Summary statistics of population and samples taken at different sampling schemes for n = 4, r = 1.

Abstract

The main objective of a demographic household survey (DHS) is to provide estimates of a number of basic demographic and health variables. This is done through interviews with a scientifically selected probability sample that is chosen from a well-defined population.

The 2007 Nauru Demographic and Health Survey (2007 NDHS) was one of four pilot demographic and health surveys conducted in the Pacific under an Asian Development Bank ADB/ Secretariat of the Pacific Community (SPC) Regional DHS Pilot Project. The primary objective of this survey was to provide up-to-date information for policy-makers, planners, researchers and programme managers, for use in planning, implementing, monitoring and evaluating population and health programmes within the country. The survey was intended to provide key estimates of Nauru's demographics and health situation. The findings of the 2007 NDHS are very important in measuring the achievements of family planning and other health programmes. To ensure better understanding and use of these data, the results of this survey should be widely disseminated at different planning levels. Different dissemination techniques will be used to reach different segments of society.

The primary purpose of the 2007 NDHS was to furnish policy-makers and planners with detailed information on fertility, family planning, infant and child mortality, maternal and child health, nutrition, and knowledge of HIV and AIDS and other sexually transmitted infections.

NOTE: The only dissemination used was wide distribution of the report. A planned data use workshop was not undertaken. Hence there is some misconceptions and lack of awareness on the results obtained from the survey. The report is provided on the NBOS website free for download.

Geographic coverage

National Coverage - Districts

Analysis unit

• Households
• Children (0-14yrs)
• Individual women of reproductive age (15-49 yrs)
• Individual men of reproductive age (15yrs+)
• Facilities providing reproductive and child health services

Universe

The survey covered all household members (usual residents), - All children (aged 0-14 years) resident in the household - All women of reproductive age (15-49 years) resident in all household - All males (15yrs and above) in every second household (approx. 50%) resident in selected household

Results: The 2007 Nauru Demographic Health Survey (2007 NDHS) is a nationally representative survey of 655 eligible women (aged 15-49) and 392 eligible men (aged 15 and above).

Kind of data

Sample survey data [ssd]

Sampling procedure

IDG NOTES: Locate sampling documentation with SPC (Graeme Brown) and internal files. Add in this sections. Or second option dilute appendix A Sampling and extract key issues.

ESTIMATES OF SAMPLING ERRORS - Refer to Appendix A of final NDHS2007 report or; - External Resources - 2007 DHS- Appendix A and B Sampling (to be created separatedly by IDG progress ongoing)

Sampling deviation

IDG NOTES: Locate sampling documentation with Macro and internal files. Add in this section. Or second option dilute appendix B Sampling and extract key issues.

ESTIMATES OF SAMPLING ERRORS - Refer to Appendix B of final NDHS2007 report or;

• External Resources
  • 2007 DHS- Appendix A and B Sampling (to be created separatedly by IDG progress ongoing)

Extract:

In the 2007 NDHS Report of the survey results, sampling errors for selected variables have been presented in a tabular format. The sampling error tables should include:

.. Variable name

R: Value of the estimate; SE: Sampling error of the estimate; N: Unweighted number of cases on which the estimate is based; WN: Weighted number of cases; DEFT: Design effect value that compensates for the loss of precision that results from using cluster rather than simple random sampling; SE/R: Relative standard error (i.e. ratio of the sampling error to the value estimate); R-2SE: Lower limit of the 95% confidence interval; R+2SE: Upper limit of the 95% confidence interval (never >1.000 for a proportion).

Mode of data collection

Face-to-face [f2f]

Research instrument

DHS questionnaire for women cover the following sections:

• Background characteristics (age, education, religion, etc)
• Reproductive history
• Knowledge and use of contraception methods
• Antenatal care, delivery care and postnatal care
• Breastfeeding and infant feeding
• Immunization, child health and nutrition
• Marriage and recent sexual activity
• Fertility preferences
• Knowledge about HIV/AIDS and other sexually transmitted infections
• Husbands background and women's work

The men's questionnaire covers the same except for sections 4, 5, 6 which are not applicable to men.

