Since the launch of Census 2011 the AIRO mapping team have developed a series of interactive mapping tools to visualise the results on a national and local authority/regional authority level (see here). This has been a joint project with the Central Statistics Office (CSO) and to date has been very successful with a high number of users viewing and interacting with this publically funded dataset. The aim of this collaboration was to improve access to the results of Census 2011 and thereby make a contribution towards improving evidence informed planning in Ireland. Over the last couple of weeks the team have been working on the recently released Place of Work, School or College Census of Anonymised Records (POWSCAR) dataset.
This dataset contains 2.78 million records where the location of the place of work, school or college was coded for each person on the basis of the reply that was given to Question 34 on the census form: “What is the Full Name and Address of your place of work, school or college”.
Using this information the CSO matched the employer/school name and address against addresses on the An Post GeoDirectory. In the case of workers, where the coder could not find an exact match they coded to a near match if they could find a GeoDirectory address on the same street or in the same town as the address stated on the form. In the case of students, an exact match was only accepted for the school or college address. The coordinates retrieved from the GeoDirectory match were then linked back to the place of work, school or college Electoral District (ED) and Town and Small Area by superimposing digital boundaries. In some cases it was not possible to match the destination of the worker/student with GeoDirectory, this was a result of a very poor return of address information or the workers destination was classed as being ‘mobile’. The final dataset is effectively an origin-destination matrix that links the place of residence of the worker/student, either at Electoral Division (ED) or Small Area (SA) level, to the work/school/college destination of the worker/student at the ED, SA and 250m grid level. The dataset contain a wealth of information about each work trip such as age, gender, industry of employment, education level, mode of transport, household occupancy status, one-off housing indicator, socio-economic group etc. Similar data is available about those attending schools/college although not as detailed and much of the data is compressed for disclosure reasons – we will do a further piece on this in a couple of weeks.
In 2011, places of work, school or college with an address in Northern Ireland were also coded in the same way by utilising the NI Pointer address database. NI County, Ward, Towns (2001) and 2001 Census output areas were derived by superimposing digital boundaries. This is a big step forward in understanding the cross-border travel to work catchments that exist in Ireland and the CSO should be commended for realising the benefits of going this extra step in the development of this dataset. Where the person indicated a work, school or college address abroad these records were coded to a specific code to indicate that the person was working abroad i.e. outside Ireland or Northern Ireland.
Due to the level of detail available within the dataset it is not as ‘open’ as the rest of the Census 2011. As there are some minor risks to data disclosure, use of the dataset is restricted and is only available to bone fide researchers who are approved by CSO and signed up as Officers of Statistics for the duration of the research they propose to undertake. A key point on all of this is that All material published from POWSCAR must be approved in advance by CSO.
The following table gives a summary of the POWSCAR address coding process:
| Students | Workers | Total | % | |
| Persons in private households or establishmentsenumerated and resident in Ireland | 1,013,292 | 1,770,644 | 2,783,936 | 100.0 |
| Place of work, school or college address (Q34)was matched to a GeoDirectory address point | 929,154 | 1,362,742 | 2,291,896 | 82.3 |
| Place of work, school or college address (Q34)blank or uncodeable | 78,956 | 147,251 | 226,207 | 8.1 |
| No fixed place of work indicated at Q34 | ‐ | 148,177 | 148,177 | 5.3 |
| Works from home indicated in Q34 | ‐ | 106,055 | 106,055 | 3.8 |
| Place of work, school or college address (Q34)was matched to a NI Pointer database addresspoint | 3,117 | 6,419 | 9,536 | 0.3 |
| Place of work, school or college address (Q34)overseas | 1,447 | ‐ | 0.1 | |
| Home school indicated at Q34 | 618 | ‐ | 0.0 |
AIRO POWCAR Mapping:
To get started on work with the POWSCAR dataset the AIRO team have developed travel to work catchments for all 22 Gateways and Hubs and made these accessible via an interactive mapping tool. Each map is based on the percentage of workers within each ED that work within the selected settlement boundary (boundaries based on CSO Settlements). Our analysis here is only based on the workfore where we have information on the destination of workers and therefore excludes those classed as Mobile workers (148,177 or 10% of workers) and workers with an uncodable or Blank destination (147,251 or 9.9%). Users can select a settlement to view the extent of the catchment and then click on an ED to get information on the number/percentage of workers employed within the selected settlement.
