Looking beyond the COVID-19 Pandemic: Congregants’ Expectations of Future Online Religious Service Attendance

Abstract

Many religious congregations in the United States have adapted to COVID-19 lockdowns by offering religious services online. This study aims to understand whether congregants from a diverse set of faith traditions expect to attend online or in-person religious services after the pandemic. First, it examines how members of different religious traditions vary in their expectations of future attendance. Second, it explores whether respondents’ habituation to online attendance during the pandemic might result in greater preference for future online attendance. This study draws on a non-representative sample of 1609 members of Christian, Jewish, and Hindu communities in four US states surveyed in late 2020 and employs logistic regression models. The findings first suggest a divergence between congregation types that require in-person attendance for certain rituals versus those that do not. Second, habituation of the practice of online attendance may cultivate the desire to sustain this practice into the future. Online religious services have been well received by most congregants, and online services will likely play a useful role across congregation types, albeit at differing levels and with different audiences. Our finding that marginal congregants were more likely to prefer online religious services, while more engaged members preferred in-person attendance, is of relevance to faith leaders.

1. Introduction

Once the novel coronavirus was announced as a global pandemic, many businesses and organizations, including places of worship, were forced by lockdowns to restrict if not close down their operations (Sandford 2020). From a historical perspective, many pandemics and wars have not substantially disrupted religious practices or congregation life, but with the COVID-19 pandemic in 2020, traditional religious practices and festivals were either halted or attended virtually (Norman and Reiss 2020). Congregation closures and restrictions on religious practices even gave rise to complaints regarding human rights violations (Mazurkiewicz 2021), but most congregations in the US were able to deliver at least some liturgies and religious practices virtually (Bryson et al. 2020).

The fear of the virus spreading through communal in-person religious practices led some politicians and media outlets to label religious gatherings in the US as mass spreaders of the coronavirus and to request in-person religious gatherings to be stopped (Lisinski 2020). Worship practices that were previously conducted in person had to be conducted online, and adaptations to lockdown policies that prohibited communal practices became necessary (Oxholm et al. 2021). In such a context, many faith leaders worried that members—especially those who were not highly committed—would lose the habit of religious service attendance, lose the connection to their faith community, and drop out of religious participation altogether. Reduced religious service attendance in turn would translate into reduced giving, thus posing serious concerns about the financial viability of many faith communities (Garfield 2020).

In some cases, the shift to online modalities, including streaming religious services online and offering online classes, attracted new congregation members (Overall 2020). Since online services drew in new audiences from around the US and even different countries, the pandemic felt like a revival for some communities (Overall 2020). However, not everyone is interested in participating in virtual religious services. A Pew survey conducted in July 2020 (Cooperman 2020) showed that few religious Americans planned to attend online services after the pandemic ends and once regular in-person services resume. This, in turn, might worry religious leaders that new members who joined virtually during the pandemic may not continue to be an active part of the congregation in the future (Ross 2020).

Furthermore, certain religious traditions may have religious rituals that do not lend themselves well to online participation on a long-term basis (e.g., Roman Catholicism and Orthodox Judaism). Yet, as people generally prefer contexts that can sustain the cultural habits that they develop (Swidler 2013), the experience of online religious attendance during the pandemic may become a habitual practice for some, thus making it likely that they would prefer to continue this practice into the future. For these reasons, it is important to examine the following two research questions: (1) To what extent do expectations regarding online vs. in-person religious service attendance after the pandemic differ by congregation type? (2) Does a higher frequency of attending online religious services during the pandemic increase the likelihood of expecting to continue with this attendance modality after the pandemic?

2. Method

2.1. Participants and Data

Participants of this study were based in twelve congregations from different faith traditions (4 Catholic, 3 Jewish (2 Reform and 1 Orthodox), 2 African American Baptist, 1 Latter-day Saints (LDS), 1 Hindu) in urban and suburban areas in two metropolitan regions: Washington DC (including Maryland and Virginia) and San Antonio, Texas (n = 1609). The congregations varied in size from less than 100 to more than 3000 members. Even though there is variation in religiosity among participants, the participants of this study skewed towards more religious members, who tend to be older women. Leaders of the religious communities from which participants were sampled consider this sample as adequately representative of members who are regularly involved in their communities.

