The Vanuatu Multiple Indicator Cluster Survey (MICS) conducted in 2007 was undertaken by the Ministry of Health, Government of the Republic of Vanuatu (GoV) with financial and technical support from United Nations Children's Fund (UNICEF) - Pacific. The major objectives of the survey are to provide up-to-date information for assessing the situation of children and women in Vanuatu and furnish data needed for monitoring progress towards goals established by the Millennium Development Goals (MDGs) and the goals of A World Fit for Children (WFFC) as a basis for future action and development of a monitoring and evaluation system for Vanuatu's Poverty Reduction Strategy and United Nations Development Assistance Framework (UNDAF). The survey covered a nationally representative sample of 2,632 households; 2,692 women respondents aged 15-49 years and 1634 under-five children. Data were collected through three questionnaires:
1) the Household Questionnaire, 2) the Individual Questionnaire for Women aged 15-49, and 3) the Questionnaire for Children under-five years of age.
Independent samples for each domain (6 provinces and 2 cities) made it equivalent to 8 separate surveys to produce valid estimates for each domain simultaneously. National, and urban and rural estimates are obtained by combining these provincial data. The fieldwork began in 01 November, 2007 and concluded in 20 December, 2007.
Kind of Data
Sample survey data [ssd]
Version 01 (edited/final)
The Vanuatu dataset is yet to be uploaded.
consumption/consumer behaviour [1.1]
economic conditions and indicators [1.2]
LABOUR AND EMPLOYMENT 
in-job training [3.2]
The sample was designed with the intention of providing reliable estimates for the key MICS indicators at the national level and also for urban and rural areas separately, as well as for the 6 Provinces of Malampa, Penama, Sanma, Shefa, Tafea and Torba. Port Vila under Shefa Province and Luganville under Sanma province are the two major cities considered as two domains under the urban stratum.
Unit of Analysis
Individual (including adult women and children aged 5 and below).
The survey covered the whole resident sedentary population, with the exception of homeless.
Producers and sponsors
Authoring entity/Primary investigators
Ministry of Health, Government of Vanuatu
Ministry of Health
Ministry of Health
Government of Vanuatu
Participation in data collection and analysis
Government of Vanuatu
United Nations Children Fund
Funding and technical assistance
Ministry of Health, Government of Vanuatu
Participation in data collection
Statistics Office, Government of Vanuatu
Participation in data collection
The sample for MICS Vanuatu - 2007 is a probability-based, stratified cluster sample of 3000 households. They were selected in 120 clusters, each of size 25 households. The sample was designed with the intention of providing reliable estimates for the key MICS indicators at the national level and also for urban and rural areas separately, as well as for the 6 Provinces of Malampa, Penama, Sanma, Shefa, Tafea and Torba. Port Vila under Shefa Province and Luganville under Sanma province are the two major cities considered as two domains under the urban stratum. The Shefa and Sanma provinces
mentioned here exclude these two cities of corresponding provinces and bear rural character. The entire areas of all other provinces are considered as rural. The sample was allocated to the provinces/cities and by urban-rural in an optimum fashion to secure enough sample cases in each domain for reliable estimates to be obtained. That is, independent samples for each domain (6 provinces and 2 cities) made it equivalent to 8 separate surveys to produce valid estimates for each domain simultaneously. National, and urban and rural estimates are obtained by combining these provincial data.
The sample frame was the enumeration areas (EA) that made up the 1999 Population Census of Vanuatu, which had been updated in the 2006 Agricultural Census. Primary sampling units, or PSUs, were defined as either single EA or combinations of EAs. Combining EA was necessary whenever an EA contained fewer than 25 households, because the cluster size to be interviewed was set at 25 households as mentioned above.
The sample was selected in two stages. The first stage consisted of first stratifying the PSUs by province and within-province by urban/rural in two provinces namely Shefa and Sanma and then selecting 120 PSUs with probability proportionate to size or pps. At the second-stage, a fixed sample size of exactly 25 households was selected from each PSU, using systematic, equal-probability sampling or epsem. Thus a total of 3000 households were selected (120 clusters times 25 households). A household was defined as "a group of people those are eating from the same pot". Sample sizes for six rural provinces are 300 households each, while 500 and 700 households for Luganville and Port Vila cities respectively. It is to be noted here that the cities of Port Vila and Luganville are the urban part of the Shefa and Sanma provinces respectively. Total areas of other provinces are considered as rural. The resulting sample was not theoretically self-weighing; and sample weights have been used to adjust for major variations among the provinces and urban/rural EA with regard to different estimates. Detail sampling plan and sample allocation is shown in Appendix A of the final report provided as external resource.
It can be also mentioned here that, every fourth households in each cluster were selected for a nutrition component of the survey, which was additional to MICS nutrition modules. Thus the sample size for the additional nutrition component was exactly one-fourth of the MICS sample size in each domain
and at national level.
Deviations from the Sample Design
There was no deviation from sample design.
Of the 2,963 households selected for the sample, 2,959 were found to be occupied. Among the occupied households, 2,632 were successfully interviewed with a household response rate of 89 percent. In the interviewed households, 3,261 eligible women (aged 15-49) were identified for interview and 2,692 were successfully interviewed, giving women response rate of 83 percent. Among the interviewed households, 1,741 under-five children were identified. Of them, mothers/caretakers of 1,634 children were successfully interviewed, yielding children response rate of 94 percent. The overall response rates of women and children were found to be 73 percent and 84 percent respectively.
The sample response rates vary to some extent by urban-rural areas and by provinces. Urban area shows higher response rate for household and children than those of rural area. However, there is little variation in women response rate between rural and urban area. The response rate varied widely among
the provinces/urban domains. The household response rate ranges from 71 percent in Sanma to as high as 99 percent in Port Vila. Women's response rate varies from 76 percent in Malampa to 89 percent in Penama and Sanma. Children’s response rate vary in a narrow range among the provinces/domains, with 90 percent in Luganville to 97 percent in Sanma and Torba.
