Statistics Canada
www.statcan.gc.ca
Common menu bar links
View the most recent version.Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.
4. Data quality
Lineage
Positional accuracy
Attribute accuracy
Logical consistency
Consistency with other products
Completeness
Spatial data quality elements provide information on the fitness-for-use of a spatial database by describing why, when and how the data are created, and how accurate the data are. The elements include an overview describing the purpose and usage, as well as specific quality elements reporting on lineage, positional accuracy, attribute accuracy, logical consistency and completeness. This information is provided to users for all spatial data products disseminated for the census.
Lineage
Describes the history of the spatial data, including descriptions of the source material from which the data were derived, and the methods of derivation. It also contains the dates of the source material, and all transformations involved in producing the final digital files.
Road layer
The data in the road layer were derived from the National Geographic Database. The National Geographic Database is a spatial database that contains the road network in Canada, as well as road attributes (name, type, direction, and address ranges). The National Geographic Database was originally built from four main data sources:
- Statistics Canada Street Network Files (SNFs)
- National Topographic Database (NTDB) 1:50,000 and 1:250,000 maps
- Digital Chart of the World (DCW) 1:1,000,000 maps
- Elections Canada road data
Additional road information was incorporated from a variety of other sources, including municipal maps and road data from private companies. However, the timeliness of the National Geographic Database varies from region to region depending on the source data. Table 4.1 provides details on the distribution of features by source.
Table 4.1 Feature counts and summed length values by data source
Described below are the steps taken to incorporate data from various sources into the National Geographic Database.
The 1996 Street Network File (SNF)
In census metropolitan areas and larger census agglomerations, the 1996 Street Network Files (SNFs) from
Statistics Canada were the primary data source. The SNFs were created from various source maps
at different scales, and maintained by the Statistics Canada Geography Division over more than twenty-five
years. They contained road names, address ranges and a rich set of road arcs. The maps used to
build and maintain the SNF had various scales and different vintages. Therefore, the quality
of its geometry varies from place to place in terms of absolute positional accuracy.
The SNF information was updated, enhanced, and incorporated into the National Geographic Database during a build phase. Features that were not roads were removed. Streets were geometrically adjusted (i.e., rubber sheeted) to match the superior positional precision of the National Topographic Database. The format of address ranges was not changed, except the type of the values was changed from character to numeric. In most cases, road names in all upper case letters were converted to names in upper and lower case. The length of the field that contains the road names was also increased to accept full names instead of abbreviations.
National Topographic Database (NTDB)
The National Topographic Database (NTDB), produced by Natural Resources Canada, has a stable and
precise geometry and a standardized road classification scheme. In the more densely populated parts
of Canada, its scale is 1:50,000, while in the more northern and sparsely populated areas the scale
is 1:250,000. Unlike the SNF, the NTDB contains no civic address range or street name information. The
NTDB served as the source
of the road network for most of southern Canada, outside of census metropolitan areas and large
census agglomerations that were covered by the SNF and Elections Canada (EC) data.
The NTDB geometry is the adopted standard for the National Geographic Database: that is, all spatial data used in the creation of the National Geographic Database were vertically adjusted (rubber sheeted) and edge matched to approach the largest scale NTDB geometry.
Digital Chart of the World
The Digital Chart of the World (DCW) is a 1:1,000,000 scale digital map, built primarily for aeronautical
charts. It was used in the database to add road geometry to the sparsely populated portion of Canada,
mainly in the north. It does not contain road names or address ranges.
Elections Canada Geographic Database
In 1993, Elections Canada (EC) started to compile the EC
Geographic Database, using data from the Street Network Files, National Topographic Database and Digital Chart of the World.
Paper maps were created for areas not covered by the Street Network Files and were distributed to the
Elections Canada returning officers, who added the road names with information from the field.
Elections Canada updated the road network with new roads and added the road names but not address
ranges.
Due to the addition of new roads, the resulting geometry does not always match the initial National Topographic Database geometry. Wherever more recent Elections Canada data would improve the quality and quantity of road information, it was added to the Statistics Canada Street Network Files to form the National Geographic Database. The content derived from Elections Canada is primarily new roads and road names. These were left in the format used by Elections Canada, with upper and lower case letters, accents, road type and direction, but no civic address ranges.
Other sources
In addition to digital maps from other federal, provincial, municipal and licensed private sources,
portions of the National Geographic Database contain information from Elections Canada Returning
Officers (maps for the 38th General Election, held in the summer of 2004), Statistics
Canada Regional Offices (1996 Enumeration Area Collection maps), data from Statistics Canada's 2001
Census, and other materials prepared by private companies.
Positional accuracy
Refers to the absolute and relative accuracy of the positions of geographic features. Absolute accuracy is the closeness of the coordinate values in a dataset to values accepted as or being true. Relative accuracy is the closeness of the relative positions of features to their respective relative positions accepted as or being true. Descriptions of positional accuracy include the quality of the final file or product after all transformations.
