|
Information about the Survey of Older People in 2000 |
 | |
| Document Map |
| Jump to: Availability |
| Jump to:Design |
| Jump to:Output Variables |
| Jump to:Guide To Interpreting Data |
| Jump to:Customised Output |
| Jump to:Catalogue & Reference Numbers |
| Jump to:Commercially Available Tables |
| Jump to:Other Comments |
| Jump to:Classification(s) Used |
| Jump to:Glossary of Terms |
| Jump to:Contact Details |
Availability
| Valid From: .. | |
| To: .. | Ongoing |
| Frequency: .. | One off |
Design
Purpose: The Survey of Older People (SOP) was part of a larger programme of research commissioned by the Super 2000 Taskforce. The Taskforce was established to develop a stable retirement income framework to assist New Zealanders to plan for their retirement in confidence (The Taskforce was dis-established by the Labour Government elected in October 1999, however, the research programme was continued under the direction of the Ministry of Social Policy).
As part of its mandate, the Taskforce embarked on a programme of research on the living standards of older people and commissioned Statistics New Zealand to undertake a survey of approximately 3,000 persons aged 65 years and over.
A companion survey of approximately 500 Mäori persons aged between 65 and 69 was also conducted. The documentation for this is stored in a separate profile.
The broad research aim was to describe the standard of living of older people and investigate the determinants of variation in living standards of older people.
General Information ..Target Population
The target population for the SOP was defined as the usually resident, non-institutionalised population aged 65 years and over.
The following people were excluded:
- long term residents of homes for the older persons, retirement homes, hospitals and psychiatric institutions
- inmates of penal institutions
- permanent members of the armed forces
- overseas diplomats
- overseas visitors who expect to be usually resident in New Zealand for less than 12 months
This is similar to the population definition for other SNZ household surveys.
Based on current population estimates, the size of the target population was approximately 421,100 older people.
Population covered
The survey population for the SOP was defined as the civilian, usually resident, population aged 65 years and over living in permanent private dwellings.
The following people were excluded:
- those living in non-private dwellings (i.e. persons residing in hotels, motels, hostels, etc.)
- those not living in permanent dwellings (i.e. persons residing in tents, caravans not permanently sited, etc.).
For practical reasons a small number of households, which are part of the defined survey population, were excluded from the survey. These coverage exclusions should be made clear in any publication of the survey results.
Such households are those which are in very remote locations and inaccessible by private car. The exclusions of these households will have negligible impact on the results.
Survey Design
Introduction
The SOP was administered through the Household Labour Force Survey (HLFS), with the SOP run as a supplement to the HLFS in the March 2000 quarter. All households containing a person aged 65 and over who were eligible for and participated in the HLFS were eligible for selection into the SOP.
Sample Frame
The sample frame is the 'list' of the target population from which the sample is selected. This section describes how the sample frame for the HLFS and consequently the SOP was constructed.
New Zealand is divided into 19,102 small geographic areas defined by meshblock. A meshblock is the smallest geographic unit for which statistical data is collected and processed by Statistics New Zealand. A meshblock is a defined geographic area, varying in size from part of a city block to large areas of rural land. Each meshblock abuts against another to form a network covering all of New Zealand including coasts and inlets, and extending out to the two hundred mile economic zone. Meshblocks are added together to ‘build up’ larger geographic areas such as area units and urban areas. They are also the principal unit used to draw-up and define electoral district and local authority boundaries. These geographic areas are called Primary Sampling Units (PSUs). The only areas of New Zealand not included in PSUs are offshore islands (except Waiheke Island). Most PSUs contain between 50 and 100 dwellings with an average size of about 67 dwellings.
The PSUs are divided into groups or strata, according to various characteristics derived from 1996 census data. There are two stages to this stratification. The first stage groups PSUs according to 14 defined regions based on regional council boundaries. In the next stage, PSUs are assigned to strata based on a number of PSU characteristics. These include urban/rural classification, proportion of Maori in the PSU, proportion of Pacific Island people in the PSU, proportion of Asian people in the PSU and other socio-economic variables such as level of education, employment and income. In total 120 strata are formed.
This two-stage design was selected as being the best option to meet the time frame and provide high coverage of the population of interest at an acceptable cost.
Sample Selection
A three-stage stratified cluster design, based on the area frame described above, was used to select the sample.