It was also recognized that some countries have a need for special information that is not contained in the core questionnaire. Separate questionnaire modules were developed on a series of topics. These topics are optional and include:

• maternal mortality
• pill-taking behaviour
• sterilization experience
• children's education
• women's status
• domestic violence
• health expenditures
• consanguinity

The Papua New Guinea (PNG) questionnaire was proposed for Nauru to adapt as in comparison to the existing DHS model, this is not as lengthy and time-consuming. The PNG questionnaire also dealt with high incidence of alcohol and tobacco in Nauru. Questions on HIV/AIDS and STI knowledge were included in the men's questionnaire where it was not included in the PNG questionnaire.

Response rate

IDG NOTES: Locate response rate documentation with SPC (Graeme Brown) and internal files. Add in this sections.

1. Meta-analyses conventionally weight study estimates on the inverse of their error variance, in order to maximize precision. Unbiased variability in the estimates of these study-level error variances increases with the inverse of study-level replication. Here we demonstrate how this variability accumulates asymmetrically across studies in precision-weighted meta-analysis, to cause undervaluation of the meta-level effect size or its error variance (the meta-effect and meta-variance).
2. Small samples, typical of the ecological literature, induce big sampling errors in variance estimation, which substantially bias precision-weighted meta-analysis. Simulations revealed that biases differed little between random- and fixed-effects tests. Meta-estimation of a one-sample mean from 20 studies, with sample sizes of 3 to 20 observations, undervalued the meta-variance by ~20%. Meta-analysis of two-sample designs from 20 studies, with sample sizes of 3 to 10 observations, undervalued the meta-variance by 15-20% for the log response ratio (lnR); it undervalued the meta-effect by ~10% for the standardised mean difference (SMD).
3. For all estimators, biases were eliminated or reduced by a simple adjustment to the weighting on study precision. The study-specific component of error variance prone to sampling error and not parametrically attributable to study-specific replication was replaced by its cross-study mean, on the assumption of random sampling from the same population variance for all studies, and sufficient studies for averaging. Weighting each study by the inverse of this mean-adjusted error variance universally improved accuracy in estimation of both the meta-effect and its significance, regardless of number of studies. For comparison, weighting only on sample size gave the same improvement in accuracy, but could not sensibly estimate significance.
4. For the one-sample mean and two-sample lnR, adjusted weighting also improved estimation of between-study variance by DerSimonian-Laird and REML methods. For random-effects meta-analysis of SMD from little-replicated studies, the most accurate meta-estimates obtained from adjusted weights following conventionally-weighted estimation of between-study variance.
5. We recommend adoption of weighting by inverse adjusted-variance for meta-analyses of well- and little-replicated studies, because it improves accuracy and significance of meta-estimates, and it can extend the scope of the meta-analysis to include some studies without variance estimates.

Abstract

The Serbia Multiple Indicator Cluster Survey (MICS) is a household survey programme conducted in 2010 by UNICEF and the Statistical Office of the Republic of Serbia (SORS). The survey provides valuable information on the situation of children, women and men in Serbia, and was based, in large part, on the needs to monitor progress towards goals and targets emanating from recent international agreements: the Millennium Declaration, and the Plan of Action of A World Fit For Children. Both of these commitments build upon promises made by the international community at the 1990 World Summit for Children.

The fourth round of the Multiple Indicator Cluster Survey represents a large source of data for reporting on progress towards the aforementioned goals. The survey provides a rich foundation of comparative data for comprehensive progress reporting, especially regarding the situation of the most vulnerable children (children in the poorest households, Roma children or those living in rural areas). It also provides important information for the new UNICEF Country Programme 2011-2015 as well as the UNDAF 2011-2015. This final report presents the results of the indicators and topics covered in the survey.

