Longitudinal Immigration Survey: New Zealand - Wave 2, 2008

The Longitudinal Immigration Survey: New Zealand (LisNZ) was designed to collect longitudinal data in order to understand how well migrants settle over their first three years as residents in New Zealand.

The survey interviews migrants at around 6 months (wave 1), 18 months (wave 2), and 36 months (wave 3) after their residence uptake. The aim is to achieve a sample of around 5,000 migrants at the third interview, allowing for non-response rates and attrition rates.

All the results in the tables represent weighted estimates of migrants or percentages from the first two interviews. All the tables include the longitudinal respondents only (interviewed both at wave 1 and wave 2).

Survey population and scope

The target population for the LisNZ consists of all migrants (excluding refugees) who were at least 16 years old and were approved for residence in New Zealand from 1 November 2004 to 31 October 2005. These migrants could have already been in New Zealand at the time of residence approval or arrived in New Zealand within 12 months of their residence approval.

The target population excluded temporary visitors and all people from Australia, Niue, Cook Islands, and Tokelau. Migrants from Australia were excluded because they are entitled to enter New Zealand without applying for a residence permit or visa. In addition, migrants from Niue, Cook Islands, and Tokelau were excluded, as people from these countries have automatic rights to New Zealand citizenship. Refugees were also excluded from the target population because their routes to permanent residence, as well as their settlement experiences, are very different from those of migrants.

For practical and operational reasons, the survey population for the LisNZ was restricted to migrants in the target population who were living in the North Island, South Island, or Waiheke Island at the first interview (wave 1), and those who could understand at least one of the designated survey languages (English, Mandarin, Cantonese, Samoan, Korean, Hindi, and Punjabi).

Survey methodology

The survey involves a longitudinal design where the same migrants are interviewed three times over a period of three years. Individual migrants (as opposed to household units), are used for the selection of the LisNZ sample, the collection of data, and the analysis of the results from the survey.

The survey was designed to produce estimates for several subpopulations defined by immigration approval category and region of origin. It was estimated that 5,000 completed interviews in wave 3 would produce estimates of the required accuracy. A large initial sample was required to achieve this target and a stratified systematic random design was used to achieve this.

A sample of 12,202 migrants was randomly selected to participate in the survey. The migrants that were interviewed at wave 1 were then followed up for interview in wave 2.

Data collection method

The sample of migrants (both onshore and offshore) was selected from 1 November 2004 to 31 October 2005. As offshore migrants can arrive in New Zealand at any time up to 12 months after residence approval, the offshore arrival period is from 1 November 2004 to 31 October 2006. Consequently, the wave 1 interviews occurred between 1 May 2005 and 30 April 2007. Wave 2 interviews occured 12 months after the first interview, between 1 May 2006 and 30 April 2008.

The LisNZ was conducted using computer-assisted face-to-face interviews. The questionnaire was translated into each of the designated survey languages.

Response rate

Despite all efforts to locate migrants selected for the survey, a number of migrants could not be interviewed. From the 12,202 migrants selected for the first wave, 217 were not eligible to take part in the survey, 145 did not arrive in New Zealand in time and 984 had no initial contact address in New Zealand. Of the remaining 10,856 migrants, 7,137 were interviewed at wave 1. This corresponds to a 66 percent response rate.

Most of the ‘non-response’ at wave 1 was due to non-contact (84 percent) rather than refusal by respondents (note that the LisNZ is a voluntary survey). Non-contact was particularly high for migrants who were approved offshore, where the response rate was 57 percent, compared with 70 percent for those approved onshore.

Of the 7,137 migrants interviewed for the first wave, 6,156 were interviewed at wave 2. This corresponds to an 86.3 percent response rate. Of the 981 non-responses at wave 2, 79.0 percent were due to non-contact, 17.6 percent to refusal, and the remaining 3.4 percent to other reasons.

Estimation

Longitudinal weights are produced after each wave so that the achieved sample for each wave is weighted up to represent the longitudinal population of interest. The population of interest was all migrants who were approved for permanent residence between 1 November 2004 and 31 October 2005 and, if they were approved offshore, arrived in New Zealand within 12 months of the approval.

