ITP Sites:   ITP Site|TechBlog|TechHub in schools|NZ CloudCode|All Tech Events|Software Escrow NZ

ITP Techblog

Brought to you by IT Professionals NZ
Menu
« Back to Home

Digital-first Census sees drop in responses

Paul Brislen, Editor. 11 July 2018, 11:03 am

The decision by Statistics NZ to delay releasing the latest Census data will no doubt fuel more skepticism around any move to encourage online voting.

User turnout at the latest census is down by 5% over previous years', and that's something of a blow to those proponents of a "digital first" approach.

Around 90% of the individual forms were filled out in 2018 compared with 94.5% in 2013. The 2011 census was cancelled because of the Canterbury earthquake, but the turn-out for earlier census' was far higher than this year's.

As a result, StatsNZ is using a "revised methodology" to compensate for the missing data - something that may skew the results, but which will certainly lead to a revised approach to the next Census, due to be held in five years' time.

According to the StatsNZ website, it is quite normal for some people to not fill in census forms, or for forms to have some unanswered questions. To compensate for this, the department uses "a statistical process called 'imputation' to improve the quality of census data. Imputation involves inserting a value when a respondent has not provided a valid response".

"Given the interim position of individual response rates for the 2018 Census, we are looking at expanding our imputation approach. We are investigating how we can impute households, and cases of item non-response. Both item and unit imputation will improve data coverage and, occasionally, data quality, but not for all census variables. If we do not impute, there will be large amounts of missing data that will affect the overall quality of the dataset."

The question of whether the engagement rate is reduced because of the digital-first approach is yet to be answered, but it is sure to reignite concerns about any move to online voting, particularly in elections that typically have lower turnout rates.

While general election turnout rates in New Zealand remain quite high by international standards, there is considerable variation between age groups, with those over the age of 50 far more likely to turn out than those under 50. Māori voting has increased in recent years (up by 3.5% between 2014 and 2017) but is at a much lower rate than non-Māori voting.

Any move that could reduce turnout further is likely to be considered a backward step. Of course, until we know for certain whether the move to a digital-first Census is at the root cause of the decline or not, this all remains academic.


Comments

You must be logged in in order to post comments. Log In


Web Development by The Logic Studio