Survey ID Number
KSV_2013_MICS-RAEC_v01_M
Title
Multiple Indicator Cluster Survey 2013-2014
Sampling Procedure
The primary objective of the sample design for the Roma, Ashkali and Egyptian communities in Kosovo MICS was to produce statistically reliable estimates of most indicators, at the Kosovo level. The sample was stratified by Enumeration Areas with more than 50 Roma, Ashkali and Egyptian households and less than 50 Roma, Ashkali and Egyptian households.
A two-stage, stratified cluster sampling approach was used for the selection of the survey sample. The sample size for the Roma, Ashkali and Egyptian communities in Kosovo MICS was calculated as about 1,200 households.
According to the 2011 Kosovo Census, there are only 6,308 Roma, Ashkali and Egyptian households in Kosovo, or about two percent of all households. A Roma, Ashkali and Egyptian household is defined as a household with at least one person in the Roma, Ashkali or Egyptian ethnic groups. In order to examine the geographic distribution of the Roma, Ashkali and Egyptian households, KAS staff tabulated the total number of households with at least one person from these ethnic groups by EA. A total of 667 EAs were identified with at least one Roma/Ashkali/Egyptian household, but about half of these (338 EAs) have only one to three Roma, Ashkali and Egyptian households. The EAs were sorted in reverse order of the number of Roma, Ashkali and Egyptian households. It was found that 169 EAs have 10 or more Roma, Ashkali and Egyptian households, and these EAs account for 75.7% of all the Roma, Ashkali and Egyptian households in Kosovo.
It was decided that it would be both cost-effective and analytically appropriate to limit the MICS for the Roma, Ashkali and Egyptian populations to the EAs with 10 or more Roma, Ashkali and Egyptian households. Four EAs have more than 100 Roma, Ashkali and Egyptian households, and another 18 EAs have between 50 and 99 households with persons of these ethnic groups. There are 147 EAs with between 10 and 49 Roma, Ashkali and Egyptian households.
Some of the Roma, Ashkali and Egyptian population identified in the 2011 Kosovo Census may have moved since that time, given a potentially higher migration rate for this population group as they find opportunities in other areas in Kosovo or countries in the Region. Therefore the current number of Roma, Ashkali and Egyptian households in the sample EAs will only be known following a new listing of households to identify those with at least one Roma, Ashkali or Egyptian person.
In order to provide a good level of precision for the key maternal and child health indicators for the Roma, Ashkali and Egyptian population, it is recommended to have a sample size of about 1,200 households for these ethnic groups. The sampling strategy is similar to that used for the Kosovo MICS. At the first stage a sample of 80 EAs was selected with probability proportional to size (PPS) from the frame of EAs with 10 or more Roma, Ashkali and Egyptian households, where the measure of size is based on the number of households with persons of these ethnic groups in the frame. Following a new listing to identify the Roma, Ashkali and Egyptian households in the sample EAs, 16 of these households were selected in each EA at the second sampling stage. With a sample of 80 EAs selected at the first stage, the final sample size would be about 1,200 households.
Based on the selection of 80 sample EAs with PPS, the EAs with 50 or more Roma, Ashkali and Egyptian households were selected in the sample with a probability of 1. Since there are 22 such certainty EAs in the frame, a sample of 58 additional sample EAs were selected from the remainder of the frame with PPS.
Since the sampling frame (the 2011 census) was not up-to-date, a new listing of households was conducted in all the sample enumeration areas prior to the selection of households. For this purpose, listing teams were formed who visited all of the selected enumeration areas and listed all households in the enumeration areas. They also asked if there was anyone from the Roma, Ashkali and Egyptian communities living in the household to ascertain the total number from which the 16 households should be randomly selected. A separate three day listing training including a pilot in both urban and rural areas was conducted in August 2013 according to the recommended MICS procedures. A total of 26 enumerators were utilised for the listing exercise to cover the 80 EAs over August and September 2013.
Lists of households were prepared by the listing teams in the field for each enumeration area. The Roma, Ashkali and Egyptian households were then sequentially numbered from 1 to n (the total number of households in each enumeration area) at the Kosovo Agency for Statistics, where the selection of 16 households in each enumeration area was carried out using random systematic selection procedures. During the selection of EAs for the Kosovo MICS and the Roma, Ashkali and Egyptian communities in Kosovo MICS a total of eight EAs were selected for both surveys, hence a separate a systematic sample of 16 households was drawn for each survey from those EAs. The survey also included a questionnaire for individual men that was to be administered in one-half of the sample of households. To ensure systematic random selection the even/odd nature of the last digit of the cluster number was then used in conjunction with the even/odd nature of the last digit of the household number to select the specific households for interviews with all eligible men. That is If the last digit of the cluster number was odd then all households with the last digit as odd were selected to administer the male questionnaire, etc.
The sampling procedures are more fully described in "Kosovo Roma, Ashkali and Egyptian Communities Multiple Indicator Cluster Survey 2013-14 - Final Report" pp.194-195.
Questionnaires
The questionnaires for the Generic MICS were structured questionnaires based on the MICS5 model questionnaire with some modifications and additions. Household questionnaires were administered in each household, which collected various information on household members including sex, age and relationship. The household questionnaire includes List of Household Members, Education, Child Labour, Child Discipline, Household Characteristics, Water and Sanitation, and Handwashing.
In addition to a household questionnaire, questionnaires were administered in each household for women age 15-49, men age 15-49 and children under age five. The questionnaire was administered to the mother or primary caretaker of the child.
The women's questionnaire includes Woman's Background, Access to Mass Media and Use of Information/Communication Technology, Fertility/Birth History, Desire for Last Birth, Maternal and Newborn Health, Post-natal Health Checks, Illness Symptoms, Contraception, Unmet Need, Attitudes Toward Domestic Violence, Marriage/Union, Sexual Behaviour, HIV/AIDS, Tobacco and Alcohol Use, and Life Satisfaction.
The men's questionnarie includes Man's Background, Access to Mass Media and Use of Information/Communication Technology, Fertility, Attitudes Toward Domestic Violence, Marriage/Union, Sexual Behaviour, HIV/AIDS, Circumcision, Tobacco and Alcohol Use, and Life Satisfaction.
The children's questionnaire includes Child's Age, Birth Registration, Early Childhood Development, Breastfeeding and Dietary Intake, Immunization, Care of Illness, and Anthropometry.
For all children age 0-2 years with a completed Questionnaire for Children Under Five an additional form, the Questionnaire Form For Vaccination Records At Health Facility, was used to record vaccinations from the registers at health facilities. Although all vaccination records for children under 3 years of age were expected to be available with each parent, given the change in the immunization schedule in June 2010 it necessitated visits to health facilities to ensure accuracy in terms of data collection of immunization records given the possible complications. The MICS field staff copied the vaccination information from the immunization card of the child in the Health Facility.
The questionnaires are based on the MICS5 model questionnaire11. From the MICS5 model English version, the questionnaires were customised and translated into Albanian and Serbian languages and were pre-tested in Fushe Kosove/Kosovo Polje, Gracanice/Gracanica, and Zvecan/Zvecan municipalities during August 2013. Based on the results of the pre-test, modifications were made to the wording and translation of the questionnaires.