This was written in 2007 and revised in 2012. Links may be wrong.
I plan to rewrite it again - as of January 2020, I'm gathering church attendance data.
The Church in Adelaide appears to be unevenly distributed, with more and larger churches in the eastern half all the way along, and a relative lack of Christians in the poorer areas. This document is mostly a quantitative investigation of the matter.
This is a topic I'm always thinking about, and many throughout the Christian community will have heard me discuss it at length. Mike Mills suggested I type this up, to facilitate further thought and discussion. Thanks and acknowledgement go to ABS who make available mountains of data; NCLS*, who have some handy info as well; and the many people who've informed my thinking, including Andrew Turner and Jason Hoet.
I originally wrote this in November 2007 and I've changed it plenty since then, most recently after working with the 2011 census data. Some of this is still out of date. Two sections of this grew into their own pages: Where then shall we live? and Counting Christians in the Australian census.
My findings in an easy single-page article "Thirsty hearts in the driest state" from the Baptist Churches of SA monthly newsletter (the link may not survive).
Reasons for the Distribution
Strongly related to the geographic imbalance is a much greater cultural imbalance, with educated or middle-class people better represented in the Church. NCLS has some ideas coming out of NCLS 2001.
Education is another area where church attenders do not reflect the community as a whole. Some 23% of attenders have a university or postgraduate degree, compared to 13% of the Australian population. (Anglicans and Baptists topped the class at 28%. As of 2011, 27% of church attenders have a degree compared to 16% of Australians. These figures count only adults.)
...the education gap between the church and community is closing... in part, because of the absence of young adults.
There has been a range of theories put forward to explain the relative lack of church involvement among people from lower socioeconomic groups...
Historical identification of the churches with the ruling classes.
The use of abstract thinking alienates the less educated.
Much of church liturgy, preaching and small-group life... involves abstract thinking.
They cite Who Goes Where? Who Doesn't Care? (Kaldor, 1987). Much has been written on this topic. Philip Hughes (Christian Research Association) goes further, saying the "knowledge class" is much more present in the Church than the "business class". Alan Hirsch in The Forgotten Ways says that the ministry of typical evangelical churches are suited only to reaching a 15% sector of the Australian population: "Demographically speaking, they tend to cater largely to what might be called the "family-values segment"-good, solid, well-educated citizens... who live, largely, what can be called a suburban lifestyle".
This class imbalance would account for much of the geographical imbalance. There are a few other likely factors: A church in a poor area is likely to get less income from its members and be less able to fund ministry. Poor areas often have high numbers of ethnic groups that have less Christians.
Although we are under-represented in some poor areas, we are not necessarily over-represented in all the rich areas. Some of the analysis below (which is far from conclusive) shows all of the Noarlunga area as more Christian than all of the eastern suburbs. But the well-churched areas would be described as pleasant ones. Often when I've asked Christians why they chose to live where they did - "It's a nice area" is a very common answer. The Christian population probably includes a higher than average proportion of families with children; this may explain the relative lack of Christians in the dense inner suburbs which are not suited to children.
Over the time I've studied this topic, I've liked to present my findings in a map format wherever possible. Earlier on I wrote a program to produce HTML/bitmap pairs to do this, but later I developed a web application called Mappage which uses Google Maps. Mappage can be found at mappage.net.au, with some things of specific interest to the Christian community at here. I've set up a map here that illustrates some of the things I talk about below.
Measuring the Distribution
The first time I noticed a metropolitan bible-belt was when I visited Sydney. A large rich part of Sydney is the North Shore - everywhere north of the harbour. The northern half of this (the upper north shore) and the area to the west through the Hornsby and Baulkham Hills shires is known for large numbers of strong churches. Thus the northern 20% of Sydney, most of the way across, has a higher concentration of Christians, and much of that area falls in the upper third of Sydney in terms of wealth (although the lower north shore, which is richer still, is not so Christian). Since then I've been interested in the distribution in Adelaide. Here are some of the quantitative methods I've used.
