New Zealand Childcare Survey 2009 Data Collection

Background to the survey

The New Zealand Childcare Survey 2009 (CCS) provides information about the use of formal and informal childcare arrangements and the relationship between childcare, work, and study arrangements.

The objective of the CCS is to provide statistical data to assist in measuring childcare arrangement use, barriers to, or difficulties experienced with childcare, and the relationship between childcare and participation in paid employment.

The CCS was a supplement to the Household Labour Force Survey (HLFS) in the September 2009 quarter.

Survey scope

The target population for the CCS is the civilian usually resident, non-institutionalised population who are either children aged 0–13 years or a person who is in the parent role for at least one child aged 0–13. This means that the statistics in this release do not cover long-term residents of retirement homes; hospitals and psychiatric institutions; inmates of penal institutions; members of the New Zealand permanent armed forces; members of the non-New Zealand armed forces stationed in New Zealand and their dependents; overseas diplomats; and overseas visitors who expect to be resident in New Zealand for less than 12 months.

The survey population for individuals is the target population with the following exclusions: residents of islands other than the North Island, South Island, and Waiheke Island; people residing in non-private dwellings; people not living in permanent dwellings; people temporarily staying in non-private dwellings when contact is attempted at their private dwellings; and people temporarily overseas when contact is attempted.

The survey population for households is households in New Zealand with at least one child aged 0–13 years and an individual who is a parent or in the parent role of children in the survey population for the childcare survey.

Achieved sample

A total of 3,656 households participated in the CCS, consisting of 6,326 parents, 1,341 pre-school children aged 0–5 years, and 2,315 school children aged 5–13.

Data collection

Data collection for the CCS was done using computer-assisted interviewing (CAI). Data was collected partly by computer-assisted personal interviewing (CAPI) for selected households (approximately 30 percent of all survey respondents). The remaining households were surveyed by centralised computer-assisted telephone interviewing (CATI). The survey was carried out from 5 July to 11 July 2009 and from 26 July to 3 October 2009. There were no interviews during the period 12 July to 25 July 2009 as the reference weeks fell on school holiday periods.

Estimation / weighting

A basic survey weight is attached to each record to reflect the probability of that unit being included in the sample. For the HLFS, two types of adjustments are then applied to the basic survey weights to improve the reliability of the survey estimates. The basic weights are first adjusted for non-response, and are then further adjusted to ensure that estimates of the relevant population characteristics match known population totals.

For the CCS there are two populations, children aged 0–13 years (children) and parents of children aged 0–13 years (parents), both inclusive.

One child in a household was randomly selected to participate in the childcare survey. The basic weight of the child was adjusted to reflect the probability of selection in the CCS. It was further adjusted to ensure that estimates of the child population matched known population totals.

For the parents, the final weight, as determined by the HLFS, was adjusted to reflect the probability of selection in the CCS. It was further adjusted to ensure that estimates of the relevant population characteristics matched estimates of the number of parents of children aged 0–13 years inclusive.

Reliability of survey estimates

Two types of errors are possible in estimates based on a sample survey: sampling error and non-sampling error. Sampling error can be measured, and quantifies the variability that occurs by chance because a sample rather than an entire population is surveyed. Non-sampling errors are all errors that are not sampling errors and are not quantifiable. Non-sampling errors include unintentional mistakes by respondents when answering questions, variation in the respondent's and interviewer's interpretation of the questions asked, and errors in the recording and coding of data. Statistics New Zealand endeavours to minimise the impact of these errors through the application of best survey practices and monitoring of known indicators (eg non-response).

Sampling errors have been estimated using a jackknife method, which is based on the variation between estimates, based on different subsamples taken from the whole sample. This is an attempt to see how estimates would vary if we were to repeat the survey with new samples of individuals. For example, the estimated total number of children with no childcare arrangements is 369,400. This estimate is subject to a sampling error of plus or minus 15,500 or 4.2 percent (measured at the 95 percent confidence level). This means that there is a 95 percent chance that the true number of children with no childcare arrangements lies between 353,900 and 384,900.

Smaller estimates, such as the total number of children attending kōhanga reo (7,400), are subject to larger relative sampling errors than larger estimates. This estimate is subject to a sampling error of plus or minus 3,000 or 40.4 percent (measured at the 95 percent confidence level). This means that there is a 95 percent chance that the true number of children attending kōhanga reo lies between 4,400 and 10,400.

The following table contains a guide to the likely sampling errors of estimates of different sizes. This table can be used by finding the closest figure to the estimate of interest in the left-hand column of the table and reading off the corresponding relative sampling error in the right-hand column. For example, a total estimate of 35,600 school-age children would have a relative sampling error of about 18 percent. Estimates of less than 5,000 within the output tables are likely to have relative sampling errors between 50 and 130 percent and, hence, these estimates should be used with caution.