The map below details the extent of the travel to work catchment for the Dublin City settlement boundary, an area of about 317 sqkm and including all major employment locations in the local authorities of Dublin City, South County Dublin and DLR. The settlement boundary does not however include many of the large employment locations in south Fingal such as Dublin Airport, Swords, Malahide and Portmarnock. In total there are 457,046 people with a work destination in the Dublin Settlement boundary. Of these, 74% also reside within the Dublin City settlement and 26% reside outside the settlement boundary highlighting the high levels of inward commuting. Of the workers who reside within the Dublin City settlement boundary a total of 12.5% (48,801) commute out of the settlement to employmant destinations.
Red areas on the map highlight where >50% of the workers in an ED are employed within the settlement boundary. The orange band represents areas where over 30% of workers in an ED commute to the Dublin settlement boundary and extend to towns such as Drogheda, Ashbourne, Maynooth, Newbridge, Blessington and Wicklow town. The pale orange band is based on 10% to 30% of workers and extends much further towards the Mid East, Midlands and parts of the Border region with towns such as Kells, Navan, Mullingar, Portarlington, Portlaoise, Gorey and Dundalk.
To access the tool and view the different catchments for all Gateways and Hubs please click here. In the map viewer, click on an area to get specific information on that ED.
A link to the tool is also available on our census mapping home page on the AIRO site where you can also access other mapping tools developed over the last number of months. View Census Mapping home page
This is the first step at mapping the travel to work catchments for the main settlements in Ireland. We’re hoping to do some additional work on job’s desnity within settlements and also roll out catchments for other towns. Happy to take suggestions on what is useful for planners, policy makers and general public who would find this information of use.
Justin Gleeson & Eoghan McCarthy
Ireland after NAMA 
November 21, 2012 at 6:23 pm
“What is the Full Name and Address of your place of work, school or college”.
For agents of the state to compile a complete database of such precise information about every citizen is worrying.
The desire to pinpoint people’s exact movements from home to college or work has overtones of Big Brother.
Did this question affect the number of people returning their census form ?
November 21, 2012 at 6:40 pm
It is worth pointing out that the State does know which schools kids go to and who is employed with whom without the census. They have to do run the education system and collect taxes. What this data is trying to do is enable an understanding of commuting and to enable appropriate transport and labour market planning. Individual data is not released by the CSO, not even to other government departments (and they can’t get the data from other depts due to data protection legislation). We would be very mindful of issues such as ‘Big Brother’ but at the same time we need good, robust evidence to underpin policy analysis and formulation.
November 22, 2012 at 9:29 am
Hi there, looks great. I would like to ask for something much more simple. I would like a table and map of Ireland showing how the work and study locations are distributed, ie the number of study and work places per square km for each Small Area.
This would give an idea of how daytime economic activity is actually distributed. It would be useful and might throw up a few surprises. It would not be sensitive at all because it would not involve a cross-walk.
November 22, 2012 at 3:20 pm
Hi Antoin,
Will look into it.
Justin
November 29, 2012 at 8:01 am
First, I wish to congratulate on the work you are doing in AIRO. It is simply great (long overdue) that you are developing tools and making existing data more useful. It is also great that you are making the results of your work available to the public. Your work will certainly contribute to both drawing up policy and assessing policy using evidence – something we cannot assume applies in this Republic.
On this basis, I want to express my surprise that you have used a sample map which excludes large parts of North Dublin, but does include South Dublin.