1. Church attendance data
This would be the best way to measure if all the data were available. My first source of data was from the SABU Handbook, representing less than 10% of Adelaide's church attenders, and by no means a cross-section. In the SE quadrant of Adelaide (Mitcham, Unley Burnside, Adelaide Hills and Mt Barker LGAs*) the number of attenders was a full percent of the population. The NE (Campbelltown, Tea Tree Gully) came next with 0.7%, and the rest of Adelaide scored lower. As this data is not precise I haven't gone into much detail here. The country areas scored the lowest, but other denominations are strong in the country: Churches of Christ in the SE, Lutheran in Barossa & Riverland, Uniting in the northern country areas.
Later I found the Lutheran Church membership numbers (and an attendance statistics page). In Adelaide, half the members are in churches east of the spine, a quarter are in city or foreign language congregations, and one quarter are in congregations to the west.
2. The Spine and locations of large churches
Adelaide is quite long in the north-south direction, and we can divide it roughly in half all the way along using Main North Rd, the city and South Rd (we'll call this line the Spine). It's close to an even split in terms of population (maybe 53% on the east). It was pointed out that most of the large baptist churches are east of this line. All the way along Adelaide, the east side is richer than the west. When I heard this, I looked into it further, and listed all the large churches I could think of, divided into west and east. See in Mappage.
There are well over two in the east for every one in the west (some are only just east of the line). A lot of them are also in the city, and their members tend to come from a wider area (they're still on the map). I don't know how big all these churches are, but I've included the ones I imagine to be large (attendance over 250) based on what I've heard of them, which potentially biases the list towards baptist churches and the east, and against Catholic and Pentecostal churches and the north & south. If anyone has any more info on which are the big churches, I'm keen to know, (eg someone who works in a denom office or who gets around a lot), indeed the map is editable for anyone who wants to add to it.
Beyond the simple W-E divide, a few things stand out. The Port Adelaide Enfield LGA is empty (apart from Oakden), along with Prospect & Walkerville. That empty strip continues south of the river, although much of this area has a high Italian population and large Catholic churches. The western half of Salisbury LGA and Playford LGA are mostly empty. The sparse Noarlunga area may be due to my ignorance. The Tea Tree Gully LGA and surrounds show strongly, as do the Mitcham Hills and Happy Valley areas. I also note the irony of numerous large churches in the city while few of the people in the city go to church.
The lack of large churches in the poorer parts of Adelaide is not the concern, but the uneven distribution of Christians is. It may be that the west and north of Adelaide don't need large churches like those appearing on this map. It may be that the kinds of Christian communities best able to reach people will not always have names or buildings. But it is unfortunate when some of the small number of Christians who do live in an unchurched area are driving a long way east to go to church rather than being part of a local one.
I often speak of the uneven distribution as a bad thing, mostly reflecting migration patterns of Christians, but I was reminded of the positive side: The strong places on the map also speak of churches staying strong during the decades when the Australian church was mostly in decline, and continuing to reach their communities today.
3. Using census religion data
I have done a lot of calculations using census data to estimate the distribution of Christians. That grew into its own page: Counting Christians in the Australian census. Using a formula that takes the religion statistics from the census, multiplies numbers according to the ratio of national church attendance to national census affiliation for each denomination, making various other adjustments to get a vague estimate of the Christian-ness of any area.
Looking at the SA3 level (mostly the same as LGAs in Adelaide), the city is the least Christian by a long way, followed by Unley, Prospect-Walkerville, Burnside and Port Adelaide. Campbelltown comes first, followed by Tea Tree Gully. Playford has risen from the bottom of the pile to near the middle and I cautiously attribute much of this this to a large number of central African Christians moving there. Most country areas except the outback do better than Adelaide.
If we restrict our count to attenders from Protestant churches, Adelaide Hills comes first, followed by Tea Tree Gully, with all of the northern and southern suburbs beating all of the western, inner and eastern suburbs.
Looking at the places that score the highest, they're not the richest ones but they would be readily described as "a nice area". Niceness is hard to quantify, but in the Advertiser on 30th May 1998 there was a feature ranking the metropolitan postcodes from the best place to live to the worst (I don't remember the criteria, which were not given in much detail). The Reservoir comes in at #2, and every postcode in Tea Tree Gully, Campbelltown and Adelaide Hills LGAs comes in the top 20, while the bottom 25 consists mostly of the inner suburbs and western parts of the northern suburbs. Here's a map showing the Advertiser's list.
