Data Sources and Technical Details for Regional Councils

General Information

Changes in the data from 2008

The majority of indicators have been updated in this year’s report. No new indicators have been added but we have revised the indicator for suicide and we are now using census data for the cigarette smoking indicator, due to the loss of the data source for cigarette smoking used in previous reports.

Small numbers

Working with regional data presents a range of problems, not least of which is small numbers. This problem is particularly acute when survey data is being used; although census and administrative data, when broken down by age and ethnicity, can also present challenges.

We have assessed the indicator data and, where appropriate, collapsed smaller regions into larger areas to produce more robust results. This has most commonly affected Tasman, Nelson, Marlborough and Gisborne, all of which are unitary authorities (note these areas are covered separately in the territorial authority data sets).

Even given the above process, many of the indicators presented in this report need to be interpreted with caution because of the relatively small number of people at a subnational level. Extra care should be taken with ethnic breakdowns in some regions where groups such as Pacific peoples and Asians have a very low representation.

Census data is also randomly rounded to protect confidentiality. Data may be randomly rounded up or down so tables may not add exactly to the total. For numbers below 30, random rounding may result in unusual percentages. For some data such as Household Labour Force Survey data, school leaver data and census data small numbers may be suppressed (this is indicated by the inclusion of "..s" within tables). Statistics New Zealand’s confidentiality procedures for the 2006 Census have resulted in the suppression of some tables that conformed when data for 2001 and before was originally compiled.

Ethnicity data

Ethnicity data is provided for a number of the indicators but it should be used with caution because definitions for ethnicity vary across data sources and over time. These issues are discussed in the report Statistical Standard for Ethnicity 2005, which can be found on the Statistics New Zealand website.

For more recent censuses, an individual can choose to belong to one or more ethnic groups. In this report, census data counts a person in each ethnic group they have said they belong to. As a result, ethnicity data may add to more than 100 per cent. The exception is the 1986 Census data, where multiple ethnicities were not captured.

There was a change in the census ethnicity question between 1991 and 1996 resulting in a potential for inconsistent results. Any comparison of 1991, 2001 and 2006 ethnicity data with 1996 and 1986 data should be made with caution.

In the 2006 Census, those who gave their ethnicity as "New Zealander" were reported separately. Previously they were included with "New Zealand European". Because of the difficulty in changing historic data and to ensure consistent time series data, this report includes New Zealander with New Zealand European. The "Other" ethnic category is also consistent with pre-2006 definitions.

Other data sources, such as the Household Labour Force Survey and administrative data on school leavers, only count an individual in one ethnic group.

People Data

Ethnicity by age, proportion in the area and sex

Data notes: These figures are based on the usually resident population from the 2006 Census. The numbers exclude overseas visitors temporarily in New Zealand on census night and New Zealand residents who are temporarily overseas on census night. Residents who are away from their usual address on census night are allocated back to the area where they usually live and form part of the census "usually resident" population count of that area. The total figure provided is for all people, including those who did not specify an ethnicity. Ethnic percentages are derived from only those who specified an ethnicity.

Data source: Statistics New Zealand.

Estimated population change, births, deaths and natural increase from 1996

This is an estimate of all the people who usually live in an area of New Zealand at a given date. This estimate is based on the census usually resident population count, which excludes visitors from overseas, and is adjusted to include:

• residents who are temporarily overseas on census night
• residents who are missed or counted more than once by the census (net census undercount)
• births, deaths and net migration (arrivals less departures) of residents during the period between census night and the given date.

Data notes: Estimates are for the year ended June. Estimates are rebased after each census, which can cause discontinuity in the time trend.

Statistics on births and deaths record the number of births and deaths registered in New Zealand each year. Natural increase is the number of registered births minus the number of registered deaths.

Data source: Statistics New Zealand.

Projected population change

Population projections are estimates of the size of the population at a future date. A number of alternative series are produced in each set of projections. These use different combinations of assumptions about future fertility, mortality, and net migration.

Data notes: The projections have been updated in this year’s report. The base for this projection series was the estimated resident population at June 2006. This estimate uses March 2006 Census results adjusted for undercount and estimated population change to June 2006.

