SPC_PNG_2006_DHS_v01_M
Demographic and Health Survey 2006
| Name | Country code |
|---|---|
| Papua New Guinea | PNG |
Demographic and Health Survey, Round 2 [hh/dhs-2]
The 2006 PNG DHS is part of the world wide DHS which are designed to collect, analyze, and disseminate data on fertility, family planning, maternal and child health, HIV/AIDS and other health related issues using internationally recognized sets of questionnaires and sampling procedures.
The 2006 Demographic and Health Survey (DHS) is the second in a series of this type of survey conducted successfully in Papua New Guinea since 1996. The 2006 DHS for the first time included a male questionnaire and questions on well being and sexual risk behaviour.
The primary objective of the 2006 DHS is to provide to the Department of Health (DOH), Department of National Planning and Monitoring (DNPM) and other relevant institutions and users with updated and reliable data on infant and child mortality, fertility preferences, family planning behavior, maternal mortality, utilization of maternal and child health services, knowledge of HIV/AIDS and behavior, sexually risk behavior and information on the general household amenities. This information contributes to policy planning, monitoring, and program evaluation for development at all levels of government particularly at the national and provincial levels. The information will also be used to assess the performance of government development interventions aimed at addressing the targets set out under the MDG and MTDS. The long-term objective of the survey is to technically strengthen the capacity of the NSO in conducting and analyzing the results of future surveys.
The successful conduct and completion of this survey is a result of the combined effort of individuals and institutions particularly in their participation and cooperation in the Users Advisory Committee (UAC) and the National Steering Committee (NSC) in the different phases of the survey.
The survey was conducted by the Population and Social Statistics Division of the National Statistical Office of PNG. The 2006 DHS was jointly funded by the Government of PNG and Donor Partners through ADB while technical assistance was provided by International Consultants and NSO Philippines.
Sample survey data [ssd]
Version 2 (October 2009) - Public use
2009-10
Version 0 - data set after data entry
Version 1 - data set after data editing & cleaning
Version 2 - final anonymized data set for public use
The scope of the Dempgraphic and Health Survey 2006 includes:
| Topic | Vocabulary |
|---|---|
| Education | World Bank |
| HIV/AIDS | World Bank |
| Population & Reproductive Health | World Bank |
| Malaria | World Bank |
National level
Regional level
Urban and Rural
The survey covered all de jure household members (usual residents), all women and men aged 15-50 years resident in the household.
| Name | Affiliation |
|---|---|
| National Statistics Office | Department of Planning & Implementation |
| Name | Role |
|---|---|
| Asian Devlopment Bank | Funding & technical assistance |
| Australian Aid for International Development | Funding & technical assistance |
| New Zealand Aid | Funding & technical assistance |
| United Nations Children's Fund | Funding & technical assistance |
| United Nations Population Fund | Funding & technical assistance |
| Name | Affiliation | Role |
|---|---|---|
| Snowy Mountains Engineering Corporation Ltd | Technical assistance on Project managerial support | |
| Dr. Socorro Abejo | Philippines National Statistics Office | Technical assistance on Data analysis |
| Noel Perez | Philippines National Statistics Office | Technical assistance on Data processing |
The primary focus of the 2006 DHS is to provide estimates of key population and health indicators at the national level. A secondary but important priority is to also provide estimates at the regional level, and for urban and rural areas respectively. The 2006 DHS employed the same survey methodology used in the 1996 DHS. The 2006 DHS sample was a two stage self-weighting systematic cluster sample of regions with the first stage being at the census unit level and the second stage at the household level. The 2000 Census frame comprised of a list of census units was used to select the sample of 10,000 households for the 2006 DHS.
A total of 667 clusters were selected from the four regions. All census units were listed in a geographic order within their districts, and districts within each province and the sample was selected accordingly through the use of appropriate sampling fraction. The distribution of households according to urban-rural sectors was as follows:
8,000 households were allocated to the rural areas of PNG. The proportional allocation was used to allocate the first 4,000 households to regions based on projected citizen household population in 2006. The other 4,000 households were allocated equally across all four regions to ensure that each region have sufficient sample for regional level analysis.
2,000 households were allocated to the urban areas of PNG using proportional allocation based on the 2006 projected urban citizen population. This allocation was to ensure that the most accurate estimates for urban areas are obtained at the national level.
All households in the selected census units were listed in a separate field operation from June to July 2006. From the list of households, 16 households were selected in the rural census units and 12 in the urban census units using systematic sampling. All women and men age 15-50 years who were either usual residents of the selected households or visitors present in the household on the night before the survey were eligible to be interviewed. Further information on the survey design is contained in Appendix A of the survey report.
Table A.2 of the survey report provides a summary of the sample implementation of the 2006 DHS. Despite the recency of the household listing, approximately 7 per cent of households could not be contacted due to prolonged absence or because their dwellings were vacant or had been destroyed. Among the households contacted, a response rate of 97 per cent was achieved. Within the 9,017 households successfully interviewed, a total of 11, 456 women and 11, 463 of men age 15-49 years were eligible to be interviewed. Successful interviews were conducted with 90 per cent of eligible women (10, 353) and 88 per cent of eligible men (10,077). The most common cause of non-response was absence (5 per cent). Among the regions, the rate of success among women was highest in all the regions (92 per cent each) except for Momase region at 86 per cent. The rate of success among men was highest in Highlands and Islands region and lowest in Momase region. The overall response rate, calculated as the product of the household and female individual response rate (.97*.90) was 87 per cent.
The sample design was self-weighting. This means that raising factors of the sample units are all equal.
