Stats NZ has a new website.

For new releases go to

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+
Impact of the 22 February 2011 earthquake on the Household Labour Force Survey

The earthquake that struck the Canterbury region on 22 February 2011 had a significant impact on the operation of the Household Labour Force Survey (HLFS). In this note we provide an explanation of how Statistics New Zealand is responding to the impact on this survey. 

The HLFS is a survey of labour market activity. Each quarter, a range of statistics relating to employment, unemployment, and people not in the labour force is published. To collect data for the HLFS, Statistics NZ interviews 15,000 households across New Zealand throughout a quarter.

The earthquake caused significant disruption to HLFS surveying in and around Christchurch. This meant a different method than usual had to be used to estimate data from this area. We'll be reviewing the method  in the coming months but we are confident that it is robust and appropriate. We do not anticipate any significant changes to the national estimate.

The main impact of the earthquake on the HLFS is on data collection. The March 2011 quarter will be released as scheduled on 5 May.

Immediately after the earthquake, Statistics NZ suspended interviewing in the whole South Island for the remainder of the week. For the following week, no interviews were conducted in the Canterbury region. For the remaining four weeks of the quarter, no interviews were done in Christchurch city, or in the Selwyn and Waimakariri districts.

Since there was no interviewing in such a large area of Canterbury, any changes in labour force status as a result of the earthquake will not be observed in the March quarter release of the HLFS.

The weighting and estimation methods usually used in the HLFS are designed to compensate for normal sample loss from people not responding to the survey. Given the impact of the earthquake on the sample, these methods were not appropriate because the sample loss is larger than normal. To address this, Statistics NZ modified the estimation method.

Statistics NZ considered several options to account for the lower response rate due to the Canterbury earthquake. We also consulted with the Australian Bureau of Statistics, and then chose the most suitable method. 

The method involved modifying the non-response adjustment process for Canterbury. The sample loss from Canterbury has been accounted for by estimating the labour force status of people who were unable to respond, due to the earthquake, from the responses of people interviewed before the earthquake. Canterbury is, on average, 11 percent of the HLFS sample – we were unable to sample 36 percent of this group in the March quarter. We will discuss this modification to the estimation method further in the technical notes to the March quarter HLFS.

The March 2011 quarter’s information release will also include statistics for New Zealand that exclude the Canterbury region. A number of data users have already expressed interest in these statistics.

Given the sample loss, there will be a larger margin of error for the Canterbury regional estimates, which indicates that a greater degree of caution will need to be used with this data.

For further information please contact:
Will Bell
Wellington 04 931 4600

28 April 2011

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