The three stages of selection were:
- selection of geographic area units (PSUs);
- selection of households within these selected PSUs; and
- selection of one individual per household
Selection of geographic area units (PSUs)
The SOP was administered in the 1,760 PSUs selected for the HLFS. These PSUs were selected by drawing an equi-probable selection of PSUs, which had not been recently selected in other SNZ surveys, from each of the 120 strata.
Selection of households within these selected PSUs
Each PSU selected for the HLFS had been previously enumerated. That is, a list of all private dwellings was constructed for each PSU. For the HLFS, a systematic sample of dwellings was selected from this list. This means that every Kth dwelling on the list was selected into the HLFS sample. The sampling interval, K, was determined by the sample size required from each PSU and the number of dwellings in the PSU. The first dwelling was randomly selected from the first K dwellings. Systematic sampling ensured that the selected sample within each PSU contained households that were spread over the entire PSU.
It is important to note that the SOP provides estimates not for a single point in time, but as an average across the survey period, which was from 7 February to 7 April 2000.
Selection of one individual per household
All in-scope households that participated in the HLFS were eligible for the SOP. In-scope refers to those households which were part of the target population.
Every selected household was enumerated via the HLFS household form, which lists all the household members and demographic information such as age, sex and whether the individual was in-scope.
One eligible older person per household was selected for the SOP. From a completed HLFS household form, all individuals eligible for the SOP were identified and one eligible person was selected at random using a selection grid.
HLFS non-respondents were automatically considered to be non-respondents in the SOP.
Selection Grids
A selection grid was used to identify which of the eligible people in a household should be randomly selected for the SOP (where more than one eligible person was present). This followed the method outlined in Survey Sampling (Kish, 1965).
This method was based on a maximum of four older people in a household. The very small number of households containing more than 4 eligible older respondents were treated as if they contained exactly four.
Selection Process
Prior to the selection of the respondent, the interviewer ranked each eligible older person within the household by age. For example, the oldest eligible older person was assigned the rank of 1, the next oldest rank 2, and so on. The interviewer consulted the selection grid in order to ascertain which eligible older person of the household should be approached for an interview.
If the selected individual was unavailable to participate in the survey at the time of the HLFS, attempts were made to contact the person at a later date. No substitution for other household members to replace non-respondents was undertaken.
Extra Rotation Group
A separate sample of households that completed their obligation to the HLFS in the September 1999 quarter were added to the SOP to boost the achieved sample size.
HLFS PSUs are separated into 8 equi-sized rotation groups. Every quarter, one rotation group is rotated out of the HLFS survey and replaced with another rotation group. The new rotation group contains a different selection of dwellings from the same PSUs as the previous rotation group.
Every 5 years, SNZ's household surveys frame is redesigned. PSUs are reformed and then stratified using the latest Census figures. A new HLFS sample is selected from this frame. During a period of transition between the old and new HLFS sample, 2 rotation groups are rotated out each quarter. The last transition occurred in 1999.
One of the rotation groups that rotated out of the HLFS survey in the September 1999 quarter was added to the SOP to boost the achieved sample size. The households that contained at least one eligible older person were identified and a respondent randomly selected. The extra rotation group has been incorporated in with the main sample of older respondents.
This extra rotation group is part of the old household surveys frame and contains a different selection of PSUs to those selected in the HLFS sample for the March 2000 quarter
Expected and Achieved Sample Size
Based on the expected response rate at each stage of selection, sufficient addresses were selected to result in a final achieved sample of at least 3000 older people.
Out of the 4124 older persons households (An older persons household is one that contains at least one person aged 65 or more responding to the screening questionnaire (i.e. HLFS), 3060 older people responded to the SOP questionnaire.
Response Rate Calculation
The response rate is a weighted estimate of the total responding eligible older population as a proportion of the total eligible older population. This should be interpreted as the proportion of the eligible older population covered by the SOP. It is not the same as the unweighted calculation of the number of older persons responding as a proportion of the number approached.
This response rate can be separated into household and individual response rates that allow estimation of the magnitude of the sample loss at each stage of selection:
- the household response rate can be interpreted as the proportion of eligible older households responding to the screening questionnaire (i.e. HLFS); and
- the individual response rate can be interpreted as the proportion of eligible older people responding to the SOP questionnaire, given they responded to the screening questionnaire.
SOP Response Rate
The response to the HLFS screening questionnaire was 91% of eligible elderly households. Of these, 75% responded to the SOP questionnaire.
