Government Home > Community Counts > About Us

Government Home
Finance Home
Stats Home

Community Counts:

Help

Technical Notes



Census of Population Coverage
About Communities
Available Geographies
Data Preparation
Data Development

Census of Population Coverage

The objective of every census is to provide detailed information at a single point in time on the demographic, social and economic conditions of the population. One of its goals is to enumerate the entire population on Census Day.

Inevitably, however, some people are not counted, either because their household did not receive a census questionnaire (for example, a household living in a separate apartment in a house) or because they were not included in the questionnaire completed for the household (for example, a boarder or lodger).

Some people may also be missed because they have no usual residence and did not spend census night in any dwelling. On the other hand, a small number of people may also be counted twice (for example, a student living away from home).

To determine how many individuals were missed or counted more than once, Statistics Canada conducts post-censal coverage studies of a representative sample of individuals.

The results of these studies provide information used to adjust the census counts for the purpose of producing current (quarterly and annual) population estimates which take into account net under-enumeration in the census. Net under-enumeration is the estimated number of people missed in the Census less the number of people counted that should not have been included.

In 2001, after adjustment for net under-enumeration, the population estimate for Nova Scotia was 2.7% higher than the population enumerated in the census. This compares with 3.3% for Canada . While the Census numbers show a population of 908,005, the actual estimate of Nova Scotia population was 932,389 as of June 1, 2001 .

In 1996, after adjustment for net under-enumeration, the population estimate for Nova Scotia was 2.4% higher than the population enumerated in the census. This compares with 2.5% for Canada . While the Census numbers show a population of 909,280, the actual estimate of Nova Scotia population was 930,652 as of June 1, 1996 .

In 1991, after adjustment for net under-enumeration, the population estimate for Nova Scotia was 1.6% higher than the population enumerated in the census. This compares with 2.6% for Canada . While the Census numbers show a population of 899,945, the actual estimate of Nova Scotia population was 914,275 as of June 1, 1991 .

For the more detailed data such as education, income and labour, the data is provided through the Census 'long-form", which is completed by 20% (one out of five people) of the population.. This information is not collected for people living in institutions, such as senior citizen homes and correctional facilities. As a result, detailed data on seniors excludes over 10% of the population over 75. In addition, the net under-enumeration rate seems to be slightly higher for persons asked to complete the long form than for those who completed the short form. The reported Census figures use the information from the 'short form' to estimate the characteristics for the full population.

Back to top of page



About Communities

"Community" can mean many different things to different people. Communities can be defined in different ways for a variety of purposes, with various results. A community of individuals can associate themselves with others having common interests; people can feel an emotional or other attachment to a community location; governments can define communities by specific geographic boundaries.

The Nova Scotia Community Counts (NSCC) system is built upon the recognition that definitions of community are dynamic, and that the implications of differing community definitions are wide-reaching. In the interests of improving access to comparable, timely, and meaningful indicators and statistics, the following structure and guidelines are followed in development and presentation of statistical information.

The Community Counts system currently provides most data for the province according to four views, each broken down into five standard geographic divisions:

Health View

• Canada/provinces/territories (14)
• District Health Authorities (DHAs - 9)
• Counties (18)
• Community Health Boards (CHBs - 37)
• Communities (276)

Electoral View

• Canada/provinces/territories (14)
• Federal Electoral Districts (FEDs - 11)
• Counties (18)
• Provincial Electoral Districts (PEDs - 52)
• Communities (276)

Municipal View

• Canada/provinces/territories (14)
• Regional Development Authorities (RDAs - 13)
• Counties (18)
• Municipalities (MUNs - 55)
• Communities (276)

Justice View

• Canada/provinces/territories (14)
• Justice Centres (11)
• Counties (18)
• Justice Police Detachments (50)
• Communities (276)

The following guidelines have been used in developing the boundaries for the 276 communities within the Community Counts system:

• most communities have a population of greater than 1,000 to minimize data suppression
• communities include 100% coverage of the provincial population, i.e. all areas of province are included
• natural clusters of population, rather than existing legal boundaries for towns and counties, are used to define communities
• each populated Indian Reserve maintains its existing boundaries as a community (these account for most communities with population less than 1,000). Statistics Canada suppresses data for any geographic area with less than 40 persons and randomly rounds individual cells, which affects mainly the smallest reserves.  Because this random rounding distorts year over year comparisons, Community Counts suppresses detailed data for those communities with population less than 100.
• the dominant or major community name within the geographic area is used to represent that area
• The Nova Scotia Civic Addressing File (NSCAF) General Service Areas (GSAs) are used as basic building blocks. There are over 2,200 of these and they are combined to form the 276 communities represented in Community Counts. In a few instances, the GSAs are divided or other geographic boundaries (e.g. provincial electoral districts) are used to define communities.