“The map below details the extent of the travel to work catchment for the Dublin City settlement boundary, an area of about 317 sqkm and including all major employment locations in the local authorities of Dublin City, South County Dublin and DLR. The settlement boundary does not however include many of the large employment locations in south Fingal such as Dublin Airport, Swords, Malahide and Portmarnock.”
What on earth were you thinking of?
Twenty years ago, the Interim Report of the Dublin Transportation Initiative reported “Of all the locations in the Study Area, the Airport is the one site where it can be said with high levels of confidence that there will continue to be, year-on-year for a very long period, high increases in demand”
Since 1992, traffic through the airport has increased hugely since then. There surely has been a very significant increase in those whose principal place of employment is Dublin Airport.
In 1996, an EU-prompted study showed that there was greater potential light-rail passenger demand (trip attractors/generators) on the route from O’Connell Street, via Croke Park and Drumcondra to Ballymun than on either the then planned Tallaght or Dundrum lines. Government did not act on this information, but instead extended one LUAS line to Sandyford (and later to Cherrywood) and the other to the Point (and later to City West).
Since 1996, Croke Park has become a major venue with conferences and more frequent matches, concerts etc. I suspect that there has been some increase in employment at Croke Park, even if it is temporary.
I suggest that DCU has also expanded as a place of work and study.
I now ask you to do some more work to redress the imbalance in the map you have presented.
Living as I do in Drumcondra, I find it deeply irritating that data for transport policy makers continues to focus on areas which have been clearly shown to have less trip attractors/generators that North Dublin. AS an example of this bias, note that
1) When RPA opened the Red Line LUAS extension to Cherrywood, it did not open two stations which it had built and equipped;
2) Similarly, when Iarnród Éireann upgraded the Kildare line, it did not open a newly built station
In both cases, the reason is that expected property development did not take place.
While passenger numbers in Dublin Airport may have declined from the peaks of the mid 2000s, I suggest that associated employment is still very high.
Judging by the regular and persistent reports of traffic congestion on the main road between the city centre and the M1 (despite the Dublin Port Tunnel) suggests that the north city fringe/South Fingal (eg. Swords) merits as much attention from you as does other parts of the Greater Dublin Area.
November 29, 2012 at 6:44 pm
Donal,
Thanks for your comments. The initial mapping that we did here was just based on the settlement boundaries as defined by the CSO and included within the POWCAR database. The idea was to just get the first level of mapping out there and then get comments and suggestions from readers.
What areas would you suggest appending to the Dublin settlement boundary (as per map) that would more accurately represent the full set of Dublin employment destinations?
thanks
Justin
November 29, 2012 at 10:40 pm
Juxstin,
I thought I made it clear
1) Dublin Airport
2) Swords
3) areas between city centre and Dublin Airport- Swords.
Are you are saying that the CSO settlement areas and the POWCAR do not include the residential/educational districts of North Dublin whose inhabitants work in the places like the Airport and the industrial/distribution areas of the North City/South Fingal?
Perhaps, you would explain
1) by what principle/criteria these areas are deliberately excluded (“The settlement boundary does not however include many of the large employment locations in south Fingal such as Dublin Airport, Swords, Malahide and Portmarnock.) in a definition of
“the Dublin City settlement boundary, an area of about 317 sqkm and including all major employment locations in the local authorities of Dublin City, South County Dublin and DLR.”
2) who/what group made such a decision;
3) set out the reasons for this decision, which seems bizarre.
If this is what you have based your mapping on, it suggests
1) selectivity in what it actually covers;
2) inaccuracy in what it presented;
3) leading to tendentiousness in policy that may be based on your work.
Have I misunderstood something?.
November 29, 2012 at 8:19 am
I also suggest that you map the results of the Census question which gives data on those who have access to a car. If I remember correctly, more than than 40 percent of Dublin City households (ie. the area administered by Dublin City Council) do not have access to a car. Please contrast this with other areas in the Greater Dublin Area.