from 1998 through 2018–19. The same research shows a different pattern for choirs, while instruments, projected lyrics, and online access changed more noticeably over time.
Contents
- How often congregations sing together
- Choir participation and long-term change
- How much worship time is devoted to music
- Differences across services and congregations
- Instruments used in congregational worship
- Projected, streamed, and virtual participation
How often congregations sing together
The National Congregations Study measured whether congregational singing occurred in each congregation’s most recent main service. The congregation-level percentages were consistently high:
- In 1998, congregational singing occurred in 96.8% of U.S. congregations’ most recent main services.
- In 2006–07, the figure was 97.2%.
- In 2012, it was 96.4%.
- In 2018–19, it was 96.3%.
These figures describe congregations, not the share of individual worshipers. The attendee perspective produced similarly high but not identical results. Congregants represented in the study attended a most recent main service with congregational singing in 98.1% of cases in 1998, 97.1% in 2006–07, 98.4% in 2012, and 97.2% in 2018–19. The distinction matters: a congregation-level result gives each congregation a role in the percentage, while the attendee perspective reflects the people represented by those congregations.
| Measurement perspective | 1998 | 2006–07 | 2012 | 2018–19 |
|---|---|---|---|---|
| U.S. congregations with congregational singing | 96.8% | 97.2% | 96.4% | 96.3% |
| Attendee perspective: service included congregational singing | 98.1% | 97.1% | 98.4% | 97.2% |
The 2018–19 study also reported that 70% of non-Christian congregations had congregational singing of some kind in their worship service. That percentage is a separate comparison for non-Christian congregations; it should not be read as a replacement for the overall U.S. congregation figures.
Source: Congregations in 21st Century America, National Congregations Study, 1998, 2006–07, 2012, and 2018–19.
Choir participation and long-term change
Choir singing was less widespread than congregational singing and declined in the congregation-level measure across the study period. Choir singing occurred in 53.9% of U.S. congregations’ most recent main services in 1998. It was reported in 44.1% in 2006–07, 45.3% in 2012, and 41.9% in 2018–19.
The attendee perspective was higher in each listed year, indicating that people represented by the study were more likely to attend congregations with choir singing than the unweighted congregation percentages alone might suggest. The attendee figures were 72.3% in 1998, 58.0% in 2006–07, 57.2% in 2012, and 53.8% in 2018–19.
The National Congregations Study also reported a seasonal-exclusion comparison. When July and August interviews were excluded, choir singing was reported by 54.4% of congregations in NCS Wave I, associated here with 1998, and by 46.2% in NCS Wave IV, associated here with 2018–19. These are not additional years; they are a specified comparison using a different interview-month treatment.
| Choir measure | 1998 | 2006–07 | 2012 | 2018–19 |
|---|---|---|---|---|
| U.S. congregations’ most recent main service | 53.9% | 44.1% | 45.3% | 41.9% |
| Attendee perspective | 72.3% | 58.0% | 57.2% | 53.8% |
The 2012 National Congregations Study report gives several differences by tradition and congregation size. Choir singing was present in about 69% of Catholic congregations’ most recent main services and about 80% of Black Protestant congregations’ most recent main services. Among evangelical congregations with fewer than 100 regularly participating adults, choir presence was 35% in 2012. These figures describe different subgroup definitions, so they should be compared as separate contexts rather than combined into one overall rate.
Source: Congregations in 21st Century America and Religious Congregations in 21st Century America: A Report from the National Congregations Study, 1998–2018–19 and 2012 measures.
How much worship time is devoted to music
The median number of music minutes in a U.S. congregation’s most recent main service was 20 minutes in 1998, 2006–07, 2012, and 2018–19. The attendee perspective also reported a median of 20 music minutes in each of those four periods. Because these are medians, they describe the midpoint of the measured services rather than the experience of every congregation or worshiper.