Data were collected through online surveys between October and December 2020. Congregation leaders sent an online link of the survey to their community mailing lists and to small groups (i.e., councils, boards, committees, etc.). The generic online link was sent by leaders of the faith communities that agreed to participate, and it was sent via a variety of channels: newsletters, email lists, website, social media platforms, and so on. Weekly reminders were sent for three weeks. Each community received USD 1000 for their participation in the study and a report of the main findings.

This sample of congregations is not meant to be statistically representative of either their denominations or localities, but it is well-suited to inform us of patterns among the more active members of faith communities.

2.2. Variables

Future religious attendance. The main dependent variable of this study, expected future religious service attendance, is based on responses to the question, “Once the COVID-19 pandemic is over, which of the following would describe you best?” Response categories included “most likely attend in-person religious services only”, “most likely attend online religious services only”, “most likely attend a mix of online and in-person religious services”, or “most likely not attend religious services”. As less than 3 percent of the full sample chose “online only” and as less than 2 percent expect to stop attending altogether, we exclude those cases from the analysis and focus only on those who expect to attend either a mix of online and in-person religious services (coded as 1) or in-person religious services only (coded as 0 and treated as the reference group in the logistic regressions) after the pandemic.

Religiosity. We include a frequency scale of in-person religious service attendance before the COVID-19 pandemic (1 = “a few times a year or less”, 2 = “once a month or less”, 3 = “2–3 times a month”, 4 = “once a week”, 5 = “several times a week”, treated as continuous), a frequency scale of online religious service attendance in the last two month (1 = “never”, 2 = “once a month or less”, 3 = “2–3 times a month”, 4 = “once a week”, 5 = “several times a week”, treated as continuous) as well as a scale of length of membership in the congregation (1 = “1 year or less”, 2 = “2–4 years”, 3 = “5–9 years”, 4 = “10–19 years”, and 5 = “20 or more years”, treated as continuous). We also include measures of religious affiliation/congregation type (African American Baptist, Catholic, Evangelical, Modern Orthodox Jewish, Reform Jewish, LDS, and Hindu, with Catholics as the reference group).

Change in attendance. In response to the survey question “Overall, how have the following changed (if at all) since the COVID-19 pandemic started?” respondents indicated whether their frequency of religious service attendance “decreased since the pandemic”, remained “the same as before the pandemic”, or “increased since the pandemic”.

Control variables. We include common demographic characteristics and socioeconomic control variables such as gender, age, race (White, African American, Hispanic, Asian American and Other), marital status (1 = “married or in partnership” vs. 0 = “single/widowed/divorced/never married”), educational level (1 = “college degree or higher” vs. 0 = “less than a college degree”), household income (grouped as 1 = “up to $35,000”, 2 = “$35,001–$50,000”, 3 = “$50,001–$100,000”, 4 = “$100,001–$150,000”, 5 = “more than $150,000”, treated as continuous) as well as self-rated physical health (“In general, how would you rate your physical health?” scored from 0 “poor” to 10 “excellent”, treated as continuous).

2.3. Analytic Strategy

We compile three logistic regressions of congregants’ self-reported expectations of attending either a mix of online and in-person religious services (1) vs. in-person religious services only (0) after the pandemic. In model 1 (M1), the first research question is addressed by using the congregation type and control variables to predict expectations of future attendance of online and in-person religious service attendance modalities. The demographic and socio-economic control variables facilitate comparisons of congregation types, as the make-up of the congregations can differ in terms of age, race, income levels, and so on. Models 2 and 3 (M2 and M3) explore the second research question by focusing on the associations of online religious service attendance during the pandemic with the expectations of future attendance modalities. We control for the length of membership in the congregation and the frequency of in-person religious service attendance before the pandemic to hold these factors constant and to differentiate newer or marginal congregants from more committed or frequent attenders. Model 3 specifically focuses on changes in the frequency of religious service attendance during the pandemic as a predictor of expectations of future attendance modalities.

