By Paul A. Djupe, Claire C. Gilliland, and Shayla F. Olson

A contributing reason why religion retained a significant portion of its adherents during the pandemic was because so many houses of worship pivoted so quickly to online worship. By April 2020, Lifeway Research estimated that 92% of congregations offered an online option as congregations closed to in-person worship; in March 2021, approximately 80% of religious attendees reported their congregations still offered an online option. Those figures represent an enormous investment in new technology and practices given that according to the 2018-2019 National Congregations Study data, collected pre-pandemic, only about 20% of congregations streamed their services.

In our new article published in Review of Religious Research, we explore the resource, religious, and political biases involved in being online. We care about this shift for a few reasons. For one, streaming was almost the only way people received information from congregations during the pandemic. If the congregations able to go online were distinctive in some ways, then that pattern can affect religious presence. But we also care about these patterns for academic reasons. A growing number of researchers want to study clergy communication patterns and there’s no better way than recording them. Fortunately, many clergy do this themselves and post the results online on a congregation website or in an archive like Sermon Central. There are lingering questions about how representative those congregations who post sermons online are of the entire population of congregations. Though streaming services is not quite the same thing as posting a sermon to an archive, we at least can start to address this question and offer qualifying notes to aspiring researchers.

So, what do the patterns look like? We used a couple of different datasets to look for bias – the National Congregations Study (2018-2019) and a sample of individual citizens collected in October 2020. The NCS was collected before the pandemic, obviously, but offers a similar set of results to what we find during the pandemic. Given their resources, larger congregations were more likely to offer service streaming. Catholics and liberal Protestants were less likely to offer streaming compared to non-denominational Christians, while evangelicals and Black Protestants were more likely to stream.

When we shift attention to the pandemic period, we expect that the need to reach people helped to eliminate many of the gaps seen pre-pandemic. We used two questions – one asked about the availability of online worship six months ago (so ~April) and currently (end of October, 2020). Seen in the figure below, the blue dots show gaps created by each variable/category in April and the orange ones show gaps in October. Notably, there are fewer gaps in October (the gaps are insignificantly different from zero if the lines touch the vertical line at zero). For instance, congregations with 300+ members were more likely to offer online worship in April, but it was only congregations with 1000+ that were more likely to offer online worship in October. In another example, strong partisans were more likely to report online worship in April, a gap which disappeared by October.

Gaps in Offering Online Worship Tended to Shrink by October 2020 Compared to April 2020.

Source: October 2020 Survey of 1,790 adult Americans; only worship attenders included.

One of our key questions is to see if online congregations are more or less likely to feature political communication. This could go either way – the conventional wisdom is probably that politics is divisive and would turn off those tuning in for the first time. However, that may be changing and political engagement may be a draw for some. In this case, we are using a question that asks respondents how many different issues from a list of 15 they heard their clergy engage in the past year.

The figure below shows that being online is generally associated with more clergy political engagement but that picture varies considerably by religious tradition. We find evidence that clergy engagement with politics is higher by about one issue engagement for evangelicals and non-denominationals in online congregations and that conclusion doesn’t shift across 2020. This may be a sign of the times, as evangelical has become thoroughly conflated with conservative Republican in at least the last 20 years; the pressure to convey politics may have increased as well. We also find evidence that being online does not matter for clergy political engagement for Catholics, Mainline Protestants, and others. In the remaining traditions, being online appears to make some difference and the pandemic period entailed a shift. For instance, among Black Protestants, there was no difference pre-pandemic, but being online during the pandemic brought with it a greater concentration of political engagement (the same pattern can be found among religious attenders who identify as nones). And for evangelicals and non-denominationals being online was linked to more political engagement in both periods and the pandemic brought no change in degree.

The Bias of Clergy Political Engagement by Online Worship in the Pandemic and Pre-Pandemic by Religious Tradition

Source: October 2020 Survey.

Note: These estimates come from a model that also controls for age, education, race, gender, partisan strength, worship attendance, congregation size, and church activities involved with. Only those who attend several times a year or more are included; data are weighted.

Our analyses find added support for the link between congregational resources and offering online worship services. At the congregational and individual levels, congregation size is associated with a greater likelihood of being online. Given the vast increase in online congregation shopping during the pandemic, this resource bias is only likely to further fuel the collapse of small congregations. We also found evidence for a link between streaming services and greater clergy political speech, especially in certain religious traditions. Researchers need to be aware of these biases that could grow again as the pandemic pressure has eased, though it seems many congregations have kept their online capabilities.

At this point, we cannot say that the higher political engagement levels of online congregations have pervaded the public consciousness through online worship access. But it is tantalizing to think about the possible links when the public is associating religion with politics and especially evangelicalism with Republicanism. If online access remains high, will it further this association and link ever tighter political affiliation with religious involvement?

Professor Paul A. Djupe directs the Data for Political Research program at Denison University, is an affiliated scholar with PRRI, the series editor of Religious Engagement in Democratic Politics (Temple), and co-creator of religioninpublic.blog. Further information about his work can be found on his website and on Twitter and Threads.

Dr. Claire Chipman Gilliland is an Assistant Professor at Furman University.

Shayla F. Olson is a Ph.D. Candidate of Political Science and Scientific Computing at the University of Michigan. Her research focuses on the intersection of religion and race in American politics.