How well is Oranga Tamariki supporting caregivers?

Methodology

Target population
To be eligible to be sampled for the survey a caregiver must have had a child currently in their care or have had a child in their care within the last 24 months. Oranga Tamariki had around 3,000 caregivers who meet this eligibility criteria.

Sampling frame
A random sample was drawn from the list of eligible caregivers each fieldwork period. All caregivers were given an opportunity to take part in the survey. For March 2022 and June 2022 around one quarter of the eligible caregiver population was sampled each period. For September 2022 the remaining half of the caregiver population was surveyed.

Sample size and response rates
Table A1 provides a summary of the sample sizes and response rates for the 2021 survey and the three quarters of 2022. A response rate of around 30% is generally what might be expected for a voluntary online survey. When looking at ethnicity, and region, the respondents appear to be representative of the full population of caregivers. However, responses were not representative when looking at caregiver type (whānau & non-whānau), non-whānau caregivers were more likely to respond than whānau caregivers, non-whānau caregivers had a response rate of 33% (359 responses) and whānau caregivers had a response rate of 18% (392 responses).

Classifications
The main classifications used for analysis within this report are broken down by caregiver type (whānau/non-whānau) and ethnicity (Māori/non-Māori). These classifications are based on self-reported data from questions within the survey which asked whether the caregiver was a whānau caregiver and their ethnicity.

Weighting
At the end of the year the survey data were weighted to ensure that the sample was representative of the population of caregivers in the Oranga Tamariki caregiver database in terms of their relationship to their Oranga Tamariki child (whānau/nonwhānau) and ethnicity (Māori/non-Māori). Total population numbers were provided by Oranga Tamariki to enable the weighting.

To construct the weighting matrix for 2022, two ethnicity groupings were used (Māori/non-Māori). Responses were assigned to one of these groups based on selfdefined ethnicity. Relationship to the child (whānau/non-whānau) is also drawn from the survey response. The quarterly datasets and reports were not weighted. Weights were only applied to annual data.

Margin of error
After adjusting for the design effects of weighting the data, the maximum Margin of Error at the 95% confidence estimate for the achieved sample of n=705 caregivers is +/- 3.3%.

Comparisons to the 2021 survey
Caution needs to be taken when comparing the results from the 2022 surveys to the survey conducted in 2021. This is because much has changed in the survey, including several questions, the methodology, and response rates.

Coding of open questions
The answers given to open ended questions were analysed for selected questions. Recurring themes formed the code frames, which were developed by NielsenIQ’s coding team and quality assured and modified by Oranga Tamariki staff. These themes built on code frames from previous iterations of the survey.

Definitions and guidance on survey completion

Caring for more than one child

When a caregiver has, or is, looking after multiple children, they were instructed to think about the child who has the next birthday and think about their experiences while caring for that child.

Child
For ease of reference throughout the survey ‘child’ was used to describe children/tamariki and young people/rangatahi of all ages. Children/tamariki and young people/rangatahi in care can be aged up to 18 years old. Young people/rangatahi can also chose to be with caregivers up to 21 years old.

Caregiver
We have used the term ‘caregiver’ to describe the many types of care arrangements people have caring for a child. These may include people caring for children in their own whānau or extended family.

Incentive
As an acknowledgement of the time taken to complete the survey, caregivers could choose to participate in the draw to win a Prezzy® card. Each quarter there was one $250 card and five $50 cards to be won.

0800 number
NielsenIQ operated an 0800 number for caregivers to call if they wanted more information about the survey or if they needed any help.

Data collection
The online surveys were programmed in Decipher (the online survey software NielsenIQ uses) and were device agnostic, so respondents saw the best layout for their computer/ tablet/ phone regardless of their device brand. As completed paper questionnaires were received at the Nielsen’s office, the data were entered manually. The data entry team had the ability to select ‘No response’ for any question where a paper respondent had not selected a response.

Data entry protocols were set up to ensure consistency between data entry team members. As part of Nielsen’s quality control processes, 10% of each team member’s data-entered surveys were validated by another coder.

Notes
Some questions were only asked of caregivers who had become a caregiver within the last 12 months and some only to caregivers who currently had a child in their care.

Throughout the report you will see ‘n=’ this provides the number of responses that were received for the question or response being discussed, this gives you context where percentages are used to be able to know the number of responses.

For the results shown in this report ‘don’t know,’ ‘not applicable,’ and ‘prefer not to say’ responses have been excluded.