The correlation is remarkable - the "nice" postcodes seem to be the most Christian ones. Here is a graph plotting the Advertiser ranking for those 100 districts with its estimated church attendance figure (to be precise, halfway between the total % and the protestant % - preventing the graph being thrown one way or the other by the Catholic areas).
While attenders from denominations with a high multiplier will tend to be measured correctly, those with low multipliers might not be. There are nominals in every district, and a much higher or lower number of attenders will get washed out. This doesn't seem to be the case with the Uniting figures, which correlate strongly with the total protestant figures and are highest in areas with large Uniting churches. Anglicans in Adelaide number 14% in the census and I'm using a multiplier of 5.7%, so they contribute an average of 0.8% to the attendance numbers. I don't think this introduces any large errors, as I don't know of any large concentrations of Anglican attenders. The Catholic Church accounts for nearly half of attenders nation-wide but I don't have any data on distribution of attenders beyond the census.
4. Using contact lists
Sometimes I've looked at a contact list from something in the Christian youth community which supposedly pertains to the whole city or state to look at where everyone lives, and I've sometimes noticed a bias toward the south and east. The best sample I got was the list of people registered as Encounter Schoolies volunteers in 2007. The mission there was in its ninth year and quite large, so I thought the numbers from different regions would be an indicator of the strength of the Church in those regions. I was given the postcode column from the database, and for several postcode ranges I calculated the number of schoolies volunteers per 1000 of young adult population. The results were fairly uneven: The west/NW and outer north scored below 1, while the inner south, Adelaide Hills (not including the YWAM base at Norton Summit) and part of the outer south scored over 5. The sample was over 350, and one might expect a bit of a southward bias as it's easier to get to Victor Harbor from the southern suburbs. This sample would also reflect "spare capacity" of churches rather than their overall numbers.
5. Conclusion of quantitative analysis
The main thing that stands out is that in the north (Salisbury, Tea Tree Gully and Playford LGAs) there is a definite eastward bias, with Tea Tree Gully beating Salisbury and Playford in every measure. This corresponds with the wealth gradient. Even within Salisbury and Playford, the eastern areas like Playford Hills have more Christians. Port Adelaide Enfield LGA is noticeably weak in all measures, particularly the Parks area and surrounds. The city centre has many churches but very few attenders. The inner suburbs within the Circle Line bus route apart from Unley LGA have no large churches on the map and score low in the other measures. Further east and west the scores are mixed. Adelaide Hills and Mt Barker LGAs score strongly in all measures. The southern LGAs of Onkaparinga, Marion and Mitcham collectively score higher than average, with the Reservoir area standing out.
The 2006 census occupation figures for the number of "Ministers of Religion" resident in each area follow along the same lines. The Reservior and Mt Barker are highest with 18 ministers per 10,000 population. Less than 4 (these are very rough) are Paralowie, the Peachey belt, Parks, Glenelg, Thebarton and Port Adelaide.
I thought the main feature of the distribution would be Christians in higher numbers in rich areas and I was going to say how good it would be if the next ten years saw the numbers shift in favour of the poor areas from the rich ones. But as it seems the "nice" factor is the one that predicts where Christians are living, my complaint is somewhat diluted, with the nice factor correlating negatively with the rich factor. (In fact the rich factor could be estimated closely by proximity to the city plus a bit of the nice factor).
One section of this page, which considered the implications of the distribution, grew into its own page: Where then shall we live? It gets away from the stats and discusses a number of issues which might affect our choice of where to live.
Does anyone have any thoughts or questions? Anyone want more details on another part of Australia? Or does anyone have any other similar data? This is an ongoing area of study for me and I'm keen to pool insights and data with anyone else who studies this topic.
I'd love to hear, by email or comment.
NCLS: National Church Life Survey - refers (a) to a 5-yearly census of Australian churches and (b) the organisation that carries it out and does other research.
LGA: Local Government Area (often called council area, shire or "City of ---").
Reservior area: Aberfoyle Park, Flagstaff Hill and some of Coromandel Valley. (Or postcode 5159, which includes Happy Valley as well).