Stats NZ has a new website.

For new releases go to

www.stats.govt.nz

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
New Zealand Income Survey: June 2008 quarter - Revised
Embargoed until 10:45am  –  09 October 2008
Technical notes

Background to the survey

The New Zealand Income Survey (NZIS) is run annually as a supplement to the Household Labour Force Survey (HLFS) during the June quarter (April to June). The NZIS was run for the first time in the June 1997 quarter.

All respondents to the HLFS are asked to participate in the NZIS. Before 2003, data was accepted from a proxy only if people were unable to answer the survey on health or language grounds. From 2003, proxy interviewing was extended to a slightly wider range of situations, as long as the following conditions were met:

  • the practice was acceptable to the respondent for whom the information was given
  • the proxy was thoroughly knowledgeable about the respondent's income details
  • every effort had been made to collect information from the respondent in the first instance.

Questions relate to the respondent's most recent pay period, except for questions on annual income, self-employment income and investment income, which cover the 12-month period prior to the interview.

The following items are collected:

  • actual and usual gross wages and salaries for main job and up to two other jobs by:
    • ordinary time
    • overtime
    • other income
  • weeks and hours worked corresponding to the components of wages and salaries as above
  • sources of latest, actual, gross government transfers received
  • total latest, actual, gross government transfers received, in total and from each agency (Ministry of Social Development, Inland Revenue and Accident Compensation Corporation (ACC))
  • total latest, actual, gross private superannuation payment(s) received
  • total latest, actual, gross 'all other private transfers including pensions and annuities' received
  • weeks covered for the transfer payments defined above
  • total annual gross income received from self-employment including overtime
  • total annual gross income from investment
  • total annual gross income from all income sources (range categories as used in the population census).

Some forms of income were not asked for in detail in the NZIS, including income from hobbies and casual jobs. This may mean that total income estimates from the survey underestimate actual total income.

Suppressed estimates in this release

Cells with estimates of less than 1,000 have been suppressed and appear as 'S' in the tables. These estimates are subject to sampling errors that are too great for most practical purposes.

Introduction of computer-assisted interviewing

The introduction of the computer-assisted interviewing (CAI) strategy involved a change in mode of data collection from traditional pen and paper interviewing to computer-assisted personal interviewing (CAPI) for newly selected households, and a change from decentralised telephone interviewing to centralised computer-assisted telephone interviewing (CATI) for the remaining households in the sample. The main benefits of CAI are improved data quality through better-controlled interviews, and modest resource savings through the elimination of some processing steps and the centralisation of others.

CAI was progressively implemented into the HLFS and NZIS, commencing with one-eighth of the sample for the June 2005 quarter. In the June 2006 quarter, newly selected households (one-eighth) were interviewed via CAPI and three-eighths of the sample were interviewed via CATI. For the June 2007 quarter, CAI implementation was complete in the NZIS, with CATI representing seven-eighths of the sample and the newly selected households again interviewed via CAPI. During the transitional stage to CAI, estimates for the CAI part of the sample were compared with estimates from the remainder of the sample, and no statistically significant differences were detected.

Definitions

A full set of definitions is available from Statistics New Zealand. The collected data was cash only, pre-tax (gross) income wherever possible, and did not include any non-cash fringe benefits. In this Hot Off The Press, all tables showing wages and salaries include usual income figures rather than actual income figures.

In paid employment is used to describe those who are receiving self-employment income and/or income from wages and salaries. Those who are not in paid employment are those who do not fit into the 'in paid employment' category. Those people may or may not have a source of income.

Hourly earnings in the tables relates to the number of hours usually worked and the usual income, rather than the hours actually paid for and actual hours worked.

Government transfers in the tables relates to the sum of income from benefits, family support, student allowances, ACC, New Zealand Superannuation, and veterans and war pensions.

Other transfers in the tables relates to the sum of private superannuation and other private transfers, including pensions and annuities.

Working age population is the population the NZIS represents. It is the same population as the HLFS, and consists of the usually resident, non-institutionalised civilian population of New Zealand aged 15 years and over.

Quintile is one-fifth of the population. The bottom quintile in terms of income represents the 20 percent of the population with the lowest personal incomes, while the top quintile represents the 20 percent of the population who receive the highest incomes.

Quintile boundary is the dollar value at which the quintile falls. Given that the bottom quintile has income below $190 per week (rounded figures), the quintile boundary between quintiles one and two is $190.

Responses to the New Zealand Income Survey

Statistics New Zealand was unable to collect valid data from all eligible respondents. The most common reasons for this were that a respondent was not able to be contacted, or that a respondent was not able to provide the relevant information about their income when asked.

Of the 85.8 percent of eligible households who responded to the HLFS, 84.2 percent of eligible individuals gave a valid response to the NZIS.

An analysis of the valid responses to the NZIS calculated as a percentage of those who responded to the HLFS by major sub-population is given below:

Responses to the NZIS as a percentage of HLFS respondents

  Response rate (%)
 Sex

    Male

82.0
    Female 86.3
 Ethnicity
    European 84.6
    Māori 82.5
    Pacific peoples 83.7
    Other 83.5
 Labour force status
    Employed 80.6
    Not in labour force 91.3
    Unemployed 91.9
    Full-time employed 80.0
    Part-time employed 82.4
 Overall 84.2

Imputation

For those records where there was not a valid and usable response, a form of imputation known as 'hot-deck imputation' was used. In this imputation method, a 'donor record' replaces the non-usable record in the dataset. The donor record is chosen randomly from an imputation pool of records that have similar characteristics to that of the record to be imputed.

