We really don’t have generalizable data on how widespread (or not) the practice [of unordained women serving as deaconesses] is in the PCA. How many churches have deaconesses? How many deaconesses are there in the PCA? The purpose of this project is not to pick a fight, but to shed light, in the hopes that it will lead to more productive debate at PCA General Assembly.

Overture 26 from Northwest Georgia presbytery proposes a change to chapter 7 of the Book of Church Order that would disallow unordained people from being “referred to as, or given the titles connected to, the ecclesial offices of pastor, elder, or deacon.” RE Brad Isbell wrote that the overture effectively addresses “a big ecclesial deal” and helps the PCA “get ahead of things for once,” since there seems to be some lack of clarity (or at least consistency) on the issue. More recently, Isbell provided some examples of the practice in the PCA. On the other side of the debate, TE Tim LeCroy warned of the coming fight with the “far right of our denomination”: “Watch out! Do you have unordained women serving as deaconesses?” But we really don’t have generalizable data on how widespread (or not) the practice is in the PCA.

How many churches have deaconesses? How many deaconesses are there in the PCA? The purpose of this project is not to pick a fight, but to shed light, in the hopes that it will lead to more productive debate at PCA General Assembly.

Method

We drew a random sample of presbyteries in the PCA, stratifying by US Census region. We stratified by region so that at least two presbyteries were chosen from each region to ensure geographic representation. We sampled more presbyteries in the South region, a region densely populated with PCA churches and presbyteries. Random sampling is important because it allows for generalizable inference. Randomization is important because, since each presbytery had an equally likely chance of being chosen, it allows us to say that our findings are generalizable within a certain margin of error. This method is similar to what pollsters using during election season to claim that a candidate is polling at some level, plus or minus some margin of error.

We also drew a random sample of presbyteries from two sister denominations in the North America Presbyterian and Reformed Council (NAPARC): the Orthodox Presbyterian Church (OPC) and the Reformed Presbyterian Church of North America (RPCNA). Since these denominations are smaller, we did not do a stratified randomization. In total, we sampled 14 presbyteries from the PCA, 4 presbyteries from the OPC, and 2 presbyteries from the RPCNA (see Table 1). These sister denominations are good comparisons for this analysis for a few reasons: (1) the PCA has fraternal relations with both of them; (2) they share doctrinal standards (Westminster); (3) the PCA and OPC do not allow for deaconesses as an ordained office, while the RPCNA does.

Together, these 20 presbyteries have 465 congregations. We excluded 36 churches from our analysis if no functioning website could be found or if the website was predominantly in a language other than English. We retained over 90% of all churches sampled in each of the three denominations.

The purpose was to capture what is clearly portrayed on each church’s website, rather than to conduct a deep investigation into each church’s website. As such, we typically spent no more than 30 seconds per website to count the number of Teaching Elders (TEs), Ruling Elders (REs), deacons, and deaconesses, typically summarized on a “Leadership” or “Officers and Staff” page.

To be counted as a deaconess, a website had to explicitly identify the woman as a deaconess. Ours is therefore a very conservative estimate because if there were ever any reason not to count a woman as a deaconess, we did not count her. Deaconess with parenthetical note? Nope. Mercy team? Nope. Women to pastors? Nope.

Limitations

Before we proceed to share what we found, the reader should keep in mind that this kind of research is subject to several limitations.

Because the practice of having ordained deaconesses is de jure not allowed in the PCA and the practice of having unordained deaconesses is contested (see this year’s Overture 26, for example), PCA churches may be pressured not to report deaconesses, even if the practice is de facto in place. Indeed, many of the websites we found mentioned “deaconesses” or “women on the diaconate” without listing the number or names. These churches were coded as having zero deaconesses for the purpose of our analysis. As such, we suspect that our findings are a lower bound estimate, underreporting the practice in the PCA.

Our method does not allow us to account for churches that forgo ordination of deacons and commission “mercy ministry teams” in lieu of a diaconate (e.g., Evergreen Church). or in addition to a diaconate (e.g., Christ Presbyterian in Santa Barbara, CA). These were not counted in our data. Others list deaconesses with a clarifying note that these are not considered ordained officers (e.g., University Reformed Church).

Different readers will come to different conclusions about our assessment of titles and practices, so we expect there will be competing views about the inferences that can be drawn from the data. Nonetheless, we believe it will be helpful to both sides of the debate to have some data on the issue.

Findings

1. PCA churches have nearly as many publicly listed deaconesses on average as the RPCNA, a denomination that allows for women to hold the office, but the practice is less widespread in the PCA.

In our website searches, we found that PCA churches have 0.19 deaconesses listed on their websites on average while RPCNA churches have 0.27 deaconesses listed on their websites on average. However, only one in twenty-five PCA churches listed deaconesses and six presbyteries did not have any deaconesses (Ascension, Canada West, Georgia Foothills, Ohio Valley, Tidewater, and Westminster), while the practice was almost three times as common in the RPCNA. The PCA churches listing deaconesses had 4.3 deaconesses on average and RPCNA churches doing the same had 2.0 deaconesses on average. The 101 OPC churches in total listed zero deaconesses on their websites.

2. PCA churches are more likely to give the impression of being “staff led” and OPC and RPCNA churches are more likely to give the impression of being “officer led.”

This is not to say that PCA churches are not in practice “officer led,” but as far as what is reported on their websites, they are more likely to give the impression of being “staff led.” A smaller proportion of PCA churches list only a TE on their websites (17%) than either OPC churches (21%) or RPCNA churches (21%). Similarly, a smaller proportion of PCA churches list only elders (ruling or teaching) on their websites (21%) than either OPC churches (46%) or RPCNA churches (41%). In contrast, PCA churches are more likely to list only TEs and staff (16%) than either OPC churches (one church) or RPCNA churches (zero churches).

3. The practice of having deaconesses appears to be common in some presbyteries, less prevalent in others.

On average, most churches in our analytic sample list between one and two TEs, two and four REs, and two and four deacons. There are, of course, some exceptions to that rule. Churches in Central Carolina and Tidewater had over five ruling elders on average and, in the latter presbytery, over six deacons on average. Much of this variation is explained by church size and membership.

There is more variation when it comes to having deaconesses. It should come as no surprise that not all presbyteries are the same with respect to this practice. Six presbyteries in our sample did not have a single deaconess listed on their churches’ websites (Ascension, Canada West, Georgia Foothills, Ohio Valley, Tidewater, and Westminster). Metro Atlanta had the most, with 0.79, followed by Central Florida (0.34), Pacific Northwest (0.28), and Great Lakes (0.19).

Matthew Lee is a ruling elder at Covenant Presbyterian Church (PCA) in Fayetteville, AR, where Liam Carr serves as a deacon.