Survey ID Number
GEO_1999_RHS_v01_M
Title
Reproductive Health Survey 1999-2000
Sampling Procedure
Results of the 99GERHS are based on in-person, face-to-face interviews with 7,798 women at their homes. The survey was designed to collect information from a representative sample of women of reproductive age throughout Georgia. Of the total, 6,143 respondents were selected from the universe of all females between the ages of 15 and 44, regardless of marital status, who were living in households in Georgia (excluding South Ossetia and Abkhazia) when the survey was carried out. In addition to the household sample, a separate sample of 1,655 internally displaced (IDP) women, who formerly resided in Abkhazia and South Ossetia and currently are living in state facilities, was performed in parallel with the household survey. This strata was added to provide a complete picture of reproductive health and women's needs in Georgia. Although about half of the IDPs in Georgia live in private dwellings (either alone or with relatives), an important segment continues to live in improvised households in communal centers (located in hotels, schools, kindergartens, farms, factories and other official buildings). Currently, it is estimated that over 100,000 IDPs are living in collective centers (UNHCR, 1999). The IDP sample of the 99GERHS was selected from the universe of IDP families living in government facilities (collective centers); these women would have otherwise been omitted from the survey, which used households in residential dwellings as the sample frame. The 1,655 women selected in the IDP sample were representative of all IDP women living in state facilities in Georgia and detailed information about their reproductive health status was published separately in the 99GERHS preliminary report (Serbanescu et al., 2000). In this final report, the IDP sample, with proper statistical weighting due to the fact that they were over-sampled, was combined with the household sample to allow the survey results to represent all women of reproductive age residing in Georgia, regardless of their housing arrangements.
Field work was conducted between November 7, 1999 and March 31, 2000. The desired sample was about 6,000 respondents for the household sample, including an oversample of women in the Imereti region, and 1,500 respondents for the IDP sample. Because the response rates were higher than expected, the actual sample size exceeds the projected sample size.
The household survey utilized a multistage sampling design using an updated sampling frame prepared by the State Department of Statistics (SDS) for the Multiple Indicator Cluster Survey conducted by UNICEF in collaboration with NCDC in July 1999. The MICS survey was designed to collect nationwide data (excluding Abkhazia and South Ossetia due to political instability) with subnational estimates. Twelve regions of the country were combined into seven survey regions and separate sampling was performed in each survey region. Grouping of regions was done taking into account the geographic location and similarity of socio-economic characteristics of the population. The SDS sampling frame contains all Georgian regions, districts, sectors, census enumeration units, census areas, and household addresses. The size of the smallest unit, the census area, contains 20-60 households; the following unit by size is the census enumeration unit incorporating 4-5 census areas with a size from 67 to 900 households; the sector is the combination of 3-5 census enumeration areas. All sectors are grouped in 53 raions (districts) that make up 12 regions (regrouped in seven regions for the MICS sampling frame). Some of the seven regions grouped for the UNICEF survey are small in size and do not always allow for independent estimates (e.g., Kakheti, Adjara). Thus, in this report the Kakheti region is part of the North-East region and Adjara is part of the West region.
The first stage of the three-stage sample design was a selection of census sectors with probability proportional to the number of households. This was accomplished by using a systematic sample with a random start in each strata; this first stage selection included 300 sectors as follows: Tbilisi (73), Imereti-Urban (28), Other-Urban (59), Imereti-Rural (27) and Other-Rural (113). In the second stage of sampling, clusters of households were randomly selected in each census sector chosen in the first stage. Cluster size determination was based on the number of households required to obtain an average of 20 completed interviews per cluster (38 households, on average). The total number of households in each cluster took into account estimates of unoccupied households, average number of women aged 15-44 per household, the interview of only one respondent per household, and an estimated response rate of 90% in urban areas and 92% in rural areas. Finally, in each of the households selected, one woman between the ages of 15 and 44 was selected at random for interview (if there was more than one woman in the household).
The 99GERHS sample includes two oversamples: a) a regional oversampling and b) an oversampling among the internally displaced population living in government facilities. Imereti region was oversampled for programmatic reasons. As in several other recent reproductive health surveys in eastern Europe conducted with CDC technical assistance (the 1996 and 1999 three-oblast surveys in Russia, the 1999 national survey in Romania and the 1999 national survey in Ukraine), the oversampling in Imereti region illustrates how surveys may be designed and integrated in the development, monitoring, and evaluation of new reproductive health programs. The oversampling of Imereti region was specifically designed to measure the impact of a region wide Women's Reproductive Health Project, a multi-faceted effort involving national and international cooperating agencies (USAID and American International Health Alliance). The project aims at reducing the reliance on induced abortion by increasing access and availability to effective contraceptive methods and promoting healthy behaviors among women, such as routine gynecologic exams, cervical and breast cancer screening. The project encompasses various interventions, such as the establishment of modern women's health clinics, training of health professionals, development of EEC messages, social marketing, and provision of high-quality contraceptive supplies and services.
The IDP oversample was also applied for programmatic reasons—to evaluate the reproductive health status of the IDP women at the end of the reproductive health program implemented by UNHCR since 1994. In addition, it was dictated by the fact that the SDS household sample did not include internally displaced households living in non-residential government facilities (collective centers). This sample also used a three-stage design. The first stage constituted of a selection of 74 collective centers throughout Georgia, proportional to the number of IDPs living in all collective centers. Most centers were located in Samegrelo, Zemo Svaneti, and Guria regions (23), Tbilisi (20), and Imereti region (19). Because information on unoccupied IDP households and the average number of women aged 15-44 per IDP household were not available, cluster size was inferred from the household sample in urban areas, since most IDP collective centers were in urban areas. Similar to the household sample, in each of the DP households selected, only one woman between the ages of 15 and 44 was selected at random for interview. The IDP sample was labeled as the sixth strata of the 99GERHS. In the preliminary report, data for the IDP women were presented separately for programmatic reasons. In the final report, however, data for all reproductive-age women in Georgia are presented as a whole and the IDP status is identified in most stratified analyses.