Official Statistics

From Mathsreach

Jump to: navigation, search
Former New Zealand Government Statistician, UK Chief
Statistician and Chair of the NZIMA Board, Len Cook, right,
spoke with Jenny Rankine.

The cancellation of the 2011 Census, which was to have taken place on March 8 is only the third
time a major event has stopped the Census. In 1931 it was abandoned because of the depression,

and in 1941 it was cancelled due to a world war. This time it was deferred because of disruption caused by the Christchurch earthquake.

The Census was to have been processed in Christchurch, but Statistics New Zealand buildings in Christchurch have suffered extensive damage. A replacement date has yet to be determined but it is unlikely to be in 2011.

Cook says that for New Zealand, more than larger countries such as the United Kingdom, “the
cancellation of a population census creates a hole in knowledge we have about ourselves that we
cannot fill adequately by other means. The Census is fundamental to public trust in government and
the quality of much public policy.”

He ranks New Zealand’s official statistics “right up there with the best with the use of statistics
methods and information technology. We’ve been very innovative in releasing Census data
with Spacetime mapping products, in the use of tax records for economic statistics, in automated
scanning of forms and organising internet access for 2011.”

Professor Stephen Haslett of Massey University is one of many who have worked on linking official
statistics with administrative data records “to get better detail and accuracy for less money”. For
example, a linkage of the unemployment register and the Household Labour Force Survey obtained
“much better estimates of the International Labour Organisation’s definition of unemployment” than
either data set alone. He had to take into account the changing relationship between the two data
sets over time with different political decisions about benefit criteria.

Says Cook: New Zealand “put a lot of effort into high quality, regularly-updated design for
the Household Labour Force Survey and the Household Income and Expenditure Survey, and
business surveys such as monthly retail trade.” It can be difficult to get accurate information
about attitudes, he says. “Surveys of wealth have been quite difficult; people don’t always know
what they’re worth. Areas that are stigmatised, for example, alcohol consumption in household
expenditure, are often quite significantly underreported.

Developing methods to adjust for that under-coverage is always a challenge.”
“On questions of identity, official statistics are always going to follow how society responds.
For example, if it’s not common to ask people about their sexual identity for other reasons,
it’s very difficult to do so in a statistical enquiry. Initially when we started measuring same-sex
partnerships, we did it indirectly.”

“Being a small country is a great asset. For balance of payments data, we collect
information from the invoices of the largest firms. Informal ways like newspapers are more
reliable to identify what new firms we should look at, which is an advantage over larger

“Official and research surveys have extraordinarily good response rates;
neighbourhood tracking is very strong here. The Dunedin and Christchurch studies that
follow people for a lifetime have been able to get world-leading response rates.”

Cook says that in almost every field “there has been a huge increase in the information
available from the public – such as state administrative records and welfare information.
It has created tremendous opportunities to understand a lot more about our society
without collecting information directly. We’re able to produce a lot of geographic information
about what’s going on in small areas.”


NZ Census

Animations from official statistics around the world