133 Molesworth Street
PO Box 5013
Wellington 6140
New Zealand
T+64 4 496 2000
16 September 2022
P Shaw
By email: [FYI request #20225 email]
Ref: H2022010220
Tēnā koe P Shaw,
Response to your request for official information
Thank you for your request under the Official Information Act 1982 (the Act) to Manatū Hauora
(Ministry of Health) on 15 August 2022. You requested the below documents:
“1. Full report titled 'Attitudes to self-isolation', and dated February 2022.
2. Full report titled 'Health seeking behaviours', and dated March 2022.”
Please find the requested reports and their appendices attached. Some identifiable information
has been withheld under section 9(2)(a) of the Act to protect the privacy of natural persons. I
have considered the countervailing public interest in release in making this decision and
consider that it does not outweigh the need to withhold at this time.
Please consider the following when interpreting the results from Appendix 5 from the February
2022 report and Appendix 4 from the March 2022 report.
The data tables are developed as part of the basic analytical framework that helps Manatū
Hauora to look at the results in a meaningful way. They show the proportion of respondents who
answered each option of a question and the total number of respondents who answered the
question.
Please note that the proportion is weighted. This means that if 100 respondents answered the
question and the results showed 75% said yes, it does not mean that 75 respondents said yes.
Weighting is a statistical technique to minimise the effect of imbalances in the sample, so it is
more representative of Aotearoa New Zealand. Hence, weighting is an adjustment of the
percentage of respondents who said yes, to better reflect the population percentage. Please
see the appendix on methods in the report for the variables used for weighting.
The tables also include a breakdown by several variables, such as age groups or gender.
Please note that the percentages shown are also weighted and the number of actual
respondents is shown below each sub-group. We are careful with how we interpret the data, to
make sure we share accurate results.
A note of caution when looking at sub-groups is that many have very few respondents who are
part of that sub-group. Numbers of respondents in each sub-group are reduced for some
questions, as not everyone answers every question or the number of respondents in the survey
that are part of the sub-group is very small.
In such a situation, we would recommend that you ignore these small sub-groups as on their
own, they are imprecise results with large ‘errors.’ ‘Error’ is a technical term that refers to the
estimated difference of the percentage obtained from the survey to the ‘true’ percentage in the
population. It is something that can be estimated in statistics.
There is a relationship between the number of respondents and the size of the error. Table 1
describes this negative relationship – as the number of respondents increase the size of the
error decrease.
You may read the table in the following way. When the number of respondents for whole survey
or sub-group of respondents is 385, the true percentage is potentially 5% higher or 5% lower
than the percentage shown in the table at a 95% confidence level (e.g., a result of 75% would
mean that we are 95% confident that the true percentage lies between 70-80%).
Table 1: Relationship between the number of respondents in a survey or sub-group and the size
of the error.
Size of the error
±1%
±5%
±10%
±20%
Number of respondents
9604
385
97
25
Besides error there are other things that could influence the results, such as bias. Bias is also a
potential difference of the percentage obtained from the survey to the true percentage in the
population. Bias, however, differs from error in that it cannot be estimated in statistics, it can
sometimes be known which direction the bias occurs. It can be a positive, negative, or
bidirectional. In this study, there is a potential bias due to the survey being conducted online
and not everyone in Aotearoa New Zealand has access to or ability to use the internet. We
know neither the percentage difference of any question, option or sub-group nor the direction of
bias due to the study’s recruitment methodology.
Respondents complete this survey without the help of an interviewer, as such there is a chance
that they might not understand what we are asking. It is also possible for respondents to
misinterpret the question or not answer some questions and these have impacts on the results
as well, the size of which is unknown. Sometimes respondents answer in ways we do not
expect, for instance, they may answer what they think we want them to say, rather than what
they genuinely think or do. We have made all attempts to minimize these issues, but we cannot
rule out the possibility that they may occur.
A survey is a common tool used in health research and we have applied common techniques to
improve the accuracy of the findings. This guide has been written to help you think through and
identify numbers and tables that cannot be interpreted easily.
These notes are generic and should be read together with the technical notes that specify the
context of each survey.
Please see NZ Statistics for more information:
https://www.stats.govt.nz/methods/2013-census-
definitions-and-forms/.
Page 2 of 4
I trust this information fulfils your request. Under section 28(3) of the Act, you have the right to
ask the Ombudsman to review any decisions made under this request. The Ombudsman may
be contacted by email at:
[email address] or by calling 0800 802 602.
Please note that this response, with your personal details removed, may be published on the
Manatū Hauora website at:
www.health.govt.nz/about-ministry/information-releases/responses-
official-information-act-requests. Nāku noa, nā
Dan Bernal
Acting Group Leader, Intelligence, Surveillance and Knowledge Public Health Agency, Ministry of Health
Page 3 of 4
Appendix 1: List of documents for release
#
Date
Document details
Decision on release
1
February 2022
Horizon Self-isolation Survey
Some information withheld
Report
under section 9(2)(a) of the
Act to protect the privacy of
1A.
Excel Spreadsheet: Appendix 5
natural persons.
2
March 2022
Horizon Health Seeking
Behaviours Report
2A.
Excel Spreadsheet: Appendix 4
Page 4 of 4
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 1
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL
Document 2
Act 1982
INFORMATION
RELEASED UNDER THE OFFICIAL