Data source: Statistics New Zealand.

Household composition

This table shows household composition by family type.

Data source: Statistics New Zealand, census results.

Health

Life expectancy

The number of years a hypothetical newborn male or female could expect to live if they were subject throughout their lives to the age-specific mortality patterns prevailing over a three-year period centred on their birth year.

Data source: Statistics New Zealand, New Zealand Life Tables 2005-2007.

Cigarette smoking

The proportion of the population aged 15 years and over, in the area, who currently smoke cigarettes.

Data notes: Earlier reports provided data from surveys conducted by ACNielsen Ltd. However this data is no longer collected. The only alternative source for subnational data is the census. In this report, we provide data from the 1981, 1996 and 2006 censuses. It should be noted that while the question on smoking asked in the census has changed over time it remains relatively consistent and can be used for time series comparisons. The use of the census has allowed us to provide more information including age and ethnic breakdowns.

The data is for the population aged 15 years and over as the census captures information on all ages 15 years and over (data used in the national social report is for 15–64 year olds as the survey used covers only this age group). While unadjusted age-specific smoking prevalence rates are provided in one table, the remaining tables provide age-standardised rates to adjust for different age structures across subnational areas (the World Health Organisation standard population has been used).

Data source: Statistics New Zealand, Ministry of Social Development.

Suicide

The number of suicide deaths per 100,000 population, expressed as an age-standardised rate, for the population aged 5 years and over.

Data notes: Classification of a death as suicide is subject to a coroner’s inquiry, and only on the completion of an inquest can a death be officially classified as suicide. In some cases, the inquest will be heard several years after the death, particularly if there are a number of factors surrounding the death that need to be investigated first. Consequently, a provisional classification may be made for a suicide before a coroner’s verdict is released. This means more recent data remains provisional. The International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Australian Modification (ICD-10-AM) codes used for mortality are X60–X84, Intentional self-harm.

Data was rebased to match regional council boundaries using data coded to meshblocks. Some regions have been collapsed into larger regions to improve data quality. Data for the 2009 report has been reworked and no longer provides time series information. Because of the small numbers involved, the data is now presented as an average for 2002–2006. This approach means we can provide separate data for males and females. Previously published numbers and rates will differ from those included in this report. The data is for the population aged 5 years and over. The data has been age standardised to take into account differences in the age structures of regional populations (the World Health Organisation standard population has been used).

Data source: Ministry of Health, customised data.

Hospitalisations for intentional self-harm

The number of hospitalisations for intentional self-harm per 100,000 population, expressed as an age-standardised rate for the population aged 5 years and over.

Data notes: Hospitalisation for intentional self-harm is a recognised proxy measure for attempted suicide, which measures the number of people who intentionally harmed themselves and were admitted to hospital. Along with suicide deaths the measure is an indicator of mental health. The motivation for intentional self-harm varies. It includes people with the desire to end their lives but also others where suicide is not the main intention. The International Statistical Classification of Diseases and Related Health Problems, 10th Revision, Australian Modification (ICD-10-AM) codes used are X60–X84, Intentional self-harm.

The hospitalisation data used has been filtered to remove patients who were admitted via the emergency department and who stayed less than two days in hospital. This allows better comparisons between areas. Because of the small numbers involved, data for the 2009 report has been reworked and no longer provides time series information. Data is now presented as an average for the 2002–2006 period. This approach means we can provide separate data for males and females. This change and the change in the data-filtering methods mean the numbers will differ from previous reports. The data is for the population aged 5 years and over.

Data source: Ministry of Health, customised data.

Knowledge and Skills

Participation rate 3 and 4 year olds

The number of children aged 3 and 4 years enrolled in early childhood education (ECE) programmes as a proportion of the estimated population aged 3 and 4 years. ECE programmes include licensed ECE services (kindergartens, playcentres, education and care services, home-based services, casual education and care services, and te kōhanga reo) and licence-exempt ECE services. The data excludes the correspondence school as it cannot be located in a region.