The imputation pools were constructed on the basis of the following HLFS variables: age group, sex, ethnicity, highest qualification, labour force status (modified), full-time/part-time work status and region. Imputation classes were combined in a priority order when there were fewer than 10 donor records in any particular class.

Hot-deck imputation was chosen for three reasons:

  • to maintain an accurate income distribution of the New Zealand population
  • to allow the calculation and analysis of household income for a larger number of households
  • to account for likely biases due to non-response.

There was little change in the income distribution or average income figures in the dataset due to imputation. For further information about the imputation method, or the effects of imputation on the final dataset, please contact Statistics NZ.

Reliability of estimates

The HLFS sample comprises approximately 15,000 private households, sampled randomly from rural and urban areas throughout New Zealand. The final NZIS dataset consists of approximately 28,000 valid person records, including approximately 5,000 imputed person records.

Two types of error are possible in estimates based on a sample survey: sampling error and non-sampling error. Sampling error is a measure of the variability that occurs by chance because a sample rather than an entire population is surveyed. Sampling errors are available on request. Non-sampling errors include errors arising from biases in the patterns of response and non-response, inaccuracies in reporting by respondents, including inaccuracies as a result of proxy interviewing, and errors in the recording and coding of data. Non-sampling errors are not quantified.

Compositional effects

Movements in income statistics are influenced by many factors. As well as changes in levels of income, movements are also influenced by the composition of the population from survey to survey. These changes occur between males and females, different ethnic groups, different labour force statuses, numbers of full-time and part-time workers, different industries or within industries, and different occupations or within occupations.

Rebase of the population estimates

Update: A five-yearly revision of the Household Labour Force Survey (HLFS) series is currently underway. The revision will incorporate updated population weights derived from the most recent population census (2006). This population rebase will have an impact on estimates from the NZIS, and updated tables with revised figures will be available on the Statistics NZ website in 2009. Users will be advised of a date closer to the time through the website and the Expert Data Users newsletter.

Ethnic statistics

Prioritisation of ethnic groups occurs when one ethnic group is assigned to an individual who has answered with more than one ethnicity.

People with multiple responses to the ethnicity question are assigned to one ethnic group using the following prioritisation:

  1. Māori
  2. Pacific peoples
  3. Other ethnic groups
  4. European.

For example, if an individual responds as both Māori and European, then the ethnic group assigned to this person would be Māori. Unlike the working-age population for different sexes and age groups, the ethnic working-age populations are not benchmarked to census-based demographic projections. This, along with other sample design restrictions, causes a high degree of volatility in the ethnic statistics of the NZIS. Movements in the working-age population estimates of certain ethnic groups may therefore be a reflection of this volatility, rather than a real change in the estimated ethnic demographic.

Update:In the December 2007 quarter, the HLFS began collecting ethnicity data using the 2005 standard classification of ethnicity. The prioritisation method of grouping ethnic data has been continued this quarter. From the June 2009 quarter the NZIS will publish ethnicity data using the single/combination output method. This will be a complete break in the ethnicity series, as the prioritisation of ethnic groups will no longer be produced.

Using the single/combination ethnicity output, people are counted just once according to the ethnic group or combination of ethnic groups they have reported. This means that the total number of responses equals the total number of people who stated an ethnicity.

The 2005 New Zealand standard classification of ethnicity enables the HLFS and the NZIS to collect and output more detailed ethnicity data, especially for the Asian ethnic group, which was not previously collected. The single/combination ethnic groups which will be published from June 2009 quarter onwards are listed below:

  • European only
  • Māori only
  • Pacific peoples only
  • Asian only
  • MELAA(1) only
  • Other ethnicity only European/Māori
  • Two or more groups not elsewhere included
  • Total all ethnic groups.(2)

(1) MELAA = Middle eastern/Latin american/African.

(2) 'Total all ethnic groups' includes residual categories.

For more information about the 2005 New Zealand standard classification of ethnicity please read the Review of the Measurement of Ethnicity.

Household statistics

The household categories incorporate the concept of dependent children rather than just children. A child is a person of any age who usually resides with at least one parent (natural, step, adopted or foster), and who does not usually reside with a partner or child(ren) of his or her own. Statistics NZ defines a 'dependent child' as a child aged under 18 years and not in full-time employment.

The household income statistics table included in this release excludes households where all members are outside the ages of 18 to 64 years. This exclusion primarily affects 'couple only' and 'one person' households. These households typically contain two distinct groups of the population: couples and single persons who are likely to be in the labour force, and couples and single persons who are primarily retired. Because these groups can have very different income characteristics, the household income table excludes elderly households where all members are aged 65 and over. The income figures for 'couple only' and 'one person' households for those aged 65 years and over are available from Statistics NZ on request.

Other possible uses of the data

The tables in this Hot Off The Press are not a full set of the possible analyses that could be carried out from the NZIS data.

More information

For more information, follow the link from the technical notes of this release on the Statistics NZ website.

Copyright

Information obtained from Statistics NZ may be freely used, reproduced, or quoted unless otherwise specified. In all cases Statistics NZ must be acknowledged as the source.

Liability

While care has been used in processing, analysing and extracting information, Statistics NZ gives no warranty that the information supplied is free from error. Statistics NZ shall not be liable for any loss suffered through the use, directly or indirectly, of any information, product or service.

Timing

Timed statistical releases are delivered using postal and electronic services provided by third parties. Delivery of these releases may be delayed by circumstances outside the control of Statistics NZ. Statistics NZ accepts no responsibility for any such delays.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Top
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+