Datasets documented here cover Roma Settlements sample representative of the population living in Roma settlements in Serbia. A total of 1,815 Roma households were selected: 1,311 households with children and 504 households without children. A stratified, two-stage random sampling approach was used for the selection of the survey sample.

Geographic coverage

National

Analysis unit

• individuals
• households

Universe

The survey covered household members in Roma settlements, all women aged between 15-49 years, all children under 5 living in the household, and all men aged 15-29 years.

Kind of data

Sample survey data [ssd]

Sampling procedure

The primary objective of the sample design for the Roma settlements Multiple Indicator Cluster Survey was to produce statistically reliable estimates of most indicators, at the level of Serbia, and for urban and rural areas.

A stratified, two-stage random sampling approach was used for the selection of the survey sample.

The target sample size for the Roma settlements was calculated as 1,800 households and 100 enumeration areas, considering the proposed formula and budget available. For the calculation of the sample size, the key indicator used was the percentage of children aged 0-4 years who had had Acute Respiratory infections.

The resulting number of households from this exercise was about 2,700 households, which is the sample size needed to provide a large number of children under 5 (about 1,300) for drawing reliable conclusions. Therefore, in order to reduce the number of households in the sample, but not to lose estimation reliability, the stratification of the sample into categories with and without children aged 0-4 years was needed. The required number of households in each category was obtained supposing an overall sample of 1800 households, 100 clusters and same number of households with children under 5 per cluster. Assuming one child under 5 per household and considering the required number of sample children, the total sample size was calculated as 1,300 (13 per cluster) households with children under 5 and 500(5 per cluster) of households without children under 5.Thus, the overall number of households to be selected per cluster was determined as 18 households.

Stratification of enumeration areas for Roma settlements was done according to type of settlement (urban and rural), and territory, to the three strata: Vojvodina, Belgrade and Central Serbia without Belgrade.

Sample allocation of enumeration areas according to territory and type of settlement was not proportional to the number of Roma households. In order to produce estimates with better precision for territories and urban/rural domains, the number of enumeration areas for Vojvodina and rural domains was increased.

The sampling procedures are more fully described in "Multiple Indicator Cluster Survey 2010 - Final Report" pp.261-263.

Mode of data collection

Face-to-face [f2f]

Research instrument

The questionnaires for Roma settlements are the Generic MICS questionnaires based on the MICS4 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 household listing form, education, water and sanitation, household characteristics, child discipline and hand washing.

In addition to a household questionnaire, questionnaires were administered in each household for women age 15-49, children under age five and men age 15-29. For children, 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 ICT, child mortality, desire for last birth, maternal and newborn health, illness symptoms, contraception, unmet need, attitudes toward domestic violence, marriage/union, sexual behavior, HIV/AIDS, and life satisfaction.

The children's questionnaire includes child's age, birth registration, early childhood development, breastfeeding, care of illness, and anthropometry.

The men's questionnaire includes man's background, access to mass media and ICT, marriage/union, contraception, attitudes toward domestic violence, sexual behavior, HIV/AIDS, and life satisfaction.

The questionnaires were developed in English from the MICS4 Model Questionnaires, and were translated into Serbian. The Serbian versions were pre-tested in Belgrade during September 2010 and modifications were made to the wording and translation of the questionnaires based on the results of the pre-test.

Cleaning operations

Data was entered using the CSPro software. The data entry was carried out on 10 microcomputers by 20 data entry operators and 4 data entry supervisors. In order to ensure quality control, all questionnaires were double entered and internal consistency checks were performed. Procedures and standard programmes developed under the global MICS4 programme and adapted to Serbia’s questionnaire were used throughout.

Data processing began simultaneously with data collection and was completed in March 2011. Data was analysed using the Statistical Package for Social Sciences (SPSS) software programme, Version 18, and the model syntax and tabulation plans developed by UNICEF were used for this purpose.

Response rate

The response rate of households is 96 percent. (Of the 1815 households selected for the sample, 1782 were found to be occupied. Of these, 1711 were successfully interviewed.)

