August 01, 2018

BERL Data Newsletter - August 2018

The BERL data newsletter has been designed to provide commentary on data issues and news, including methodological changes, new measures, reviews, delays, and other technical issues and news.

This newsletter covers:

  • Delays in the Census data
  • Further changes in IDI

2018 Census update

Statistics New Zealand’s first outputs from the 2018 Census will now be released in March 2019, instead of being released in October 2018 as previously announced. The outputs released in March 2019 will be usual resident population counts and occupied dwelling counts at national, regional, local authority, and SA2 geographic levels.

The delay has been caused by a lower than expected response rate to the 2018 Census with only 90 percent of individuals and households providing full or partial responses. This is down from a 94 percent response rate to the 2013 Census. The lower response rate means that Statistics New Zealand will need to engage in a lengthier imputation phase in order to account for the missing responses.

A news article out on 8 August 2018 from the New Zealand Herald raised two issues. The first is that Census field staff were not able to hand out paper forms to households without internet access. And second, Census field workers were told to skip properties in hard to reach locations during the push after Census day to increase the response rate. Both of these factors throw up red flags about the quality of the census information for rural areas.

So what does this mean for users?

It means that custom data requests will not be processed until at least June 2019 or even later. In addition it means that currently there are unaddressed issues regarding Statistics New Zealand’s plans for ensuring the quality of data for smaller population groups, such as iwi populations, and for rural populations.

Given the delay in the release of the 2018 Census data, we are currently investigating how we can use other data sources in place of the 2018 Census data.

Changes to the IDI

Statistics New Zealand, with the June 2018 refresh of the IDI environment, has made the following changes:

  • Applications are now processed over a six-week batch processing cycle. This means that new IDI applications will need to wait until the start of the next cycle before entering the approval process. All new projects with a potential IDI element will need to take this new process into account when determining their timelines.
  • New data was added into the IDI for the Programme for International Assessment of Adult Competencies (PIAAC); GSS20186; HES 2016/17 and 2017/18; MOH Maternity and Pharmaceutical; Police; and Community Finance Initiative. The new data added into the IDI includes new variables added for existing datasets (MOH Maternity and Pharmaceutical), new years added onto existing datasets (GSS2016, HES 2016/17 and 2017/18), and completely new datasets (PIAAC, Police and Community Finance Initiative).

As a reference the diagram below shows all the datasets in the IDI as at June 2018.

Source: Statistics NZ