Data notes: Rates of participation are only "apparent" because children may be enrolled in more than one ECE centre. This can result in participation rates exceeding 100 per cent. The measure does not provide information on the length of the participation or on the quality of the programmes, both of which are relevant to positive educational outcomes. Ethnic data is not provided because of difficulties in getting robust ethnic population estimates at a regional level. In the 2009 report the data has been rebased from 2001 which means the data will not be comparable to previous reports.

Data source: Ministry of Education, customised data.

Early childhood education attendance by Year 1 students

This is an alternative measure of early childhood education based on the indicated experience of entry-level Year 1 school children.

Data notes: The measure does not provide information on the length of participation or on the quality of the programmes, both of which are relevant to positive educational outcomes. Unlike participation rates, ethnic data can be presented.

Data source: Ministry of Education, customised data.

School leavers with higher qualifications

The proportion of school leavers who had attained a qualification at National Certificate of Educational Achievement (NCEA) Level 2 or above, and the proportion who leave with Bursary and above.

Data notes: This data is based on the results from schools rather than on the residential location of students. Thus, results are coded to the region in which the school is located. NCEA Level 2 is achieved by having 60 credits at NCEA Level 2 or above, plus 20 credits at any level even if gained for any other National Qualification. There are no literacy or numeracy requirements for this qualification. This level of qualification is higher than that presented in previous reports where Sixth Form Certificate or NCEA Level 2 with at least 14 credits (30 credits in 2004 and 2005) was used. The change has resulted in a discontinuity in the data series so only 2005–2007 data is presented in this report.

Data source: Ministry of Education, customised data.

Educational attainment of the adult population

The proportion of 25–64 year olds with at least upper secondary school education, defined in the International Standard Classification of Education (ISCED 97) as Level 3 and above. This includes any qualification including and above School Certificate in one or more subjects.

Tertiary is based on ISCED Level 5A/6 and above and includes a post-graduate degree, certificate or diploma, and a bachelor's degree.

Data notes: Because of small sample sizes, some regions have been aggregated to provide robust data. Even given this, the data is subject to sampling error and should be used with caution. Figures below 1,000 have been suppressed. This has most often occurred for ethnic data where there is a low representation of some ethnic groups within a region. Data is an annual average percentage based on the December year. Data from 1991 has been revised for this report.

Data source: Statistics New Zealand, Household Labour Force Survey, customised data.

Paid Work

Unemployment

The number of people aged 15 years and over who are not employed and who are actively seeking work and who are available for paid work, expressed as a percentage of the total labour force. The total labour force is the population aged 15 years and over who are either employed or unemployed. The unemployed are defined in the Household Labour Force Survey as those who are without a paid job (or unpaid work in a relative's business) and who have actively sought work in the four weeks before the survey and who are available to take up work. "Actively seeking" includes doing any actions such as contacting an employer, asking friends and relatives and contacting an employment agency or Work and Income, but excludes only checking newspaper advertisements. The employed are those who worked for pay or profit for one hour or more in the week before the survey or who worked unpaid in a relative's business or who have a job but did not work that week because of leave, sickness or industrial disputes.

Data notes: The data is based on a sample survey and is, therefore, subject to sampling error. Because of small sample size, some regions have been aggregated to provide robust data. The definition of unemployed excludes some people who regard themselves as unemployed, including the "discouraged unemployed" – those who do not meet the "actively seeking work" criterion. This group is classified in the "not in the labour force" category. The unemployment rate also excludes those who have part-time employment but who are seeking to work more hours. Data is an annual average percentage based on the December year. Figures below 1,000 have been suppressed. Data from 1991 has been revised for this report.

Data source: Statistics New Zealand, Household Labour Force Survey, customised data.

Employment

The proportion of the population aged 15–64 years who are in paid employment for at least one hour per week. See Unemployment, above, for a definition of employed. The definition used here is consistent with the definition in the social report and relates to the population aged 15–64 years, rather than to those aged 15 years and over. This is because, if 15 years and over were used, results would be skewed by differences in the proportions of the sub-populations over 65 years, particularly when comparing males with females and comparing different ethnic groups.