Because the sample was stratified by region, it is not self-weighting. For reporting the national level results, sample weights were used. Variable hhweight is to be used to weigh records at the household level. Variables wmweight and chweight have been computed respectively to weigh records at the women and child level.
Dates of Data Collection (YYYY/MM/DD)
Mode of data collection
In each district a team of people was selected – one supervisor for the district, controllers (one controller per 5-6 interviewers) and interviewers (whose number depended on the number of clusters in the region). For conducting the fieldwork, 10 teams were established - each was composed of three to four people, two interviewers (health workers), one laboratory technician and one driver. The MICS Coordinator provided overall supervision.
Several levels of control system were imposed:
1. During the field work, the controllers from the Ministry of Health conducted the first level of control immediately after receiving questionnaires from the interviewers.
2. Controllers and supervisors from the Ministry of Health conducted a second level of control on a sample of 10% of households.
3. The third level of control was carried out by supervisors from the WHO, UNICEF and Ministry of Health of on a sample of 5% of households.
Type of Research Instrument
The 3 questionnaires for the [Country] MICS were based on the MICS Model Questionnaire with some modifications and additions. A household questionnaire was administered in each household, which collected various information on household members including sex, age, literacy, marital status, and orphanhood status. The household questionnaire also includes education, child labor, water and sanitation, and salt iodization modules.
In addition to a household questionnaire, questionnaires were administered in each household for women age 15-49 and children under age five. For children, the questionnaire was administered to the mother or caretaker of the child.
The questionnaire for women contains the following modules:
- Child mortality
- Tetanus toxoid
- Maternal and newborn health
- Contraceptive use
The questionnaire for children under age five includes modules on:
- Birth registration and early learning
- Vitamin A
- Care of Illness
The questionnaires were developed on the basis of the MICS-3 model questionnaires in English language and were translated into Vanuatu national language, Bislama, and back translation was done to ensure the accuracy of the translation. The questionnaires were pre-tested. Based on the results of the pre-test, modifications were made to the wording of the questions, the response categories, and the key words.
Ministry of Health
Government of the Republic of Vanuatu
The data were entered in 5 microcomputers using the specially prepared software in CsPro. The data were entered in the regional offices of the VNSO, with 10 staff trained prior to data processing. In order to ensure quality control, the software was programmed to check the internal consistency of data entered. The SPSS-10 statistical package was used for data tabulation and analysis.
Estimates of Sampling Error
To estimate the standard errors for MICS indicators we used the estimation of variance for the proportion given in the formula:
Vp’= Def*p (1-p)/(n-1), where:
p – proportion for the variance estimate,
n – sample size, and
Def – effect of sample planning for the observed group of indicators.
The standard error is the square root of Var xd'.
To calculate the variance for the whole population, the estimations of variance for the separate domains were summed.
The approximate design effect was derived from the estimation of the variance of the simple random sample, and from the estimation of the variance proposed in the ultimate cluster method. The design effect was calculated for all groups of variance and separately for all observed domains. All differences denoted as significant in the text are significant at the 95 percent confidence level, unless otherwise indicated.
As a basic check on the quality of the survey data, the percentage of cases missing information on selected questions is shown in Table 3 in the Survey Report. Fewer than one percent of household members have missing information on their level of education but three percent are missing data on the year of education. Among female respondents, 0.2 percent did not report a complete birth date (i.e., month and year). Three percent of women who had a birth in the 12 months prior to the survey did not report the date of their last tetanus toxoid injection. These low levels of missing data suggest that there were not significant problems with the questions or the fieldwork.
The data on weight and height are the most likely among the selected information to be missing. Approximately five percent of children are missing this information, which may be the result of the child not being present, refusal, or some other reason. By international standards, this percentage is relatively low in comparison to other surveys in which anthropometric measurements are taken.
The single year age distribution of household members by sex exhibits some distortions centered around age 15 for females and on age two for males. There appears to be significant heaping of women on ages 14-17 and perhaps a slight dearth of women ages 18-19. For both sexes, some digit preference is evident for ages ending in 0 and 5, a pattern typical of populations in which ages are not always known.
Data has been anonymized and is available for download at http://www.childinfo.org/MICS2/MICSDataSet.htm
MICS2 has put greater efforts in not only properly documenting the results published in the MICS2 country reports, but also to maximize the use of micro data sets via documentation and dissemination. For those MICS2 countries that granted UNICEF direct access to the micro data sets and documentation, a rigorous process was completed to ensure internal and external consistency, basic standards of data quality, corresponding documentation and, standardization of variable and value labels across countries.
For each country four SPSS data files were produced, corresponding to the four main units of analysis: households, household members, women in reproductive age (15-49 years of age) and children under the age of five. An additional Word file contains basic characteristics of the data such as year of the survey, sample sizes, weights, dictionary of variables and labels, and any existing limitations of the data files.
Data is available for download at http://www.childinfo.org/MICS2/MICSDataSet.htm
Data and metadata is provided to you for exclusive use. The data and/or metadata may not be transferred to any other user without prior authorization from UNICEF.
The dataset provided here is licensed.
"National Statistics Office of Vanuatu, Multiple Indicators Cluster Survey 2007 (MICS 2007), Version 01 of the licensed dataset (December 2008), provided by the Pacific Microdata Library. http://pdl.spc.int/index.php/home"
Disclaimer and copyrights
The Vanuatu National Statistics Office and UNICEF provide you with the data as is, without any warranty or responsibility implied. VNSO and UNICEF accepts no responsibility for the results and/or implications of any analysis and/or other actions conducted with this data.