2. Definitions and survey methods

2.1 Who is a temporary employee?

In theory, a job is temporary if the worker does not have any expectation of continuous long-term employment. A temporary employee has been hired for a specific time period, is working until the completion of a specific project, was hired to temporarily replace another worker, is filling a seasonal job, or is employed only when needed by their employer.

In practice, some individuals in jobs that are technically 'temporary' have relatively long-term relationships with their employer and do have some expectation of continuing work. The most significant group of 'quasi-temporary' employees who are considered in this paper are 'permanent seasonals': employees who said that their main job was permanent but also said that it was only available at certain times of the year, ie was seasonal. Some of the casual and fixed-term employees who were surveyed in SoWL had also been in their jobs for several years and may have had an expectation of continuing (or even long-term) employment, despite indicating that their job was technically not permanent.

2.2 How temporary work was measured in the Survey of Working Life

A temporary employee was defined in the survey as an employee whose job only lasts for a limited time or until the completion of a project. In practice, employees who answered 'no' to a question on whether their main job was permanent, or 'yes' to one of the temporary job questions, were classified as temporary. The initial question that was used to distinguish between temporary and permanent employees was worded as follows:

'A permanent employee is guaranteed continuing work. They can stay in their job until they decide to leave or their employer makes them redundant. In your job, are you a permanent employee?'

Employees who said that their job was not permanent were asked a series of follow-up questions designed to identify their type of employment relationship, so that they could be classified as a fixed-term worker, a casual worker, a temporary employment agency worker, or having some other type of employment relationship. All employees, including those who said that their job was permanent, were asked whether their job was seasonal. People who initially said that their job was permanent but later indicated that their job was seasonal, were classified as temporary, because seasonal jobs do not provide continuous work throughout the year.^[1]

People who gave more than one 'yes' response to the sequence of questions on the nature of their temporary job were classified to one job type using a prioritisation scheme. This framework prioritised 'yes' responses in the following order:

1. temporary agency worker
2. casual worker
3. fixed-term worker
4. seasonal worker (employment relationship not further defined)
5. other temporary worker.

The full set of questions that were used to classify temporary employees by their type of employment relationship is given in Appendix 1.

Previous qualitative research in New Zealand has shown that many temporary and casual employees do not have a good understanding of the terms of their employment, and would struggle to define their job as casual, fixed term, or something else.^[2] The SoWL did not ask respondents to classify themselves, and instead classified them using their responses to more concrete questions such as 'were you hired to temporarily replace another worker?'. However, this strategy did not eliminate the potential for confusion, and it is likely that there is some measurement error in the survey's estimates of the total number of temporary employees and the numbers in each type of temporary work.

2.3 Survey design and data collection

The Survey of Working Life was administered as a supplement to the Household Labour Force Survey (HLFS), and shared the same sample and collection methods. Copies of the HLFS and SoWL questionnaires can be downloaded from the Statistics NZ website.

All eligible responding individuals in the March 2008 quarter HLFS who were employed in the reference week were asked to participate in the SoWL. Eighty-four percent responded, giving a total sample size of 14,510 individuals. Proxy responses from other members of the same household were not accepted (except under certain limited conditions). Interviews were carried out by phone or in person.

The pattern of response by employed people in the Household Labour Force Survey sample to the SoWL was not strictly random. Non-respondents were disproportionately male, aged under 25 years, members of the Maori and Pacific ethnic groups, people without any educational qualifications, and people who had worked relatively long hours in the reference week. However, the survey weights developed by Statistics New Zealand were designed to increase the representativeness of the final estimates, offsetting to some degree those response biases. The SoWL sample was benchmarked to the HLFS employed sample by detailed age and gender group, Mäori /non-Mäori ethnicity, month and region.