Data notes: The data is based on a sample survey and is, therefore, subject to sampling error. Because of small sample size, some regions have been aggregated to provide robust data. The definition of employment includes those working one hour or more per week, so it will include some people who are likely to regard their status as closer to unemployment than to employment. For example, people who are on the unemployment benefit and who are searching for work but who work a few hours a week will be counted as employed. Data is an annual average percentage based on the December year. Figures below 1,000 have been suppressed. Data from 1991 has been revised for this report.

Data source: Statistics New Zealand, Household Labour Force Survey, customised data.

Median hourly earnings

Real median hourly earnings from all wages and salaries for employees earning income from wage and salary jobs.

Data notes: The data is based on a sample survey and is, therefore, subject to sampling error. Because of small sample size, some regions have been aggregated to provide robust data. Hourly earnings relate to the number of hours usually worked and the usual income rather than the number of hours actually worked and the actual income. Proxy interviewing may be used to collect data on income under certain circumstances. Past data has been inflation indexed to reflect June 2008 dollars. The New Zealand Income Survey is carried out in the June quarter.

Data source: Statistics New Zealand, New Zealand Income Survey, customised data.

Workplace injury claims

The number of workplace accident insurance claims reported to the Accident Compensation Corporation (ACC) per 1,000 full-time equivalent employees, excluding those who received accident and emergency treatment only (one part-time employee equals 0.5 full-time employee). Full-time equivalent employee data is as estimated by Statistics New Zealand's Household Labour Force Survey.

Data notes: Data was collected by territorial authority. The regional estimates were calculated by Statistics New Zealand by summing-up the raw data for each territorial authority. The data does not include workplace accidents where no claim was made to ACC. In some cases, there are delays between when the accident happened and when the claim was reported to ACC.

Data source: Statistics New Zealand.

Economic Standard of Living

Population with low incomes

The proportion of people living in households with real gross income less than 60 per cent of the median household equivalised national gross income benchmarked at 2001. This allows for an assessment of the number of people living in households with low incomes both at the benchmark period (2001) and in earlier or later periods.

Data notes: Households were used for this exercise rather than families because they are consistently defined over the census periods provided. Household incomes were equivalised using the Revised Jensen Household Equivalence Scale. This makes allowance for the impact of household size and structure on income. Incomes for the 1986, 1991, 1996 and 2006 censuses were inflation-adjusted to provide real gross income, so accurate comparisons can be made between each census. The non-response to the census question on which the table is based was quite high, particularly for some subgroups such as Pacific families. This non-response may skew the results.

Data source: Statistics New Zealand, customised census data.

Household crowding

The proportion of the population living in "crowded housing" with fewer bedrooms than required given household size and composition, measured against the Canadian National Occupancy Standard.

Data notes: The Canadian National Occupancy Standard states that:

• no more than two people shall share a bedroom
• parents or couples may share a bedroom
• children under 5 years, either of the same sex or of the opposite sex, may share a bedroom
• children under 18 years of the same sex may share a bedroom
• a child aged 5–17 years should not share a bedroom with a child under 5 years of the opposite sex
• single adults 18 years and over and any unpaired children require a separate bedroom.

The Canadian Crowding Index is not an objective index of crowding. The extent to which household members see themselves as living in crowded circumstances depends on many factors, including social and cultural expectations. Furthermore, it cannot be assumed households requiring one or more additional bedrooms (based on the Canadian index) suffer negative social outcomes.

The Canadian Crowding Index is used here as it is sensitive to both household size and household composition. The measure sets a bedroom requirement for households based on precise criteria. It is useful for ascertaining crowding levels and for identifying the extent of bedroom under-utilisation. Data for 1986–2001 has been revised and will differ slightly from earlier publications.

Data source: Statistics New Zealand, customised data.

Civil and Political Rights

Voter turnout - local authority elections

The proportion of all enrolled electors (both resident and ratepayer) who cast a vote in local authority elections for regional councils.

Data notes: To be eligible to vote, a person must be at least 18 years old and meet residential and certain other criteria. Data is for contested elections (where positions are not contested no election is held). Data in this report has been updated to include the 2007 local authority elections. Information on the four unitary authorities is based on territorial authority election results.

