The 2009-2010 Household Income and Expenditure Survey (HIES) is the third household expenditure survey conducted in Papua New Guinea (PNG), and is the second HIES conducted by the National Statistical Office (NSO) of PNG. The previous survey, the Household Survey of PNG, was conducted in 1996 by Unisearch and the Institute of National Affairs. The first household expenditure survey (the ‘HES’) was conducted in 1975-1976 by the NSO. It remains the source of the expenditure weights of the Consumer Price Index.
The 2009-2010 HIES collected information on key topics such as family demography, education, health, employment and consumption. The scope of 2009-2010 HIES was broader than the previous survey, with more indicators and more households surveyed, including questions on living standards and other related subject areas. One key goal was to collect data on income and expenditure, and to enable a rebasing of the Consumer Price Index (CPI). This will assist future policy decisions and analysis to be better based on reliable evidence, such that they can better support improvements in the living standards of Papua New Guinea’s people.
As the 2009-2010 HIES was a multi-topic survey, the following types of information were collected:
1. Household level information on housing characteristics, ownership of consumer durables, non- food consumption, access to various types of public services, and incidence and resolution of different types of disputes.
2. Person-level information on age, sex, education, health, employment status, receipt of remittances, and personal security. This was supplemented by anthropometric data for children aged six years or younger.
3. Personal record of all food and nonfood purchases for 14 consecutive days for all household.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Version 02: There was an issue with the initial weight that only yield population as 5.047 million instead of 6.8-6.9 millions. To solve the issue, post_stratified weight (weight_ps) has been recalculated. The correct weight (weight_ps) is provided in weights_new dataset.
The 2009-10 Papua New Guinea Household Income and Expenditure Survey covered the following topics:
General services available in the village
Health services: location and accessibility
Self-employment and family business activities
Main meals at this household
Employment & labour
Producers and sponsors
National Statistical Office
Government of Papua New Guinea
The World Bank
Government of Papua New Guinea
The World Bank
United Nations Children's Fund
Australian Aid for International Development
PNG is the largest nation in the South Pacific in both land area and population. It is comprised of around 600 islands, and the interior of the country is mountainous. Administratively, the country is divided into 22 provinces, within four geographic regions:
Southern Region comprising of the following provinces: Western, Gulf, Central, Milne Bay, Northern Oro, and the National Capital District
Highlands Region comprising of the following provinces: Southern Highlands, Enga, Western Highlands, Chimbu, Eastern Highlands, Hela, and Jiwaka
Momase Region comprising of the following provinces: Moroboe, Madang, East Sepik, and West Sepik
Island Region comprising of Manus, New Ireland, East New Britain, West New Britain, and the Autonomous Region of Bougainville.
A two-stage stratified cluster sample design was used in order to ensure independent estimates for the "rural", "urban", and "metro" areas of each of these regions. "Metro" denotes the two large urban centres of Port Moresby in Southern region and Lae in the Momase region. All the other urban areas in all regions were included in the "urban" stratum. The sample was thus divided into 10 strata. Sample size in each cluster was set to ensure reliable strata-level representative estimates; in addition, the sample size in the metropolitan areas was set large enough to support the development of a new CPI expenditure basket. In each stratum, households were chosen in two stages. In the first stage, using the sampling frame of the 2000 Census, clusters (or Primary Sampling Units) (PSU) were chosen using probability proportional to size. A full listing of households was done in all chosen clusters and a pre-set number of households were then selected randomly from this list. In the non-metropolitan areas, 18 households were chosen per cluster whereas in the metropolitan areas, 6 households were chosen per PSU. Replacement census units or replacement households in each cluster are used in the case of emergencies or when selected households could not be interviewed because of refusal, absence, illness in the family etc, were also selected using the same methodology.
For the 2009-2010 HIES sample, the selection probabilities and raising factors are determined in accordance with the sample design described above. The probability of selecting a PSU(ij) in stratum i is
p(ij) = m(i)n(ij)/n(i)
where n(ij) is the number of households in the PSU (as reported by the 2000 Census), ni is the total number of households in the stratum (also as per the 2000 Census) and mi is the number of clusters selected in the stratum.
The probability of selecting household ijk in PSUij of stratum i (in non-metro areas) is
p(ijk) = p(ij) (18/n’(ij))
where n’(ij) is the number of households in the PSU, following the household listing operation. The raising factor or weight w(ijk) for household ijk is the inverse of the selection probability p(ijk).
Dates of Data Collection
Data Collection Mode
Data Collection Notes
Training and Fieldwork
Because of the complexity of the survey, training was conducted for a full month (end of February-end of March 2009) to allow interviewers, supervisors, and data entry operators to master techniques and processes. The training covered all aspects of survey operations including household listing, interviewing techniques, good understanding of all sections of the questionnaire, and data entry. After classroom tests and field practical tests, 100 qualified persons were selected and allocated into 20 teams. Two additional training sessions and refreshers were conducted over the course of the survey for newly recruited team members.
Data collection was conducted simultaneously in all provinces from July 2009 to January 2011. The survey was initially planned to be spread over 12 consecutive months to account for seasonality. However, because of funding delays, it took almost 18 months for survey completion.
The HIES is a complex application with a hierarchical set of questionnaires. For example, the main questionnaire consists of a household roster and other household information, and there are separate questionnaires for eligible members in the household. The data entry application may then contain two levels-one for the household and one for each eligible member in the household. The set of forms corresponding to the household, make up level one. The set of forms corresponding to each eligible member make up level two. Each case would consist of a level one and a variable number of level occurrences for level two. Most applications consist of a single level.
The survey questionnaires contain the following FORMS:
1) From A: Household Control Form
2) Form B: Household Schedule Form
3) Form C: Personal Schedule
4) Form D: Personal Diary
5) Form E: Personal Notepad
The use of the datasets must be acknowledged using a citation which would include:
- the identification of the Primary Investigator (including country name)
- the full title of the survey and its acronym (when available), and the year(s) of implementation
- the survey reference number
- the source and date of download (for datasets disseminated online)
Papua New Guinea National Statistical Office. Household Income and Expenditure Survey (HIES) 2009-2010. Ref. PNG_2009_HIES_v01_M. Dataset downloaded from [URL] on [date].
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.