Absolute positional accuracy
Absolute positional accuracy describes the degree to which the position of features in a geographic database reflects their true position on the ground (i.e., the closeness of reported coordinate values to values accepted as true).
The information present in the National Geographic Database (NGD) road layer is provided for the purposes of statistical analysis, census operations and electoral operations only. The absolute position of roads on the NGD varies with the source files and documents used to build and maintain the database. Therefore, the NGD is not suitable for high precision measurement applications such as engineering, property transfers, or other uses that might require highly accurate measurements of the earth's surface.
During the build phase, the road layer was rubber sheeted to match the position of those on the NTDB (which was used for reference purposes). After the build, the rubber sheeting process was applied to the materials used to maintain and improve the content of the road network. It is therefore expected that these geometrically matched arcs will have a positional accuracy similar to the corresponding reference data used during development of the database. It should be noted that the reference source selected for different geographic areas depended on a variety of factors such as population size, geographic location (urban or rural) and the availability of NTDB/Digital Chart of the World data. For example, in major urban centres 1:50,000 NTDB data were generally used as the reference data. As a result, in these areas, roads that were geometrically matched have a positional accuracy similar to roads on the NTDB data (i.e., approx. ±10 metres). In areas that used 1:250,000 NTDB and Digital Chart of the World reference data, the positional accuracy of roads that were geometrically matched is approximately ±300 metres (NTDB) and between ±2,100 and ±4,300 metres (Digital Chart of the World), respectively.
The positional accuracy of arcs that could not be matched because they were not present in the reference data is not measured. These arcs were digitized on screen from paper maps annotated by Elections Canada's returning officers and Statistics Canada's regional officers. Although accurate in their attribute information and their relative position in relation to other features, the absolute positional accuracy of these roads is unknown.
Absolute positional accuracy is not a requirement for electoral and census processes. An indication of the positional accuracy of features in the NGD can be assessed from the source of the geometry of road features in Table 4.1.
Relative positional accuracy
Relative positional accuracy describes the degree to which the position of features in a geographic database reflects their true ground relationships.
For the National Geographic Database, relative positional accuracy is important. A road must appear in the proper position relative to other roads and physical features. For example, a road should not fall on water when the road layer is overlaid on hydrographical layers, unless it is a road classified as a bridge.
During the build phase, the dataset was thoroughly tested for relative positional accuracy. The road network was overlaid onto the hydrographical, power line and railroad layers.
Attribute accuracy
Attribute accuracy refers to the accuracy of quantitative attributes and the correctness of non-quantitative attributes. Two road attributes were tested for accuracy: road name (name) and road address range ('addr_fm_left', 'addr_to_left', 'addr_fm_rght', 'addr_to_rght'). Road address range considers the completeness of addressing on individual arcs.
Road name
During the build phase, every effort was made to insure a proper transfer and association of a specific attribute (i.e., name, type, direction, and address range) to a specific geometric feature. This includes the association as well as its accuracy.
Measures on number of road name and address range attributes have been presented in Table 4.2.
Road address range
Two tests were conducted to determine the attribute accuracy of address features on the base. First, the results from the current version of the NGD were compared to the previous version of the NGD in order to identify any increases or decreases in the number of addressable roads and blockfaces. Secondly, a check was run on the 2001 addresses to determine which 2006 geographic area they fell into, then those same addresses were compared to the 2006 geographic area derived using a 2001/2006 correspondence file.
Error processing SSI file
Logical consistency
Logical consistency refers to the fidelity of relationships between all variables in a dataset. For example, a road arc that does not have a road name should not have a road type.
During the build phase, the National Geographic Database (NGD) dataset was thoroughly tested for logical consistency. Any violations of logical consistency were corrected, and 100% of the data are logically consistent.
Node-line-area relationships satisfy topological requirements as specified in the ArcInfo® data model.
Error processing SSI file
Consistency with other products
The position of the arcs in the 2006 Road Network File are generally consistent with the 2005 Road Network File and 2006 Cartographic and Digital Boundary Files, but are not necessarily consistent with those of the 2001 Cartographic Boundary Files and the 2001 Road Network and Skeletal Road Network Files.
Error processing SSI file
Completeness
Completeness refers to the presence or absence of features, their attributes and relationships. Many new road features that were not previously found on earlier digital files at Elections Canada and Statistics Canada have been added to the National Geographic Database in order to create a more complete National Geographic Database road layer for all of Canada.
Roads
Many features not found in previous road network file products were added to the 2006 Road Network File in order to improve our nation-wide road coverage. Table 4.2 shows the number of road features on the 2006 Road Network File.
Table 4.2 Number of road features in the 2006 Road Network File