Data source: Department of Internal Affairs, customised data.

Representation of women in local government

The proportion of female candidates and elected members of regional councils in local authority elections.

Data notes: Data in this report has been updated to include the 2007 local authority elections, and additional data has been presented to provide a time series from 2001. Information on the four unitary authorities is based on territorial authority election results.

Data source: Department of Internal Affairs, customised data.

Cultural Identity

Māori language speakers

The number of Māori who reported in the census they could hold a conversation about everyday things in Māori, as a proportion of the Māori population, and the total number of Māori speakers as a proportion of the total population.

Data notes: The data relies on self-reporting rather than measuring the actual level of fluency in the population.

Data source: Statistics New Zealand, customised data.

Language retention

The proportion of people who can speak the "first language" (excluding English) of their ethnic group, for ethnic groups (other than Māori) with an established resident population in New Zealand, as recorded in the 2001 and 2006 censuses. The ability to speak a language is defined as being able to hold an everyday conversation in that language. "First language" refers to the indigenous language associated with a given ethnicity rather than the first language of an individual.

Several criteria were used to identify ethnic groups with an established resident population in New Zealand. These included the total population size, the number of years since the group’s arrival in New Zealand and the age distribution and birthplace (overseas and within New Zealand) of group members. These variables provide a measure of the influence of time and of the demographic characteristics of the groups. Each variable was applied independently to a large list of ethnic groups from which 15 were selected under the broad categories of Pacific peoples, Asian and European. To be selected, a group needed to have: a New Zealand resident population of over 2,000 people; a broad age distribution to investigate the impact of age on language retention; and sufficient numbers born in New Zealand to make meaningful comparisons with overseas-born residents.

Data notes: While a direct link can usually be made between a language and an ethnic group, this is not always the case. Some ethnicities are associated with several languages and one language can span several ethnicities. While English is an official language of some groups selected in these tables, the census does not distinguish between different varieties of the English language. English has, therefore, been excluded as a first language within these tables.

Because the census variables for ethnic group and language spoken allow more than one response, there may be some individuals who appear in more than one ethnic category, and as such are double counted.

Data from the 2001 Census has been revised for this report and the 2006 Census data has been added. Revision of the 2001 data was required to ensure consistency across census periods and has resulted in a slight variation from the numbers presented in previous reports.

Leisure and Recreation

Participation in physical activity

The proportion of the population aged 16 years and over who met physical activity guidelines (ie were physically active for at least 30 minutes a day on five or more days over the last week), as measured by the 2007/2008 Active New Zealand Survey.

Data notes: The measure used for regional councils is the 2007/2008 Active New Zealand Survey whereas the national social report uses the New Zealand Health Surveys. Both surveys produce commensurate results. This report updates the 2008 report which used 2001 data. Data for the regions is based on 17 Regional Sports Trust areas which align closely with regional council areas. Some regions have been aggregated to provide more robust data. Even so, the information is based on survey data and because of small sample sizes at the regional level should be used with caution.

Data source: Sport and Recreation New Zealand’s 2007/2008 Active New Zealand Survey.

Physical Environment

Drinking water quality

The percentage of the estimated resident population who receive their water from community water supplies whose drinking water complies with either the 2000 or 2005 Drinking-water standards for New Zealand relating to E. coli and Cryptosporidium.

Data notes:The microbiological health risk for drinking water is assessed using compliance criteria based on the two microbiological reference organisms, Escherichia coli (E. coli) and Cryptosporidium. Compliance is based on the Drinking-water standards for New Zealand (DWSNZ). At this time there is a transition between the DWSNZ:2000 and the DWSNZ:2005. This transition is scheduled to take several years to complete and drinking-water suppliers may elect which of these standards to operate under. Compliance is assessed against the standard the supplier has chosen to comply with at this time. This approach is in line with the Health (Drinking Water) Amendment Act 2007. Section 14(3) of this Act allows suppliers to opt to comply with either the 2000 or 2005 standards, to ease the transition for those suppliers who are more comfortable with the 2000 standards. Compliance is measured at the treatment plant for Cryptosporidium and in the reticulation area for E. coli.