Three questionnaires were used in the 2006 DHS namely; the Household Questionnaire (HHQ), the Female Individual Questionnaire (FIQ) and the Male Individual Questionnaire (MIQ). The planning and development of these questionnaires involved close consultation with the UAC members comprising of the following line departments and agencies namely; Department of Health (DOH), Department of Education (DOE), Department of National Planning and Monitoring (DNPM), National Aids Council Secretariat (NACS), Department of Agriculture and Livestock (DAL), Department of Labour and Employment (DLE), University of Papua New Guinea (UPNG), National Research Institute (NRI) and representatives from Development partners.
The HHQ was designed to collect background information for all members of the selected households. This information was used to identify eligible female and male respondents for the respective individual questionnaires. Additional information on household amenities and services, and malaria prevention was also collected.
The FIQ contains questions on respondents background, including marriage and polygyny; birth history, maternal and child health, knowledge and use of contraception, fertility preferences, HIV/AIDS including new modules on sexual risk behaviour and attitudes to issues of well being. All females age 15-50 years identified from the HHQ were eligible for interview using this questionnaire.
The MIQ collected almost the same information as in the FIQ except for birth history. All males age 15-50 years identified from the HHQ were eligible to be interviewed using the MIQ.
Two pre-tests were carried out aimed at testing the flow of the existing and new questions and the administering of the MIQ between March and April 2006. The final questionnaires contained all the modules used in the 1996 DHS including new modules on malaria prevention, sexual risk behaviour and attitudes to issues of well being.
| Start | End |
|---|---|
| 2006-10 | 2007-01 |
| Name | Affiliation |
|---|---|
| National Statistical Office | Department of Planning and Implementation |
Training for the 2006 DHS was conducted in two phases. The first phase involved the training of master trainers from NSO in Port Moresby in August 2006. The second phase involved training of interviewers, supervisors and quality controllers in the four regional training centres in September 2006.
Data collection began simultaneously in all the provinces in October 2006 and was completed in January 2007. In total, there were more than 230 people involved in the data collection phase. Field monitoring visits were conducted through field visits to the teams by the NSO Senior Management Team. Regular communication was maintained with the teams throughout the duration of the survey.
All questionnaires from the field were sent to the NSO headquarters in Port Moresby in February 2007 for editing and coding, data entry and data cleaning. Editing was done in 3 stages to enable the creation of clean data files for each province from which the tabulations were generated. Data entry and processing were done using the CSPro software and was completed by October 2008.
Appendix B of the survey report describes the general procedure in the computation of sampling errors of the sample survey estimates generated. It basically follows the procedure adopted in most Demographic and Health Surveys.
Appendix C explains to the data users the quality of the 2006 DHS. Non-sampling errors are those that occur in surveys and censuses through the following causes:
a) Failure to locate the selected household
b) Mistakes in the way questions were asked
c) Misunderstanding by the interviewer or respondent
d) Coding errors
e) Data entry errors, etc.
Total eradication of non-sampling errors is impossible however great measures were taken to minimize them as much as possible. These measures included:
a) Careful questionnaire design
b) Pretesting of survey instruments to guarantee their functionality
c) A month of interviewers’ and supervisors’ training
d) Careful fieldwork supervision including field visits by NSOHQ personnel
e) A swift data processing prior to data entry
f ) The use of interactive data entry software to minimize errors
| Name | Affiliation | |
|---|---|---|
| National Statistician | National Statistics Office | jaka@nso.gov.pg |
| Is signing of a confidentiality declaration required? | Confidentiality declaration text |
|---|---|
| yes |
Confidentiality of the respondents' information is governed by the Statistical Services Act 1980 of the Independent State of Papua New Guinea, particularly: Section 31. INFORMATION SUPPLIED TO BE CONFIDENTIAL AND USED FOR STATISTICAL PURPOSES ONLY. (1) Subject to Subsections (2) and (3)– (a) information required to be supplied under this Act shall be used for statistical purposes only; and (b) notwithstanding anything in any other Act, a person, other than a person employed under this Act who has made the declaration of secrecy specified in Form 1, shall not be permitted to see any data relating to an individual respondent which has been collected under this Act except for the purposes of proceedings under this Act or to the release of information collected under Section 25; and (c) it shall be a principle to be followed in the publication of statistics to arrange statistical tables in such a manner as to prevent any particulars published in the tables from being identifiable by any person (other than the person by whom the particulars were supplied) as particulars relating to any particular person or undertaking. SECTION 32. INFORMATION PRIVILEGED. (1) Subject to Subsection (3), any data collected under this Act or a copy of such data in the possession of a respondent shall not be used as evidence in any proceedings in a court, or in connection with, or as evidence of, an assessment or levy of any tax imposed by the State. (2) Subject to Subsection (3), a person employed under this Act shall not be required to divulge any data collected under this Act to a court or other person having power to summon witnesses whether by producing a form or document or otherwise. |
The dataset has been anonymized and is available as a Public Use Dataset. It is accessible to all for statistical and research purposes only, under the following terms and conditions:
"National Statistical Office of Papua New Guinea, Dempgraphic and Health Survey 2006 (DHS 2006), Version 2 of the public use dataset (July 2012), provided by Pacific Data Catalog, www.spc.int/prism/data-catalog"
The user of the data acknowledges that the PNG NSO, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
(c) 2009 National Statistical Office of Papua New Guinea
| Name | Affiliation | URL | |
|---|---|---|---|
| Publication & Library | National Statistics Office | mpulah@nso.gov.pg | www.nso.gov.pg |
DDI_SPC_PNG_2006_DHS_v01_M
| Name | Affiliation | Role |
|---|---|---|
| Mathew Pulah | PNG NSO | Documentation of survey |
| Leslie Wongo | PNG NSO | Documentation of survey |
2012-07-09
Version 01 (July 2012) - First documentation of DHS 2006 using IHSN Toolkit