Listwise deletion is applied to deal with missing information, resulting in a reduced analytical sample of 1116 complete cases. Data validation checks did not indicate systematically differing missing data patterns between congregations. The statistical analyses are performed via Stata/MP 16.1 (StataCorp. 2019. Stata Statistical Software: Release 16. College Station, TX: StataCorp LLC). Based on the odds ratios of the logistic regressions, predicted probabilities are illustrated for selected variables of interest.

3. Results

3.1. Descriptive Statistics

In terms of relevant background information, all studied congregations offered at least some religious services or meetings online, although the actual content and frequency may have varied both between and within congregations over time as the pandemic evolved. Whereas none of the congregations had virtual attendance options before the COVID-19 pandemic, congregations adapted quickly and less than 14 percent of respondents have never attended an online service during the pandemic. As seen in

Table 1. Descriptive statistics of the analytic sample (n = 1116).

	Proportion	Mean	Min.	Max.
Focal variables
Expecting to attend a mix of online and in-person religious services after the pandemic (ref. expecting to attend in-person religious services only)	44%
Congregation type
Catholic	40%
Evangelical	12%
LDS	9%
Reform Jewish	23%
Orthodox Jewish	4%
African American Baptist	8%
Hindu	6%
Length of membership in the congregation		3.9	1.0	5.0
Frequency of in-person religious service attendance pre pandemic		3.5	1.0	5.0
Frequency of online religious services attendance during the pandemic		3.1	1.0	5.0
Change in frequency of religious service attendance during the pandemic
Decreased since the pandemic	43%
Same as before the pandemic	43%
Increased since pandemic	15%
Demographic and socio-economic variables
Gender: Men (ref. Women)	34%
Race/ethnicity
White	68%
Black/African American	10%
Hispanic	12%
Asian American	7%
Other	4%
Age in years		56.6	19.0	96.0
Marital status: Married/partnership (ref. not married)	75%
Education: College degree or higher (ref. less than a college degree)	82%
Household income		2.9	1.0	4.0
Self-rated physical health		7.2	0.0	10.0

Note: n = 1116 (analytic sample). Mental Health in Congregations Study, 2020. Mean = Statistical averages for continuous variables. Proportion = Frequencies expressed as percentages for categorical variables.

, 44% of respondents expected to attend a mix of online and in-person religious services after the pandemic while 56% of respondents expected to attend in-person religious services only.

The descriptive statistics suggest that most sample members had high levels of religious commitment and that most respondents have been long-term members of their respective congregations. The sample could be described as predominantly consisting of older, highly religious women. Most respondents had relatively high incomes and high levels of education. While these demographics are not representative of the US population overall or even the congregations, the faith leaders of these communities (to whom we presented survey reports) indicated that the demographics of respondents were roughly representative of those who were regularly in the pews.

Table 2. Row percentages for the frequency of in-person religious service attendance before the pandemic by congregation type, by the frequency of online religious service attendance during the pandemic and by the change in frequency of religious service attendance from before to during the pandemic (n = 1116).

Frequency of In-Person Religious Service Attendance before the Pandemic
	A Few Times a Year	Once a Month or Less	2–3 Times a Month	Once a Week	Several Times a Week
Congregation type
Catholic	3.4	2.3	8.1	61.9	24.4
Evangelical	3.0	0.0	6.0	56.4	34.6
LDS	1.1	0.0	4.2	65.3	29.5
Reform Jewish	43.4	31.0	14.7	8.9	1.9
Orthodox Jewish	15.0	12.5	20.0	40.0	12.5
African American Baptist	1.2	2.4	14.3	47.6	34.5
Hindu	25.4	11.1	11.1	17.5	34.9
Frequency of online religious service attendance during the pandemic
Never	16.8	5.2	9.7	50.3	18.1
Once a month or less	34.2	15.2	12.5	29.2	9.0
2-3 times a month	12.2	18.2	19.2	37.4	13.1
Once a week	3.6	4.0	5.3	64.0	23.1
Several times a week	2.1	3.2	4.8	39.6	50.3
Change in frequency of religious service attendance during the pandemic
Decreased since the pandemic	9.9	8.8	12.2	45.5	23.7
Same as before the pandemic	14.9	5.7	6.9	49.9	22.6
Increased since the pandemic	22.8	21.6	13.6	28.4	13.6