The SOP covered 68% of the eligible elderly population. This response rate compares favourably with that for the Time Use Survey, which SNZ conducted recently. The Time Use survey was of similar respondent load and achieved a 72% response rate.
The SOP used two questionnaires - a Household Form and a Personal Questionnaire. Additional information was taken from the Household Labour Force Survey completed by the respondent. This ensured that respondents were not asked for information more than once.
The questionnaire was administered in a face-to-face interview with the selected respondent. The duration of interviews ranged from less than an hour to well over two hours, with the average estimated at just over an hour. The length-of-interview data is based on field test data, as this information was not collected in the main survey. A number of questions were asked of the respondent's partner (if applicable). The respondent's partner could either answer these or the respondent could proxy for them.
The questionnaire design and content was driven by the information requirements of the Super 2000 Taskforce and the Ministry of Social Policy, who drew upon the work of previous surveys of a similar nature. The work drew upon the 1974 Survey of the Aged, studies conducted in Australia by Peter Travers, and input from a range of academics and researchers.
Extensive cognitive testing and a field test were conducted to refine the questionnaire content and structure.
Household Form
The Household Form used was identical to that used for the Household Labour Force Survey. Its purpose was to collect information on the demographic characteristics of the people usually resident in the selected household and to allow selection of a respondent. The information collected was:
- date of birth
- sex
- ethnic group
- country of birth
- years in New Zealand
- marital status
- household relationships
Personal Questionnaire
The personal questionnaire contained questions on the following subjects:
- Health and disabilities
- Mood
- Mobility
- Family and Social Networks
- Financial Support
- Family Support
- Unpaid Work
- Home production
- Accommodation
- General Insurance
- Superannuation
- Government Allowances
- Overseas pensions
- Private superannuation or job-related pension schemes
- Regular paid employment and self employment
- Other sources of Income
- Total Income
- Assets
- Life History
- Recent Financial Stress
- Socio-economic status
- Self rating standard of living
- Whether respondents owned a range of items and participated in a series of activities and their ranking of the importance of these.
Household Labour Force information
For each selected respondent information was taken from their previously completed HLFS personal questionnaire. This avoided having to repeat questions already previously asked. The information used from the HLFS was:
- education qualifications
- labour force participation
SOP Questionnaire and show cards
Output Variables
Output variables are pieces of individual information that can be extracted from the survey/output data. Often output variables can be cross tabulated with other output variables, for example sales by industry classification. The list below contains all current and past output variables which have been released for this survey/output.
The data dictionary is attached below
Changes in Output Variables over time
Guide to Interpreting Data
Summary of Changes to Survey/Output ..One off survey,so changes not applicable
Usage and Limitations of the Data ..
Related Data Sources ..
Sampling Errors ..
Non-sampling errors Non-sampling errors in the survey data may result from errors in the sample frame, respondent error, mistakes made during processing survey results and non-response imputation. Statistics New Zealand adopts procedures to detect and minimise these types of errors but they may still occur and they are not quantifiable.
Caveats on Release ..See above
Customised Output
Two output data sets were created as part of the SOP output.
The first data set contains the information from those respondents who signed consent forms agreeing to have their information made available to a range of researchers. This data set is called the consent data set. Access to this data set is managed by the Ministry of Social Development. Contact the Manager Information Analysis Group at the Ministry of Social Development for more information
On the consent dataset there are two weight variables included. One weights the respondents up to the survey population on an individual basis (see section 10.2). The other weight was calculated by the Ministry of Social Policy (now the Ministry of Social Development) - it averages the individual weights of core economic units (couples) and then divides through by the weighted population. It was used in the regression work that was done to calculate the material well-being scale.
The second data set contains the information collected from all respondents to the SOP. This data set is called the full data set. Access to this data set is managed by SNZ. As for the consent data, there are two weight variables on the full dataset.
Catalogue & Reference Numbers
Classification(s) used
Glossary of Terms
| Customer Service E-mail Contact: | info@stats.govt.nz |
| |
| Survey Enquiries E-mail Contact: | julie.woolf@stats.govt.nz |
Liability
Statistics New Zealand gives no warranty that the information or data supplied contains no errors. However, all care and diligence has been used in processing, analysing and extracting the information. Statistics New Zealand shall not be liable for any loss or damage suffered by the customer consequent upon the use directly, or indirectly, of the information supplied in this product.
|
