PDF Data Dictionary

Data Collection

2016 HES (Expenditure) data collection

2016 HES (Expenditure) data collection



HES (Expenditure) data collection

In all cases, contact with the selected households is made through personal visits by interviewers. The number of eligible households on the panel list is the target number of respondents for the selected area. Thus, the aim of the data collection operation is to obtain completed documents from as many eligible addresses as the financial and time constraints of the survey allow. On average this is four eligible responding households per panel.

When a household cannot be contacted on the first visit, the interviewer makes at least two further visits at different times of the day in an effort to establish contact with the household. If, after the third visit, the household has still not been contacted then the household is a non-respondent. If an address contains more than one household, the interviewer randomly selects and surveys one household.

Each household is interviewed and all those 15 years and over are asked to keep an expenditure diary for the following two weeks.

Proxy rules for interviewers

A proxy may provide information in 'family type' households where:

  • the whole household is informed about the survey. All agree to participate, but are not able to be present when the questionnaires are administered

  • children are away at boarding school

  • people don't work and have no source of income

  • people are elderly, sick or mentally incapacitated

In all proxy interviews, the interviewer must be convinced the proxy is totally familiar with the other respondent's information.

External influences at the time of collection

Changes in income and expenditure may be influenced by one-off real-world events. Events that could have influenced the HES (expenditure) 2015/16 data are the:

  • increase in the adult minimum wage from $14.75 to $15.25 (effective from 1 April 2016)

  • increase in government transfer maximum rates for people with dependent children of $25 for main benefits and student allowances (effective from 1 April 2016)

  • increase in New Zealand Superannuation rate of 2.73 percent (effective from 1 April 2016)

Recall period

The HES (expenditure) was carried out from the 1 July 2015 to 30 June 2016, therefore different households have different recall periods. Those households interviewed on the 1 July 2015 will have recall periods earlier than the survey date range, so from the 1 July 2914 to 30 June 2016.

Expenditure data is collected by the following methods:

  • 3-month recall for large or irregular expenditure types, such as health and travel

  • 12-month recall for housing-related costs and recreation and culture

  • latest payment (for regular commitments such as electricity, telephone, rates, rent, insurance and superannuation)

  • 14-day diary keeping for smaller, more regular expenditure types

Income has several recall periods, including 12-month recall and latest payment.

Under-reporting expenditure

For some types of expenditure, the estimated amount for all private households is less than expenditure reported from other data sources.

There are three main reasons for this difference:

  • expenditure by residents of non-private households, or by those ineligible for the survey (eg overseas visitors), is excluded from this survey

  • respondents to the survey forget or omit some types of purchases because they are unable to recall expenditure, or cannot refer to records at the time of the interview. Some of this has been reduced due to changes in the recall period

  • A bias associated with non-response affects some statistics.

We don't adjust the data to compensate for any under-reporting.

Response rate and description of calculation

The sample size is approximately 5,000 households.

The target response rate is 70 percent. We achieved 77.3 percent for the 2015/16 cycle(post-imputation).

We calculate the response rate by determining the weighted number of eligible households that responded to the survey as a proportion of the estimated weighted population.

Even though the response rate before imputation has been declining over time, and minimal bias is present, our non-response adjustment and the calibration to population benchmarks remove the impact of this bias.

Imputation method

Imputation replaces missing values with actual values from similar respondents. As a result of imputing records, the response rate for the year ending June 2016 improved from 75.1 percent to 77.3 percent.

Two imputation methods are used in HES (Expenditure) - nearest neighbour donor imputation and mean imputation (latter for expenditure only).

The nearest neighbour donor imputation method replaces missing values by data values from another record called a donor. A donor is selected by finding a respondent with matching characteristics to the recipient. Mean imputation uses the mean of the acceptance values to replace a missing value.

We introduced donor imputation into HES (expenditure) in 2009/10, and use it in all subsequent HES (expenditure) releases. Back data has been revised to apply imputation.

The donor imputation is applied to a household where the household does not supply all the required income or expenditure information, but supplies sufficient information to be retained in the sample.

For households where at least one significant person in the household has completed at least two modules out of three core ones (Job, Government transfers, and Investment) of the income questionnaire, we impute income questionnaires for other household member(s) who have not fully completed their income questionnaire(s). In HES (expenditure) years, we apply the same process when expenditure diaries are not supplied by all eligible members of the household. In addition, we impute age for respondents who do not provide an age.

We also impute local and regional council rates for respondents who have not provided enough information for us to calculate their rates. A form of manual imputation is used to impute interest rates.

Before imputation was introduced, we discarded households with one or more questionnaire(s) missing. With imputation, we recover some of these households.

Consistency with other periods of collection

Although we adjust survey results for various demographic variables, there can be variability in survey estimates from one survey collection period to the next. This variability is because a different group of households is selected for each survey. For example, in 2014/15 the sample was boosted to 8,000 households because it was a HES (savings) collection. Although weighting should adjust for sample size differences, there could be some variability between 2014/15 and 2015/16.

Sampling error

The table below shows the sampling errors for the level of expenditure by expenditure type in 2012/13 and 2015/16.

We advise care when interpreting expenditure estimates with sampling errors greater than 20 percent. They are less statistically reliable than estimates with sampling errors less than or equal to 20 percent.

Sampling errors for average weekly household expenditure, by expenditure type
Year ended June 2013 and 2016
Expenditure typeLevel sampling error (%)
Alcoholic beverages88
Clothing and footwear1210
Housing and household utilities68
Household contents and services76
Recreation and culture57
Miscellaneous goods and services55
Other expenditure911
Sales, trade-ins, and refunds2236
1. Data for this year is revised.Source: Stats NZ