Open Cities Index Report Series
Municipal Open Datasets: AN ANALYSIS OF DATA ACCESSIBILITY
PSD’s Open Cities Index (OCI) is Canada’s first benchmarking study for municipal open data initiatives. The OCI launched in 2015 with 34 participating municipalities. In fall 2016, the second iteration of the OCI was published with 68 municipalities, representing 61 percent of the Canadian population. The 2016 OCI Report includes an overview of Canada’s Top Twenty Open Cities and national trends in municipal open data initiatives.
Leading up to the launch of the third edition of OCI in the fall of 2017, PSD will publish four quarterly reports
exploring municipal open data challenges, solutions and best practices. Each themed report will include further data and analysis generated from the 2016 Open Cities Index study. After the publication of each OCI Report, PSD will convene a virtual round table
of municipal open data practitioners to discuss the report findings, identify areas for additional research, and facilitate collaboration and the sharing of best practices across the country.
to learn how you can participate in an upcoming virtual round table discussion.
Report 1) Municipal Open Datasets: An Analysis of Data Accessibility
Report 2) Municipal Open Data Budgets, Resources & Capacity
Report 3) Metrics & Mechanisms to Measure the Impact of Open Data Initiatives
Report 4) Models for Open Data Partnerships & Collaboration
MUNICIPAL OPEN DATASETS: AN ANALYSIS OF DATA ACCESSIBILITY
This first report in the OCI Report Series looks closely at data accessibility for municipal open data programs. An open data program can only be successful if quality data is readily accessible to municipalities. As municipalities seek to advance their open data initiatives, it is important to understand what types of data the public is seeking and how municipalities can access those datasets. The answers to these questions may vary from one community to the next, however there are some insightful trends in the types of data municipalities across Canada report as being accessible. These insights may be used by open data practitioners to further refine their open data initiatives and point towards datasets that should perhaps be made a priority for publication in 2017. If a dataset is largely inaccessible to most municipalities, for instance if that particular data is collected by provincial governments, but residents are increasingly requesting that dataset from their municipal government, a data-sharing arrangement between orders of government may be the solution. In the interim, local governments may wish to provide links to provincial open data portals to facilitate easy access to the sought-after dataset. The fourth report in the OCI Report Series will cover "Models for Open Data Partnerships & Collaboration."
In the Open Cities Index survey completed by our 68 participating municipalities, 32 specific datasets were highlighted. Through an international scan of open data indices, a review of Canadian municipal practices in open data, and through feedback gathered from participants of the 2015 OCI, these 32 datasets were deemed important for inclusion in the Open Cities Index evaluation.
"An open data program can only be successful if quality data is readily accessible to municipalities."
Access to data
For each dataset, there were 11 attributes that participants could select, if applicable, in order to better convey the quality of the data. The first two attributes allowed the participant to indicate if they did not have access to the dataset or if the data did exist in some form. The table below shows the number of participants that reported that they had access to each of the 32 datasets.
Table 1 Number of participants reporting access to datasets
The top 5 most accessible datasets:
1) Base GIS Data
2) Government Budget
3) Public Facilities and Structures
4) Park Inventory
5) Zoning (GIS)
The top 5 most accessible datasets are almost all related to asset management. As one of the main responsibilities of a municipality, and with the 'tangible nature' of infrastructure and asset management data, it is reasonable to expect that most municipalities would report having access to asset management-related datasets, such as public facilities data and park inventories. It is commonly expected that public works departments (or individual departments responsible for each asset category) would have access to most of this information, although the quality of the data depends on the quality of municipal data management practices and the frequency of condition assessments conducted by the municipality. For the second most accessible dataset - government budget - PSAB standards and provincial legislation both require municipalities to report annually on their finances, ensuring that government budgetary data is likely highly accessible for most communities. Furthermore, the public has come to expect a certain level of transparency when it comes to the budget process for governments. Proposed and adopted budgets are already made available on municipal websites, usually in PDF format. It is up to the municipality's open data team (or lone champion) to convert a PDF budget into machine-readable data for greater open data accessibility.
When it comes to the municipalities that have reported not having access to these most commonly accessible datasets, there are no clear trends across these communities. These municipalities range in both population and geographic location, therefore it is likely unique circumstances rendering these common datasets inaccessible.
The top 5 least accessible datasets:
1) Education Performance
2) Lobbyist Information
3) Health Performance
4) Air Quality
5) Restaurant Inspection
These datasets are not surprisingly the least accessible to municipalities, as reported by our participants, since most of the government functions/services associated with these datasets fall outside the jurisdiction of local government. Within Canada, the management and delivery of education programs is the responsibility of each provincial government in conjunction with local school boards. Data on education performance
within each community does exist, however it is largely data collected by provincial governments, rather than municipalities. The City of Edmonton, OCI's 2016 Most Open City, has included this data in its portal by providing a link to the Education Section of the Government of Alberta's Open Data Portal
, which provides information on education performance within each provincial district. For the residents of Edmonton hoping to access education performance data for their community, finding a link to the province's open data portal directly within Edmonton's portal allows for ease of access.
Within each province, there is an official lobbyist registry
that applies to all governments within each provincial jurisdiction. This registry is public and fully accessible - a measure to strengthen transparency in the public sector. Some municipalities, such as Toronto
have their own municipal lobbyist registries, facilitating the inclusion of this data in their respective open data portals. For most other municipalities, a local lobbyist registry does not exist, or does not exist in a format conducive to open data publication.