For 2018–19, the National Congregations Study placed the median U.S. worship service at 75 minutes, including a median 20 minutes of music. The same source reported a median Catholic worship service of about 60 minutes, with 20 minutes of music. A median Black Protestant worship service lasted about 120 minutes and included 30 minutes of music.
| 2018–19 service context | Median service length | Median music time |
|---|---|---|
| U.S. worship service | 75 minutes | 20 minutes |
| Catholic worship service | About 60 minutes | 20 minutes |
| Black Protestant worship service | About 120 minutes | 30 minutes |
The service-length figures show why music minutes need context. Twenty minutes can represent a larger share of a 60-minute service than of a 120-minute service, but the supplied measurements report the two quantities separately and do not provide a calculated percentage. The stable 20-minute overall median from 1998 through 2018–19 indicates consistency in the central U.S. measure even as other parts of worship music changed.
Source: Congregations in 21st Century America, 1998, 2006–07, 2012, and 2018–19.
Differences across services and congregations
Music can distinguish services within the same congregation. Among U.S. congregations with multiple services, 46.3% reported in 2012 that kind of music was an important difference between services. In 2018–19, the corresponding congregation-level figure was 54.9%.
The attendee perspective produced 71.9% in 2012 and 68.9% in 2018–19 for multiple-service congregations reporting music as an important service difference. These percentages use the attendee perspective and therefore are not directly interchangeable with the congregation-level results.
The 2012 report also supplied a different set of measures about differences between services. The share of congregations with multiple services reporting important differences between those services was 48% in 2006 and 30% in 2012. Separately, 15% of congregations with more than one service said that music differed between services in 2012. The wording and measures differ, so these values should remain labeled rather than treated as one time series.
Multisite congregations represented 3.4% of U.S. congregations in 2012, while 10% of U.S. churchgoers were in multisite congregations. Those figures describe organizational reach and churchgoer representation, not music participation directly. They provide context for why a service-level music comparison may occur within a broader multisite structure.
Source: Congregations in 21st Century America and Religious Congregations in 21st Century America: A Report from the National Congregations Study, 2006, 2012, and 2018–19.
Instruments used in congregational worship
Instrument use changed in different directions across U.S. congregations’ most recent main services. Drums were used in 19.9% of services in 1998, 32.5% in 2006–07, 34.3% in 2012, and 40.8% in 2018–19. The supplied figures therefore show a higher reported drum-use share in the later period.
Organ use was 53.0% in 1998, 42.0% in 2012, and 46.8% in 2018–19. Guitar use was 33.5% in 2006–07, 29.3% in 2012, and 35.3% in 2018–19. The available observations do not cover every instrument in every wave, so missing years should not be treated as zero use.
| Instrument | 1998 | 2006–07 | 2012 | 2018–19 |
|---|---|---|---|---|
| Drums | 19.9% | 32.5% | 34.3% | 40.8% |
| Organ | 53.0% | Not supplied | 42.0% | 46.8% |
| Guitar | Not supplied | 33.5% | 29.3% | 35.3% |
The instrument percentages refer to use in the most recent main service, not to the number of musicians, the number of instruments, or the share of worshipers playing them. They also do not establish whether an instrument was used throughout the service or only at one point.
Source: Congregations in 21st Century America, 1998, 2006–07, 2012, and 2018–19.
Projected, streamed, and virtual participation
Projected song lyrics appeared in 31.5% of U.S. congregations’ most recent main services in 2012 and 42.1% in 2018–19. From the attendee perspective, projected lyrics appeared in 42.1% of most recent main services in 2012 and 49.0% in 2018–19. As with singing and choir measures, the two perspectives answer different questions about congregations and the people represented by them.
In 2018–19, half of U.S. congregations recorded parts of their worship service for later listening or viewing, and 20% broadcast or streamed worship live. Among congregations that used smartphones during worship, 5% used them so people could follow service music lyrics on an app. That 5% is conditional on smartphone-using congregations and is not a percentage of all U.S. congregations.
Pew Research Center reported a participant-level measure for November 2022. Thirty-nine percent of U.S. adults who regularly watched religious services online or on television said they usually prayed, responded aloud, sang, or knelt all or most of the time while watching. This is a measure of reported behavior among regular online or television viewers, not a count of people attending an in-person congregation.
Together, the measures distinguish several forms of participation: singing in a shared service, following projected lyrics, watching a recorded service, joining a live broadcast, and responding physically or vocally while viewing remotely. They were measured in different years and populations, so they should be read as separate indicators rather than added together.
Sources: Congregations in 21st Century America, 2012 and 2018–19; Americans’ experiences with virtual religious services, Pew Research Center, November 2022.