The response rate of women is 95 percent within interviewed households. (In the interviewed households, 2234 women aged between 15-49 years were identified. Of these, 2118 were successfully interviewed.)

The response rate of children is 99 percent within interviewed households. (1618 children under the age of five were listed in the household questionnaire. Questionnaires were completed for 1604 of these children.)

The response rate of men is 78 percent within interviewed households.(1121 men aged between 15-29 years were identified. Of these, 877 were successfully interviewed.)

Overall response rates of 91, 95 and 75 percent respectively are calculated for the women’s, under-5’s and men’s interviews.

Sampling error estimates

Sampling errors are a measure of the variability between the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.

The following sampling error measures are presented for each of the selected indicators: - Standard error (se): Sampling errors are usually measured in terms of standard errors for particular indicators (means, proportions etc). Standard error is the square root of the variance of the estimate. The Taylor linearization method is used for the estimation of standard errors. - Coefficient of variation (se/r) is the ratio of the standard error to the value of the indicator, and is a measure of the relative sampling error. - Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a deft value above 1.0 indicates the increase in the standard error due to the use of a more complex sample design. - Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2se or r – 2se) of the statistic in 95 percent of all possible samples of identical size and design.

For the calculation of sampling errors from MICS data, SPSS Version 18 Complex Samples module has been used.Sampling errors are calculated for indicators of primary interest, for the national level and for urban and rural areas. Five of the selected indicators are based on household members, 18 are based on women, 8 are based on men and 12 are based on children under 5. All

The case of size-biased sampling of known order from a finite population without replacement is considered. The behavior of such a sampling scheme is studied with respect to the sampling fraction. Based on a simulation study, it is concluded that such a sample cannot be treated either as a random sample from the parent distribution or as a random sample from the corresponding r-size weighted distribution and as the sampling fraction increases, the biasness in the sample decreases resulting in a transition from an r-size-biased sample to a random sample. A modified version of a likelihood-free method is adopted for making statistical inference for the unknown population parameters, as well as for the size of the population when it is unknown. A simulation study, which takes under consideration the sampling fraction, demonstrates that the proposed method presents better and more robust behavior compared to the approaches, which treat the r-size-biased sample either as a random sample from the parent distribution or as a random sample from the corresponding r-size weighted distribution. Finally, a numerical example which motivates this study illustrates our results.

Abstract

The Child development survey (or MICS) 2012 provides valuable information on the situation of children, women and men in Nalaikh district, for measuring fulfilment of their rights. It was based largely on the needs to monitor progress towards goals and targets, pertinent to recent international agreements: The Millennium Declaration, adopted by all 191 United Nations Member States in September 2000, and the Plan of Action of A World Fit For Children, adopted by 189 Member States at the United Nations Special Session on Children in May 2002. Both of these commitments build upon promises made by the international community at the 1990 World Summit for Children.

OBJECTIVES

The Nalaikh district’s “Child Development Survey-2012” (Multiple Indicator Cluster Survey) has the following primary objectives:

• To provide up-to-date information for assessing at the district level the following national and international level policies and programmes

A) the World Fit for Children Declaration

B) Millennium Development Goals

C) National Reproductive Health Programme

• To serve the baseline for UNICEF’s Country Programme 2012-2016

• To build the capacity of the Statistics Department of the District.

Geographic coverage

District level.

Analysis unit

• Individuals

• Households

Universe

The survey covered all households, women and men age 15-49 years, and children under age of 5 and age 2-14 years.

Kind of data

Sample survey data [ssd]

Sampling procedure

The Child Development Survey is a household-based survey. Therefore, households are defined as the final sampling units. The sample for the survey was designed to provide estimates for a number of indicators on the situation of children, women and men at the district level. The total sample size was determined as 1,000 households for the district.