A basic sampling weight is attached to each migrant to reflect the probability of that migrant being selected in the sample. An initial adjustment is made to the basic sampling weight to account for the retention of a maximum of two migrants per application. Two further adjustments are then applied to account for unit non-response and to benchmark to known population totals.

The weighting classes used for non-response adjustments are based on the strata and type of application. The population totals used for benchmarking are the actual number of migrants for the survey period by sex and age group, obtained from other administrative sources.

Item non-response

Item non-response causes incomplete information when answers are not provided to some questions. The item non-response rate for the LisNZ was less than 1 percent for most variables at both waves. Imputation was not applied for wave 1 and wave 2 data of the LisNZ.

Reliability of survey estimates

Two types of error are possible in estimates based on a sample survey: sample error and non-sample error. Sample error is a measure of the variability that occurs because information has been collected from a sample of migrants rather than the entire population of migrants in a given reference period.

Sample errors have been determined at the residence approval categories. For skilled and family partner migrants, the sample errors have also been determined at the region of origin. For this release, sample errors have been calculated based on a replication variance estimation method for all numbers produced, including estimates of change between the two waves. For example, the proportion of migrants who owned a dwelling was 42.9 percent at wave 2. This proportion is subject to a sample error of plus or minus 1.8 percentage points, giving a confidence interval of 41.1 percent to 44.7 percent. In general, about 95 percent of the published results are expected to include the actual value in intervals calculated this way.

Estimated changes are also subject to sample error. For example, the proportion of migrants employed at wave 1 was 71.1 percent; at wave 2, this proportion increased to 74.3 percent. The estimated change in employment was thus, 3.2 percentage points, and is subject to a sample error of plus or minus 1.7 percentage points. The change in this estimate is larger than its associated sample error.

Non-sample errors are very difficult to measure. They include inaccuracies and bias in migrants’ responses during interviews, and errors made during data processing. Statistics NZ applies survey monitoring procedures, such as editing of the data collected from the interviews, to minimise these types of errors.

Confidentiality

All estimates provided in this release have been randomly rounded to 10. Weighted counts below 20 have been suppressed for confidentiality reasons. Percentages have been calculated after applying the rounding and suppression processes. Therefore, the estimated totals may differ from the sum of the individual cells.

Definitions

English language ability: English language ability is derived from responses to questions about migrants' language usage and their ability to read, write, speak, and understand English. English was either recorded as a language spoken best, otherwise various questions assessed migrants’ ability to read, write, speak, and understand English (each on a five-point scale). An overall score was derived from the average of the four scores.

Hourly wages: An hourly rate is derived by dividing a weekly gross income from wages and salaries (excluding self-employment income) by the number of hours usually worked (rather than the hours actually paid for). The income from wages and salaries excludes allowances, bonuses, overtime, and any other extra payments.

Immigration approval category: People who wish to migrate permanently to New Zealand must apply through one of the following residence streams of the New Zealand Residence Programme (NZRP): the Skilled/Business, Family Sponsored, or International/Humanitarian streams. An application for permanent residence includes the principal applicant (the key person assessed against the policy criteria) and any secondary applicants (other people in an application, for example, a partner and children). All people in the application are approved through the same policy, such as the Skilled Migrant Category (SMC). Each stream has a number of categories (or policies) with specific approval criteria. The following groupings are used in this release:

Labour market activity: Labour market activity is measured by categorising migrants as employed, not employed but seeking work, or not employed and not seeking work. The definitions for involvement in the labour force are aligned with, but not identical to, the concepts and definitions used in the Household Labour Force Survey. In particular, the definition of people seeking work is different as standard job search questions or questions about current availability to start work are not asked in the LisNZ.