For the Cryptosporidium measure the approach followed for the social report differs from that used by the Ministry of Health (MoH). The MoH measure is an estimate based on all plants supplying each distribution zone. The measure used in the social report is based on the worst result from the various plants supplying a distribution zone. The former approach double counts populations where a distribution zone is supplied by multiple plants, the latter approach avoids this by measuring only one plant. The approach followed in the social report is to become the standard measure when the transition to DWSNZ:2005 is completed.

In the 2008 report the measurement of compliance moved from a calendar year to the fiscal year. For this reason the data points jump from the 2005 calendar year to the 2006/2007 fiscal year. This change, combined with the transition in standards, will result in some lack of data continuity across these periods.

Data is rebased to match regional council boundaries using data coded to territorial authority. A best-fit approach is followed because some territorial authorities have boundaries that span more than one region.

The data is derived from a survey which lists all water supplies and the number of people they service. Some people have access to more than one water supply (eg a second home in another town) and therefore may be counted twice. This can result in a rate greater than 100 per cent.

Safety

Recorded criminal offences

The total number of recorded criminal offences and the recorded criminal offence rate per 10,000 population.

Data notes:Before a crime can be recorded the matter needs to come to the attention of the police. Research indicates that many crimes are never reported to the police. Crimes most likely to be reported include those that involve insurance claims and those where injuries require medical treatment. All reports of incidents, whether from victims, witnesses or third parties, or incidents discovered by the police, and whether crime-related or not, result in a police officer registering an incident report (more information can be found on the Statistics New Zealand website). This data excludes traffic offences. It should be noted that changes in administrative practices, legal requirements and public awareness of crime can influence time series trends. Therefore, caution should be used when interpreting this data.

This indicator is more valid when used to compare the level of crime between regions, and less valid when used to measure trends in crime over time or the absolute level of criminal activity.

In this report, recorded offences have been aggregated to regional council areas based on information recorded at police station level. A best-fit approach is used as some areas covered by police stations fall into more than one regional council area. Data for Tasman and Nelson has been combined because police station data cannot be split into separate regions. The data is for the calendar year.

Data source: New Zealand Police, customised data.

Road casualties - deaths and injuries

The number of people killed or injured in motor vehicle crashes per 100,000 of the population. Pedestrians or cyclists killed or injured by motor vehicles are included.

Data notes: The New Zealand Transport Agency derives its data from two main sources: injury data from the traffic crash reports completed by police officers who attend fatal and injury crashes; and mortality and hospitalisation data from the New Zealand Health Information Service. Police policies regarding traffic crash reports can influence the reported level of incidences over time.

Social Connectedness

Telephone and internet access in the home - telephone access

The proportion of the population with telephone access in the home.

Data notes: Data for the 2009 report has been revised and will differ from previous reports. In past reports percentages were worked out using all households including those where telephone access was not specified. For this report the not-specified group has been excluded.

Telephone and internet access in the home - internet access

The proportion of the population with internet access in the home.

Data notes: This question was first asked in the 2001 Census. Data for the 2009 report has been revised and will differ from previous reports. In past reports percentages were worked out using all households including those where internet access was not specified. For this report the not-specified group has been excluded.

Data source: Statistics New Zealand, customised census data.

Contact between young people and their parents

The percentage of secondary school students (in Years 9–14, aged 18 years or younger) reporting that most weeks they spent enough time with their parents (or someone who acts as a parent).

Data notes: The data is based on 9,107 secondary school students and was collected in 2007/2008. The data excludes young people who are no longer at school. The Youth’07 survey was the second survey of this type. The first survey, reported in previous social reports, was conducted in 2001 but a change in the questions asked means time series data cannot be reported. Regional data was derived using the home address of students coded to meshblock areas.

Because of small sample size some regions have been aggregated to provide more robust results. Even so, estimates from sample surveys are subject to error, especially at a subnational level, and the data should be used with caution.

Data source: Adolescent Health Research Group University of Auckland, customised data from the Youth’07 survey conducted in 2007.