In total for the Nalaikh district, 40 khesegs were selected systematically with probability proportional to size. After a household listing of the selected PSUs or the selected khesegs was carried out by the khoroo’s governor, 25 households were selected using systematic random sampling in each PSU.

Data were collected from the households in the sample, and for reporting the district level results, sample weights are used. A more detailed description of the sample design can be found in the Final Report (Appendix A) attached as a Related Material.

Sampling deviation

During the data collection fieldwork in July-August 2012, we had encountered a problem due to nonappearance of families at the registered addresses, and absence of family members, because of seasonal resort and vacation period. In spite of this, we managed to collect survey data from all selected PSUs.

Mode of data collection

Face-to-face [f2f]

Research instrument

Based on the five core questionnaires contents of the Mongolia Child Development Survey, conducted nationwide in 2010, specific supplementary module and questions were added for the Nalaikh “Child Development Survey 2012”. Based on the current priorities and needs, the questionnaire for men age 15-49 years was taken in its entirety for this round of CDS.

Altogether, five types of questionnaires were used:

1. A Household Questionnaire

2. A Questionnaire for Woman, age 15-49

3. A Questionnaire for Child under 5

4. A Questionnaire for Child, age 2-14

5. A Questionnaire for Man, age 15-49

In addition to the administration of the questionnaires, fieldwork teams tested the salt used for cooking in the households for iodine content, observed the place for hand washing and measured the weights and heights of children age under 5 years.

In this round CDS 2012, internal migration questions (country specific module in household questionnaire) were asked for all household member listed in household listing module (HL).

The Questionnaire for Child under 5 was administered to mothers or caretakers of all children under 5 years of age living in the households. Normally, the questionnaire was administered to mothers of under-5 children; in cases when the mother was not listed in the household roster, a primary caretaker for the child was identified and interviewed.

The Questionnaire for Child age 2-14 was administered to mothers or caretakers of children age 2-14 years living in the households. Normally, the questionnaire was administered to mothers of children age 2-14; in cases when the mother was not listed in the household roster, a primary caretaker for the child was identified and interviewed.

All questionnaires modules are provided as Related Materials.

Cleaning operations

The data collected from the selected households were entered on computers using the CSPro 4.0 software program by one data entry supervisor and two data entry operators from 20 August to 10 September 2012. In order to ensure quality control, all data were double entered and internal consistency checks were performed before finalization of the database. The procedures and standard programs developed under the global MICS4 programme and adapted to the Nalaikh CDS's customized questionnaires with additional module and questions were used throughout.

The data were analyzed using the standard SPSS 18.0 (Statistical Package for Social Sciences) software program and the model syntax and tabulation plans developed by UNICEF were customized for Nalaikh CDS 2012 questionnaires.

Response rate

In total, 1,000 households selected for the sample, and of these 956 were found to be available for the survey. Of these, 949 households were successfully interviewed for a household response rate of 99 percent. In the interviewed households, out of the total 799 men and 929 women, age 15-49 years, enlisted for the survey, 705 men and 889 women were successfully interviewed, yielding a response rate of 88 and 96 percent respectively. In addition, 433 children under age of 5 and 896 children age 2-14 years were listed in the household questionnaire. Questionnaires were completed with mothers/ caretakers for 429 of these under-5 children and for 894 of children age 2-14, which corresponds to response rates of 99 and 100 percent respectively, within interviewed households.

Nalaikh district’s overall response rates stand at 88 percent for men, 95 percent for women age 15-49 years, 98 percent and 99 percent are calculated for mothers/ caretakers of children under 5's and children age 2-14's respectively.

However, the response rate for men age 15-49 years’ interviews is relatively lower than the response rates for other interviews, because of the dynamic mobility nature of men, particularly of young men.

Sampling error estimates

The sample of respondents selected in the Nalaikh District Multiple Indicator Cluster Survey 2012 is only one of the samples that could have been selected from the same population, using the same design and size. Each of these samples would yield results that slightly differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.

