Abstract
Following the 1999 election in New Zealand, a Bill was introduced into Parliament to prevent the publication of polls during the month prior to a general election. Two parties, New Zealand First and National, feared that polls had adversely influenced their electoral performances. The claim is reviewed, the methodologies used to estimate the effects of polls on the vote discussed, and a test is applied for New Zealand in 1999. The data is from a pre- and post-election panel, the pre-election component based on rolling cross-sections estimated on a daily basis throughout the campaign. It finds evidence for a small but very significant polling effect.
Get full access to this article
View all access options for this article.