, like education performance, largely falls under the jurisdiction of provincial governments. The City of Toronto, however, does have some health indicators included in its suite of open datasets. The Wellbeing Toronto
platform includes data on female fertility, premature mortality, and cancer screenings.
in a community is measured through the Air Quality Health Index (AQHI). The federal government, in conjunction with provincial governments, have the infrastructure to measure air quality across the county. Within each province, the data is released in various formats either in real time or in annual reports. The City of Edmonton has included this data in its portal by linking to the Alberta Air Quality
website which includes an API for exploring the data.
are managed differently within each province. In most jurisdictions, restaurant inspections are handled by a department within the provincial government. In British Columbia
, and Saskatchewan
, restaurant inspections are managed by regional health authorities, which are independent from municipalities. However, in Toronto
, and Montreal
, restaurant inspections are handled by the municipality themselves and so the data is included in their respective open data portals. Within Ontario, upper tier local governments are also responsible for inspections within their region. Depending on the circumstances of each province, municipalities will have different hurdles to jump in accessing restaurant inspection data. As a highly sought after dataset (a topic for a forthcoming OCI Report), municipalities may wish to make the effort to collaborate with other organizations, if possible, to provide restaurant inspection data centrally in the municipality's portal. Again, in an ideal world, a resident, researcher, or developer seeking data specific to one community, should be able to find that data in the most logical location: within the open data catalogue or portal of the community in question.
Data variation by province
Evidently access to data differs depending on the type of data. Looking at the breakdown by province, we find even further variation across jurisdictions. Although 9 provinces and territories were represented in the 2016 OCI, 4 provinces (Alberta, BC, Ontario and Quebec) will be considered for this analysis as there were at least 4 participating municipalities from each. The table below shows the ranking of the datasets according to access for each province, as well as the overall ranking for all 68 municipalities.
Table 2 Ranking of dataset accessibility by province - 1 indicates most accessible and 32 indicates least accessible
The top 5 most accessible datasets for Alberta municipalities appear to be in line with the overall top 5 discussed above. Looking beyond the top 5, there are some apparent differences. In Alberta, municipalities are mandated to complete their own property assessments for property tax purposes. Therefore, municipalities readily have access to this information and so property assessment data is ranked much higher in Alberta than nationally (3rd most accessible dataset in Alberta versus 16th nationally).
Recreational programs data is also ranked much higher in Alberta (1st versus 8th nationally), while city services and procurement contracts are both ranked lower in Alberta than in the national ranking (these datasets are tied for 12th in Alberta compared to 6th in the overall ranking.)
Similar to Alberta, the top 5 datasets in BC are mostly in line with the national ranking. Property assessments are ranked significantly higher in BC than in the national ranking (1st versus 16th), however unlike in Alberta, a provincial body, BC Assessment is responsible for conducting property assessments. There may be an arrangement between BC municipalities and BC Assessment to facilitate the transfer of this data. Again like Alberta, recreational programs data is ranked higher, while procurement contracts is ranked lower. Also in BC, company register data is much more readily available than in the rest of Canada (10th in BC versus 25th overall).
Since more than half of the participants in the 2016 OCI are Ontario municipalities, it is no surprise that the Ontario ranking generally follows the national ranking. The most significant divergence is with the recreational programs data, which is the 13th-most accessible dataset in Ontario compared to 8th in the national ranking.
With only 4 participants from Quebec, it is more difficult to draw significant conclusions about data accessibility across Quebec municipalities. However, there are a few notable trends. Recreational programs data is ranked significantly lower in Quebec (24th) than nationally (8th). Considering the ranking for the four provinces in comparison to the national ranking, it seems the eastern and western provinces have a discrepancy in access to recreational programs data. Our four Quebec municipalities also reported significantly greater access to real-time transit data, ranking this dataset at number 1, compared to 25th in the national ranking. This variation could be due to the small sample size from Quebec, consisting of all fairly large cities, since it is larger cities that are more likely to have an advanced transit system with the capability to collect real-time transit data.
Data variation by local government type
Upper tier and lower tier municipalities differ in their responsibilities and in the services that they deliver to the public. Therefore, data accessibility will also vary across the type of local government. The table below shows the ranking of dataset accessibility for the group of 10 upper tier municipalities in the OCI sample in comparison to the national ranking.
Table 3 Ranking of dataset accessibility by local government type
The top 3 lists for both upper tier municipalities and for all municipalities are similar, although there are several notable differences. The biggest divergence is with health performance data, which is ranked 16th among the upper tiers in comparison to 30th in the national ranking. As previously discussed, in Ontario many health services are managed by regional health authorities, which are often aligned with the upper tier local governments of the region. Therefore, these upper tier governments have easier access to this information in comparison to single and lower municipalities. This is the same case for restaurant inspections in Ontario, with upper tiers reporting restaurant inspection data as the 9th most accessible dataset, compared to 28th for all municipalities.
Election data is ranked significantly higher nationally than it is among upper tiers (8th nationally versus 21st for upper tiers). This is likely due to the unique circumstances surrounding the election of regional municipal councillors and regional chairs. Finally, recreational programs data is ranked significantly lower for upper tiers (27th for upper tiers versus 8th overall), likely because upper tier governments do not typically provide the same level of recreational programming as their lower tier counterparts. In an OCI Round Table discussion with a sample of upper tier local governments, it was discussed that there are other datasets beyond our list of 32 that are more pertinent to upper tier municipalities - for example, several upper tiers reported that in lieu of park inventory data, they have access to beach inventory data, with many regional governments in Ontario having responsibility over the beaches in their area. Moving forward, PSD will work with upper tier municipalities to further refine the Open Cities Index to reflect their unique responsibilities and services.
Join the Discussion
On November 24th at 1pm ET, Open Cities Index member municipalities will gather virtually to discuss the findings in this report and share best practices in improving data accessibility. Following the virtual round table session, PSD will share a synopsis of the discussion with participants, which will include a "Municipal Guide to Open Data Accessibility."
Please contact us
to discuss how your municipality can participate in our virtual round table series and start advancing your open data initiative.