The sampling error measures for each of the selected indicators are:

• Standard error (se): Sampling errors are usually measured in terms of standard errors for particular indicators (means, proportions etc). Standard error is the square root of the variance of the estimate. The Taylor linearization method is used for the estimation of standard errors.

• Coefficient of variation (se/r) is the ratio of the standard error to the value of the indicator, and is a measure of the relative sampling error.

• Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a deft value above 1.0 indicates the increase in the standard error due to the use of a more complex sample design.

• Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2.se or r - 2.se) of the statistic in 95 percent of all possible samples of identical size and design.

For the calculation of sampling errors from MICS data, SPSS Version 18 Complex Samples module has been used.

Abstract

In 2015, Innovation Edge commissioned the development of South Africa's first national-level preschool child assessment tool. Finalized in 2016, the primary purpose of the Early Learning Outcomes Measure (ELOM) (www.elom.org.za) is to provide the country with a national instrument to fairly assess children age 50-69 months from all socio-economic and cultural backgrounds. With this tool, we will be able to monitor early learning program outcomes, guide program improvement, and test program effectiveness. This study is conducted to assist Innovation Edge with the 2nd phase in the process for the development of ELOM--Age validation. In this study, ELOM tools was administered to children enrolled in public schools at the commencement of their Grade R year in 2016, in schoolf of all five quintiles in three provinces. The goal of the ELOM age validation process was to construct a sample that was likely to be as representative as possible of children eligible to enter Grade R in January 2016, drawn from across South Africa's socio-economic distribution, and including five major language groups.

The ELOM includes both direct assessment of children's performance as well as an assessment of the child's social and emotional functioning and orientation to tasks. The ELOM Direct Assessment consists of 23 items measuring indicators of the child's early development in five domains: Gross Motor Development Fine Motor Coordination and Visual Motor Integration Emergent Numeracy and Mathematics Cognition and Executive Functioning Emergent Literacy and Language

Geographic coverage

It is envisioned that the ELOM will be applied to children from a range of cultural and socio-economic settings. However, as finances did not permit a national sample, three provinces were chosen for the study. The sample for this study then aimed to be representative of public school Grade R students in the target language groups who were between the ages of 54-66 months in selected school districts in North West (Setswana speakers only), the Western Cape (English, isiXhosa and Afrikaans speakers) and KwaZulu Natal (isiZulu speakers only).

Analysis unit

Individuals and institutions

Universe

Desired target population is Grade R children in South African public schools between the ages of 54-66 months.

Kind of data

Observation data

Sampling procedure

A two-stage clustered sample design was employed. In the first stage, and in each district, probability proportional to Grade R population size sampling was used to randomly select schools within each of the five School Quintile bands. Two schools in traditional, more rural areas in each of North West and KwaZulu-Natal were recruited independently of this exercise to explore the influence of more "traditional" approaches to child rearing. In the second stage, learners were selected within Grade R classes using simple random sampling. Minimum of nine children per school were selected per cluster.

Mode of data collection

Other

Abstract

The Malawi MDG Endline Survey (MES) was carried out in 2013-14 by National Statistical Office as part of the global MICS programme. The global MICS programme which the MES is part of was developed by UNICEF in the 1990s as an international household survey programme to collect internationally comparable data on a wide range of indicators on the situation of children and women. MICS surveys measure key indicators that allow countries to generate data for use in policies and programmes, and to monitor progress towards the Millennium Development Goals (MDGs) and other internationally agreed upon commitments. The basic objective of the MES 2014 is to provide information on indicators for monitoring progress of attainment of the Millennium Development Goals and Malawi Growth and Development Strategy and other development programmes. Through collection and calculation of status of indicators of the Millennium Development Goals and other key social statistics indicators, the MES data will also be used to update the socio-economic database for policy and research.