Census geographies (i.e. census blocks, dissemination areas, census tracts, and census sub-divisions) were used in earlier drafts of community definitions, but were found to have too many limitations and idiosyncrasies (e.g. blocks several kilometres long in many coastal areas). In particular, Census blocks in rural areas are large and bounded by roads which often resulted in divided communities. Census geographies have been defined, in many cases, without the use of water features. For Nova Scotia , this presents a major limitation to using Statistics Canada boundaries as building blocks.

Back to top of page



Available Geographies

The following geographies are available within the Community Counts:

Communities (Listed by County)

Community Health Boards (Listed by District Health Authority)

Federal Electoral Districts

Justice Centres

Justice Police Detachments

Municipalities (Listed by County)

Provincial Electoral Districts

Provinces and Territories

Regional Developmental Authorities

Back to top of page

Data Preparation

There are many different ways in which data are gathered and organized by a wide variety of government departments, agencies or others within the province (e.g. Statistics Canada divisions for the Census and postal codes for Taxfiler data; Nova Scotia Department of Health District Health Authorities and Community Health Boards; federal, provincial and municipal electoral districts; school boards and districts; municipalities and populated place names, among others). In order to provide a common representation of these data sources, input data from this wide variety of sources must be standardized.

As indicated in our About Communities section, there are several geographic categories for presentation and summarization of data. Rationalization of these geographies has been, and will continue to be, a major challenge to providing comparable data from a variety of sources over time. It is difficult to get comparable data for any given geographic division used by Statistics Canada across various Census periods. Taxfiler data, which is postal code-based, is developed on a totally different geographic basis than the Census, and also changes from year to year.

For these reasons, the Community Counts system uses a data modeling approach that takes input data from any given set of geographic categories and models the output data for each of the geographic divisions within the Community Counts system. Each input data set will have methods and approaches developed appropriate to the data in order to present information as accurately as possible for the defined geographic categories. Each of the data preparation steps, and especially the resulting output data, is subject to a series of quality assurance tests. The following describes the three step approach used for Census of Population data:

Step 1: Development of Concordance File

For each geographic division used in Community Counts, a concordance file is generated. The concordance file uses the available levels of core data (i.e. census division (CD), census-subdivision (CSD), census tract (CT) or dissemination area (DA)) and pro-rates it to the corresponding community/CHB/PED/DHA. A separate file is created for each of the output geographic divisions. The pro-rated share is initially based on the number of residences in the input geographic division attributable to the output (NSCC) geographic division. The shares are rounded to the nearest five percentage points.

In order to minimize data suppression, the highest level of data that fits the output geographic division is used. For example, many Community Health Boards have the same boundaries as CSDs; therefore the CSD information, rather than the lower level DAs or CTs, is used to model data for the CHB. A further adjustment is made to balance between the precision of pro-rating DAs versus the effort to avoid data suppression through the use of CT data. For example, if one of eight DAs within a CT is split between two communities, with most of the other 7 DAs in the first community and only a small percentage of the split DA in the second community, then CT level data, rather than DA level data, would be pro-rated to model the population for each of the two communities.

To following links demonstrate the transformation of the Census data for each available Census year (1991, 1996, 2001, and 2006).

2006

2001

1996

1991

Step 2: Data Preparation

Using the Census standard data products as much as possible, data is extracted for each of the available geographic layers in a standard format for each data set needed. This is repeated for each Census year. Changes in both data content and data summarization categories present extra challenges, e.g. education attainment is reported for ages 15 and over in 1991 and 1996, but for ages 20 and over for 2001.

Step 3: Data Modeling

An Access database routine has been developed and is run for each data set using the above files as input. This generates the required data for the Community Counts system.

Back to top of page

Data Development

Currently, the Community Counts system has most available standard Census of Population data fully developed. Data preparation of input Taxfiler data is ongoing. In addition to these Statistics Canada data sets, Community Counts staff members are working with provincial government departments and other agencies to further develop our database. Within the next year, it is expected that the Community Counts system will include selected administrative data from various Nova Scotia Departments.