Note: n = 1116 (analytic sample). Mental Health in Congregations Study, 2020. The table present row percentages for the frequency of in-person religious service attendance before the pandemic.

provides further details on the levels of religiosity by cross-tabulating the frequency of in-person religious service attendance before the COVID-19 pandemic by congregation type and two indicators of religious attendance during the pandemic. We find that 95% of LDS, 91% of Evangelical, 86% of Catholic, 82% of African American Baptist, 53% of Orthodox Jewish, 52% of Hindu and 11% of Reform Jewish congregants had been attending in-person services once a week or more before the pandemic. Considering the self-reported change in the frequency of service attendance from before to during the pandemic, decreases mainly occurred among those who had been attending in-person religious services frequently before the pandemic (e.g., 69% of respondents who reported decreases but only 42% of those with increases had been attending in-person services once a week or more before the pandemic).

3.2. Expected Future Attendance

The three logistic regressions (M1–M3) in

Table 3. Logistic regression results of expectations of attending a mix of online and in-person religious services (vs. in-person religious services only) after the COVID-19 pandemic (n = 1116).

Logistic Regressions of Expectations to Attend a Mix of Online and In-Person Religious Services (1) vs. In-Person Religious Services Only (0) after the Pandemic
	M1	M2	M3
Focal variables
Congregation type, Evangelical (ref. Catholic)	2.83 ***	2.19 **	3.08 ***
	(1.79–4.48)	(1.35–3.57)	(1.93–4.90)
Congregation type, LDS (ref. Catholic)	1.65 +	1.17	1.81 *
	(0.95–2.87)	(0.64–2.14)	(1.02–3.19)
Congregation type, Reform Jewish (ref. Catholic)	10.56 ***	5.86 ***	5.70 ***
	(6.98–15.98)	(3.41–10.08)	(3.40–9.54)
Congregation type, Orthodox Jewish (ref. Catholic)	1.09	1.05	0.84
	(0.47–2.53)	(0.39–2.85)	(0.35–2.03)
Congregation type, African American Baptist (ref. Catholic)	2.88 **	2.17 +	2.89 *
	(1.29–6.44)	(0.91–5.14)	(1.28–6.53)
Congregation type, Hindu (ref. Catholic)	6.17 ***	3.66 *	4.92 ***
	(2.54–14.99)	(1.33–10.05)	(1.98–12.24)
Length of membership in the congregation		0.97	1.00
		(0.84–1.12)	(0.87–1.14)
Frequency of in-person religious service attendance pre pandemic		0.55 ***	0.81 **
		(0.46–0.65)	(0.70–0.93)
Frequency of online religious services attendance during the pandemic		2.34 ***
		(1.99–2.76)
Change in frequency of religious service attendance during the pandemic
Decreased since the pandemic (ref. no change)			0.89
			(0.66–1.20)
Increased since the pandemic (ref. no change)			2.26 ***
			(1.45–3.53)
Demographic and socio-economic variables
Gender: Men (ref. Women)	0.86	0.93	0.87
	(0.64–1.16)	(0.67–1.28)	(0.64–1.18)
Race/ethnicity, Black/African American (ref. White)	3.71 ***	2.71 *	3.51 ***
	(1.80–7.64)	(1.22–5.98)	(1.68–7.34)
Race/ethnicity, Hispanic (ref. White)	2.69 ***	2.05 **	2.43 ***
	(1.66–4.35)	(1.21–3.46)	(1.49–3.96)
Race/ethnicity, Asian American (ref. White)	2.80 *	2.39+	2.69 *
	(1.23–6.39)	(0.93–6.18)	(1.16–6.23)
Race/ethnicity, Other (ref. White)	1.41	1.13	1.32
	(0.66–3.03)	(0.49–2.61)	(0.61–2.85)
Age in years	1.02 ***	1.01	1.02 **
	(1.01–1.03)	(1.00–1.02)	(1.01–1.03)
Marital status, Married/partnership (ref. not married)	0.83	0.85	0.82
	(0.58–1.18)	(0.58–1.24)	(0.57–1.18)
Education, College degree or higher (ref. less than a college degree)	1.04	1.04	1.07
	(0.71–1.51)	(0.70–1.56)	(0.73–1.56)
Household income	1.03	1.04	1.01
	(0.88–1.20)	(0.88–1.23)	(0.86–1.18)
Self-rated physical health	0.92 *	0.92 *	0.93 *
	(0.86–0.99)	(0.86–1.00)	(0.86–1.00)
Observations	1116	1116	1116

