If you want to understand the 1999 election read Proportional Representation on Trial. Click on the cover image for further information

Funding for the 1999 NZES was provided for by Foundation for Research, Science, and Technology  (FRST) and the University of Waikato. Funding for the pre-election survey was from the University of Waikato.

The principal researchers (with current affiliations) are:

Jack Vowles, j.vowles@exeter.ac.uk Susan Banducci, s.a.banducci@exeter.ac.uk Jeffrey Karp, j.karp@exeter.ac.uk University of Exeter

Peter Aimer, p.aimer@auckland.ac.nz Raymond Miller, rk.miller@auckland.ac.nz Ann Sullivan, a.sullivan@auckland.ac.nz University of Auckland

The 1999 Election
New Zealand's second election under proportional representation was held on November 27, 1999.

The 1999 Study
The 1999 Study was at the time the most sophisticated political survey ever conducted in New Zealand. The first component of the 1999 NZES is a pre-election telephone survey of 3500 New Zealanders beginning on October 18 and continuing daily until election day on November 27. Using a rolling cross section design we surveyed a random sample each day of  80-90 persons eligible to vote. The pre-election survey measures voting intentions, political attitudes, and evaluations of MMP. Here is a detailed description of the methodology. The ratio of refusals to completed interviews was 43:57. 

The 1999 post-election sample was primarily by post, with a small telephone supplement. It is made up of three sections, the first a ‘new sample’ sampled directly from the electoral rolls (N=1059, with at the most conservative estimate a response rate of 58 per cent for the postal component alone, or 64 per cent including the telephone top-up). Further sections are made up of a panel of respondents form the 1990, 1993, and 1990 Election Studies (n=2342), and a panel of respondents first sampled in the Campaign Study (N=2489). Due to panel attrition response rates for these sections are somewhat lower, but analysis of response patterns between these and the new sample indicate no significant apparent biases. 

As part of the 1999 NZES a random sample of 1000 Maori identifiers were selected for face to face interviews in a post election survey that promises to be a landmark in research on Maori political opinion. 

As in 1993 and 1996 a survey of political candidates has also been conducted. Candidates nominated by the Labour, National, ACT, Alliance, Greens, and New Zealand First parties were sent a 22-page questionnaire, with questions arranged under the following sub-headings: Political Background and Activity; Candidate Selection and Campaigning; The Role of an MP; Opinions and Policies: Government and the Electoral System; Background Information. A similar survey was sent to party candidates in 1996 and party and conference delegates in 1993.

Questionnaires were sent to a total of 452 candidates. The overall response rate was 62 per cent, or 282 usable questionnaires, distributed among the parties as follows (party response rates in brackets): Labour 56 (64%), National 44 (52%), ACT 52 (72%), Alliance 48 (68%), Greens 52 (72%), and New Zealand First 27 (40%), 

No individual respondent can be identified in the dataset. Respondents, can, however, be grouped according to shared characteristics - such as party affiliation, age, sex, income, religion etc. In keeping with our undertaking of anonymity, however, the researchers will not process the data in a way that results in very small groups, and therefore the possibility of attributing a response to an individual, by inference. For similar reasons a number of social and demographic variables will be deleted from the dataset when it is released to other researchers.