• Employed: All individuals in the working-age population who worked for one hour or more per week, either as an employee or in self-employment; or worked without pay for one hour or more in work which contributed directly to the operation of a farm, business or practice owned or operated by a relative; or had a job but were not at work due to own illness or injury, personal or family responsibilities, bad weather or mechanical breakdown, direct involvement in industrial dispute, or leave or holiday.
• Not employed but seeking work: All individuals in the working-age population who were without a paid job and seeking work. Note that this is not the same as the official measure of unemployment derived from the Household Labour Force Survey.
• Labour force: All individuals in the working-age population who were employed or not employed but seeking work.
• Out of the labour force (or doing other activities): All individuals in the working-age population who were neither employed nor seeking employment. For example, people who were retired, or had personal or family responsibilities; people attending educational institutions; and people permanently unable to work due to disabilities.
• Working-age population: All individuals aged 16 years and over, regardless of their labour market activity. This definition differs from the standard International Labour Organisation (population aged 15 years and over).
• Employment rate: Proportion of the working-age migrant population who were employed or self-employed.
• Seeking-work rate: Proportion of migrants who were looking for work (and who were currently not working) out of all those migrants in the labour force.

Level of satisfaction with life in New Zealand: At each wave, migrants were asked how satisfied or dissatisfied they were with life in New Zealand. At each wave, they could give one of the following options:

• very satisfied with life in New Zealand
• satisfied
• neither satisfied nor dissatisfied
• dissatisfied
• very dissatisfied.

For the purpose of this release, these responses have been regrouped:

• very satisfied/satisfied with life in New Zealand
• neither satisfied nor dissatisfied
• dissatisfied/very dissatisfied.

Level of satisfaction with main job: Refer to the level of satisfaction with life in New Zealand definition.

Longitudinal respondents: People responding in all waves of the survey are called longitudinal respondents. Those people who have responded at waves 1 and 2 of the LisNZ are the current longitudinal responding individuals. In this release, all estimates and proportions refer to the longitudinal respondents.

Perception of safety in New Zealand: At each wave, migrants were asked how safe or unsafe they felt in New Zealand considering crime only. At each wave, they could give one of the following options:

• very safe in New Zealand
• safe
• neither safe nor unsafe
• unsafe
• very unsafe.

For the purpose of this release, these responses have been regrouped:

• very safe/safe
• neither safe nor unsafe
• unsafe/very unsafe.

Region of origin: Region of origin is derived from country of nationality/citizenship. For an applicant with dual citizenship, citizenship refers to nationality recorded on the passport used for their residence application. The following regions of origin were derived based on these criteria:

Region of settlement: Derived from the 16 New Zealand regional councils. For the purpose of analysis, these have been grouped into three regions:

• Auckland: Auckland regional council area
• North Island except Auckland: Includes Northland, Waikato, Bay of Plenty, Gisborne, Hawke’s Bay,Taranaki, Manawatu-Wanganui, and Wellington regional council area
• South Island: Includes West Coast, Canterbury, Otago, Southland, Tasman, Nelson, Marlborough regional council area.

Skill levels: Immigration New Zealand’s (INZ) definition of skilled employment is primarily based on the Australian and New Zealand Standard Classification of Occupations (ANZSCO 06). The five skill levels in ANZSCO are defined in terms of formal education and training, previous experience, and on-the-job training. The determination of boundaries between skill levels is based on the following definitions:

Spell: At wave 2, the respondents were asked about the activities they had been doing since their last interview. For each activity (time spell), the following information was gathered or coded from the responses collected in the questionnaire:

• Type of activity: Paid job, self-employed, family unpaid job, casual work, looking for work, doing other activity.
• Start and end dates: If the respondent is still doing this activity, then no end date is recorded. When the respondent has been doing two activities, two time spells are recorded with overlapping start and/or end dates.
• Occupation and industry codes: For the employment spells only.

Time spell data are used, for example, to calculate the number and duration of jobs respondents had between two waves.

Wave: In a longitudinal survey, interviews are conducted with the same respondents repeatedly over time. The LisNZ is thus made up of cycles, or waves, of interviewing. The wave length is one year between the wave 1 and wave 2 interviews and 18 months between the wave 2 and wave 3 interviews.