Note: 95% confidence intervals in parentheses. Exponentiated coefficients of logistic regressions. Mental Health in Congregations Study, 2020. *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1.

model respondents’ expectations of future attendance of a mix of online and in-person religious services (1) vs. returning to in-person religious services only (0) once the COVID-19 pandemic has ended.

With Catholics as the reference group and while accounting for the demographic and socioeconomic variables, Evangelicals (2.83 times the odds, 95 percent confidence interval 1.79–4.48), Reform Jewish (10.56 times the odds, 95% CI 6.98–15.98) and Hindus (6.17 times the odds, 95% CI 2.54–14.99) in our sample are much more likely to want to attend a mix of online and in-person religious services. Compared to Catholics, African American Baptist congregation members had 2.88 times the odds (95% CI 1.29–6.44) of expecting to attend a mix of online and in-person religious services, but this became statistically insignificant (p < 0.1) in model 2. The difference between Catholics and Reform Jewish and Hindu congregants is large: in terms of predicted probabilities, only 25% of Catholics but 64% of Hindu and 74% of Reform Jewish congregants intend to attend a mix of online and in-person religious services after the pandemic, even when accounting for demographic differences (Figure 1). There were no statistically significant differences between Catholic, LDS, and Orthodox Jewish respondents.

In terms of demographic control variables, African Americans have 3.71 times the odds (95% CI 1.80–7.64) and Hispanics have 2.69 times the odds (95% CI 1.66–4.35) of Whites of expecting to attend a mix of online and in-person religious services, thus suggesting that Whites are more likely to prefer in-person services only. There were no significant gender differences in model 1. However, self-rated physical health of the respondents predicts substantive differences in their expected future attendance modality, i.e., better health is associated with a preference for in-person religious attendance. Higher age is associated with a higher expectation of future online service attendance (e.g., 35% of congregants aged 30 vs. 56% of congregants aged 90 expect to attend a mix of online and in-person religious services). Both the health and age associations may correlate with the relative risks or levels of worrying about COVID-19. One may have expected the younger generations of congregants to be more attuned to virtual online modalities, but at least among our relatively religious sample, this is not the case.

The results of model 2 strongly show that more frequent in-person attendance before the COVID-19 pandemic is associated with significantly lower expectations of attending a mix of online and in-person religious services after the COVID-19 pandemic (0.55 times higher odds per unit increase in frequency of attendance before the pandemic, 95% CI 0.46–0.65). As illustrated in Figure 2 and when accounting for the covariates from model 2, respondents who only attended a few times per year or less before the pandemic have a predicted probability of 70% of expecting to attend a mix of online and in-person religious services, while for those who attended several times per week before the pandemic, the probability drops to 30%. The length of membership in the congregation was not statistically significantly associated with the dependent variable.