Suppressed estimates in this release

All estimates provided in the output tables have a relative sampling error (measured at the 95 percent confidence interval) of less than 100 percent. Some estimates have been suppressed (replaced by 'S' in the tables) for reliability and confidentiality reasons. These suppressed estimates had a relative sampling error of 100 percent or more and/or reflect a low number of responses (weighted count of 1,000 or less).

Response rate

The target response rate for the CCS was 80 percent. This represents the minimum acceptable by Statistics NZ. The achieved response rate for the number of parents or those in a parent role was 82.4 percent.

Non-response to the CCS was partly due to the increased burden of it being a supplement to the HLFS and partly because proxy responses were not accepted in all situations (even though they were accepted for the HLFS). A proxy response is a response by one member of a household on behalf of another.

Editing and imputation

A minimal approach to editing was implemented for the CCS. With CAI, the computer software runs checks for validity and consistency as responses to questions are captured. If required, the software prompts the interviewer to clarify answers with the respondent at the time of interview. This keeps the number of inconsistent answers low. A further round of validity, logic, and error checks was performed on the data as part of the data processing stage.

Non-response to the CCS occurred either when an eligible individual in the sample did not respond to the questionnaire or did not provide enough information to determine a response. Any non-responding parents from the HLFS sample were dealt with by adjusting the weights of the responding individuals.

Item (or partial) non-response could occur within the responses for the CCS. This includes a response of 'don't know' or 'refused' to non-core questions. Since all the core CCS questions had been answered, the record was deemed to be a full response. No imputation has been applied to any item non-response.

If a response to age, sex, or full-time/part-time status was missing in the HLFS then it will have been imputed by the HLFS and used in the CCS. There is no other imputation for variables collected in the CCS.

Mean (average) and median

The mean or average is calculated as the total divided by the number of units in the population. A mean can be sensitive to extreme values. Unusually high or low values will have a large impact on the estimate of the mean.

The median is the value at which half of the units in the population have lower values and half have higher when all values have been ordered from highest to lowest. It corresponds to the 50th percentile. The median is less sensitive to extreme values than the mean.

Percentages

Percentages in this release are calculated excluding those who did not specify a response. For example, information on hours spent in informal care was provided for 131,600 of the 136,300 pre-school children attending informal ECE and care. The percentage of those who attended for 10 hours or less is therefore calculated as 69,000 divided by 131,600 (52.4 percent), not as 69,000 divided by 136,300 (50.6 percent).

Rounding

All estimates provided in this release have been independently rounded to the nearest hundred. For this reason estimated totals may differ from the sum of individual cells. All percentages have been calculated using unrounded figures and have been rounded to one decimal place in this release.

Survey content and structure

The CCS questionnaire contained the following three modules:

1.Early childhood education (ECE) module

2.Out-of-school services (OSS) module

3.Parent module.

Early childhood education (ECE) / out-of-school services (OSS)

For each household where there was more than one child aged 0–13 years present, one child was randomly selected by the computer. Information about this child was then collected in either the ECE or OSS module, based on that child's age. If there was only one child in the household then the information was collected about that child.

Where the selected child had more than one eligible parent in the household (as determined by the HLFS) a question was asked as to which parent could talk about the child's care arrangements. Where there was no obvious parent relationship to the selected child, the questionnaire asked which of the adults in the household (those aged 15 or over) was able to talk about the child's care arrangements.

Children aged four years and under were automatically asked the ECE module, while those aged seven years or older were automatically asked the OSS module. For children aged five or six years, the parent was asked whether the child attended school last week. If the child did attend school they were asked the OSS module otherwise they were asked the ECE module.

The ECE and OSS modules both contained questions on formal and informal care arrangements for the selected child in the household. The following information was collected:

• types of childcare arrangements (formal and informal) parents use for the child

• days and hours spent by child in each type of care arrangement

• cost of childcare used

• reasons for not using formal care if none was used

• use of government subsidies to assist with childcare

• reasons for not using government subsidies to assist with childcare when not used.

In addition, for those who completed the OSS module, information was asked about care arrangements used during the last school holidays. For those who were interviewed in the week starting 5 July 2009 these questions related to the April school holidays; for all other respondents these questions related to the July school holidays. Information collected included types of arrangements used, cost of arrangements, hours spent in school holiday care, and reasons for not using a school holiday programme when applicable.

Parent information

Questions in the parent module were asked about the parent who had provided information about the childcare arrangements and their partner (where applicable). The same parent who completed the ECE or OSS module answered on behalf of themselves and their partner. The following information was collected:

• work arrangements used in the week prior to interview

• work arrangements used in the week prior to interview specifically to look after a child

• difficulties experienced (if any) getting childcare in previous 12 months when working or wanting work

• work consequences experienced (if any) by those who had experienced difficulties

• difficulties experienced (if any) getting childcare in previous 12 months when studying

• study consequences experienced (if any) by those who had experienced difficulties

• types of care used for any other children in the household that person is parent to (including use of government subsidies)

• receipt of government benefits

• income range (parent and partner combined where applicable).