In model 3, an increase in the frequency of attendance during the COVID-19 pandemic, as compared with no changes, is associated with higher expectations to attend a mix of online and in-person religious services vs. expecting to attend in-person religious services only after the pandemic (2.26 times the odds, 95% CI 1.45–3.53). As seen in Table 2, those respondents who experienced an increase in the frequency of attendance during the pandemic had lower levels of in-person religious attendance before the pandemic. Moreover, for the vast majority, the increases in service attendance during the pandemic happened online. Additional analyses revealed that of those respondents who reported increases in service attendance during the pandemic, 81% had been attending online services 2–3 times per week or more frequently during the pandemic and 72% never attended in-person services during the pandemic. It further appears to be the case that online religious services created new “recruitment” opportunities and that respondents who joined congregations virtually in this manner are happy to continue with the online medium. Such speculations are also shared by leaders of some of these communities, as we discuss below.

4. Discussion

As there has been minimal research published in the US regarding the future of religious practice in congregations following the pandemic, the current study sheds light on how the pandemic affected both congregations in terms of the shift towards online religious services and congregants in terms of changes in religious commitment, and how those factors are associated with expectations of future online religious service attendance modalities after the pandemic. While the results of this study are not generalizable to faith communities in general, the study makes an empirical contribution to the knowledge base, providing data on specific localities (Washington, D.C. and San Antonio) and time frames (October–December 2020) which can be contextualized in future research.

Relating to the first research question, large differences in expected future religious service attendance modality were observed between religious traditions, with Catholic and Orthodox Jewish congregants expressing much higher likelihoods of expecting to attend only in-person services after the pandemic. Even though dispensations were provided by Catholic Bishops during the pandemic for virtual services, Catholics cannot receive the Eucharist online, and the dispensations were always understood to be of a temporary nature (O’Laughlin 2021). Similarly, many Orthodox Jewish rituals and practices require an in-person presence in a synagogue, and online religious services cannot offer a suitable replacement for those rituals. While our sample examined a Modern Orthodox congregation, more strict Orthodox traditions prohibit the use of technology on Shabbat entirely, making zoom services impossible, although this proscription is being debated and may lead to changes in the future (Cohen 2020; Rosenberg 2020).

In contrast, African American Baptist, Reform Jewish, and Hindu congregants in our sample were more likely to expect to attend a mix of both online and in-person religious services after the pandemic. In our sample, Reform Jewish members had low levels of pre-pandemic attendance, suggesting that some of those congregants became more connected to their faith and the community during the pandemic because of online services. The option of online attendance allows some members to participate in a lower-commitment, lower-cost way of engaging in communal religion, which may allow these congregations to retain marginal members who might otherwise drop out entirely. This reflects an appreciation for new types of online offerings in these committed communities that are not constrained the features of religious services that cannot be conducted online. According to qualitative interviews with the leaders of the Reform Jewish and Hindu communities in particular, the virtual services they offered drew in new members from outside their congregations and even outside the country. This might be an outcome of online services that encouraged new members to join a faith community, which they would not have done before.

The observed differences between congregations could point at future trends in which Catholic and Orthodox Jewish congregants mainly return to in-person services, while other congregations could focus on online services as a means of attracting new members or increasing the attendance rate of existing members. Thus, the pandemic appears to have contributed to a divergence in attendance modalities between different congregation types. Nonetheless, differences in religious tradition might partly reflect difficulties experienced by minority populations in attending in-person services during the COVID-19 pandemic (Routen et al. 2021).

In terms of the second research question, and while statistically controlling for other congregational differences, the results strongly suggest that the frequency of online religious service attendance during the pandemic is positively linked with expectations of a mix of online and in-person religious service attendance vs. returning to in-person religious service attendance only. We can speculate that many respondents tended to find positive aspects such as accessibility, comfort, or flexibility in online services which they would like to continue to benefit from in the future. Even if Catholics and Orthodox Jewish place strong values on in-person attendance, the results suggest that occasional online services might still be received positively, at least for the less committed members. While the present study can only provide insights for a limited time frame and while views on online services may change, the results of the study suggest that online attendance, for some, might have become a cultural practice that they will want to continue to foster even beyond the pandemic (Bourdieu 1990; Swidler 2013).

However, we also find that respondents with more frequent in-person religious attendance before the pandemic were much more likely to expect to return to in-person-only services rather than a mix of online and in-person services. In other words, congregants mainly expect to continue with the modality of attendance that they have grown accustomed to, which may be in-person services for the committed congregants and online services for more marginal ones. The findings therefore speak to the importance of habituation and habit building.

Our results show that increases in attendance during the pandemic were linked with preferences to attend online religious services in the future and that in most cases these increases happened from the relatively less committed members or congregation types with lower levels of in-person attendance before the pandemic. Previously less committed congregants might have become more religious as they turned to religion to cope with pandemic-related stresses (Cowden et al. 2021) or because online services were easier to fit into their schedule.

Nonetheless, for existing (pre-pandemic) committed congregants, the challenges faced during the pandemic might have also increased religious commitment and salience, especially levels of intrinsic religiosity in terms of the frequency of prayer or the importance of religion (Jacobi et al. 2022), since religion and religious practices provide psychological comfort during trying times (Cones 2021). Taken together, the findings speak to the multidimensionality of religious involvement during the pandemic and the differential impact of the pandemic on congregants of differing levels of baseline religiosity.

Limitations

This study is based on a limited sample and is not meant to be representative of the US population or of the congregations studied. The reported denominational differences are not necessarily representative of members of those denominations in general, especially if one assumes differential survey response rates between congregations. Regarding expectations of future attendance, the responses might be based on the experiences of the local congregation: if a faith community offered limited or poor-quality online services, then even if respondents would prefer to attend online, they may indicate in our survey that they are most likely to return to in-person only. Nonetheless, the strengths of the study lie in the large sample size, the ability to analyse self-rated changes (decreases, stability, increases) in service attendance, and in the level of detail regarding expectations for future attendance. Moreover, further analyses (not shown) in which multilevel models were employed to account for the statistical nesting of congregants (level 1) in congregations (level 2) yielded nearly identical results and the intra-class correlation (ICC) was low at less than 2%, suggesting that the presented logistic regression results are accurate and a good representation of the data.

Despite its limitation, the study provides a timely and nuanced glimpse of how congregations in the contexts and periods studied were affected by the COVID-19 pandemic and the extent to which the online services offered by new technologies might contribute to the sustained vitality of some congregations after the pandemic.

5. Conclusions and Implications

This article has presented new insights on the pandemic-related shift towards online religious services in faith communities and explored congregants’ expectations of future online and in-person religious service attendance after the pandemic.

The first set of findings identified divergences between congregation types, namely that Catholic and Orthodox Jewish respondents especially were more likely to intend to return to in-person services than other congregation types. While this may not be entirely surprising, it does demonstrate that online attendance modalities are to some extent constrained by the practices and contents of the different religious traditions. Whereas online services might become a crucial part of some congregations, it appears likely that they will only exist as supplemental or niche offerings in most Catholic and Orthodox Jewish congregations. The increases in attendance because of online religious services (in Reform Jewish and Hindu congregations) are promising, but if the increases prove to be of a lower-cost and lower-commitment nature, then those congregations might find it challenging to build and maintain cohesive faith communities.

Second, a habituation pathway has been identified, namely that the frequent practice of online attendance by individual congregants is associated with a higher likelihood of intending to continue attending online services in the future. Third, congregants who had been attending in-person services frequently before the pandemic were less likely to expect to shift to online services, which again points to the effects of habituation for attendance. Increases in attendance during the pandemic mostly occurred from respondents who were relatively less frequent attenders before, suggesting that engagement in online attendance for more marginal members had a positive impact. While highly religious members, especially if they are Catholic or Orthodox Jewish, are likely to return to in-person services, faith leaders more generally may want to invest in online service offerings to reach out to newer and less committed members.

Future in-depth qualitative research is needed to better understand the diverse meanings of religious service attendance across members of different faith traditions. Such research should especially pay more attention to what people find meaningful in new forms of online service attendance. Finally, this study focused on the specific aspect of religious service attendance, but future research should examine differences in intrinsic and extrinsic religiosity among faith community members and how each has changed during the pandemic and beyond.