The 2014 MES has as its primary objectives: - To provide up-to-date information for assessing the situation of children and women in Malawi - To generate data for the critical assessment of the progress made in various areas, and to put additional efforts in those areas that require more attention; - To furnish data needed for monitoring progress toward goals established in the Millennium Declaration and other internationally and nationally agreed upon goals and to serve as a basis for future action; - To collect disaggregated data for the identification of disparities, to allow for evidence based policy-making aimed at social inclusion of the most vulnerable; - To contribute to the generation of baseline data for the post-2015 agenda; - To validate data from other sources and the results of focused interventions.

Geographic coverage

The sample for the 2014 MES was designed to provide estimates for a large number of indicators at the national level; for urban and rural areas; the three regions (Northern Region, Central Region and Southern Region); and the 27 districts of Malawi excluding Likoma.

Analysis unit

Household Women (15-49 years) Men (15-49 years) Children under 5 years

Universe

the survey covered ز - All Household -All Women (15-49 years) - All Men (15-49 years) living in the onethird of households -All Children under 5 years

Kind of data

Sample survey data [ssd]

Sampling procedure

The sample for the 2014 MES was designed to provide estimates for a large number of indicators on the situation of children and women at the national level; for urban and rural areas; the three regions (Northern Region, Central Region and Southern Region); and the 27 districts of Malawi excluding Likoma2. The urban and rural areas within each region were identified as the main sampling strata and the sample was selected in two stages. Within each stratum, a specified number of census enumeration areas were selected systematically with probability proportional to size. After a household listing was carried out within the selected enumeration areas (EAs), a systematic sample of 25 households was drawn in each sample cluster. A total of 1,140 sample EAs and 28,479 households were selected for the 2014 MES. One of the selected clusters (Cluster 0152) in Rumphi district was not visited because it was inaccessible due to heavy rains and poor road network during the fieldwork period. The sample is not self-weighting. For reporting national level results, sample weights are used. A more detailed description of the sample design can be found in Appendix B, Sample Design.

Mode of data collection

Face-to-face [f2f]

Research instrument

Four sets of questionnaires were used in the survey: 1) a household questionnaire which was used to collect basic demographic information on all de jure household members (usual residents), the household, and the dwelling; 2) a questionnaire for individual women administered in each household to all women age 15-49 years; 3) a questionnaire for individual men administered in every third household to all men age 15-49 years; 4) an under-5 questionnaire, administered to mothers (or caretakers) for all children under 5 living in the household.

The questionnaires are based on the MICS5 model questionnaire. From the MICS5 model English version, the questionnaires were customised and translated into Chichewa and Tumbuka and were pre-tested in Kasungu district during October 2013. Based on the results of the pre-test, modifications were made to the wording and translation of the questionnaires. A copy of the 2014 MES English version questionnaires is provided in Appendix H and the translated version (Chichewa or Tumbuka) can be obtained from the NSO on request.

In addition to the administration of questionnaires, fieldwork teams tested the salt used for cooking in the households for iodine content, observed the place for handwashing, and measured the weights and heights of children age under 5 years.

Cleaning operations

Data were entered using the CSPro software, Version 5.0. The data were entered on 30 desktop computers and carried out by 30 data entry operators and 4 data entry supervisors. For quality assurance purposes, all questionnaires were double-entered and internal consistency checks were performed. Procedures and standard programs developed under the global MICS programme and adapted to the MES questionnaire were used throughout. Data processing began simultaneously with data collection in December 2013 and was completed in May 2014. Data were analysed using the Statistical Package for Social Sciences (SPSS) software, Version 21. Model syntax and tabulation plans developed by UNICEF were customized and used for this purpose.

Response rate

• Of the 28,479 households selected for the sample, 27,030 were occupied. Of these, 26,713 were interviewed, giving a response rate of 98.8 percent.
• In the households interviewed, 25,430 women age 15-49 years were eligible for interviews and of these 24,230 were interviewed producing a response rate of 95.3 percent.
• For the male survey, 7,818 men age 15-49 years were identified, and 6,842 successfully interviewed, yielding a response rate of 87.5 percent.
• Concerning children under the age of 5 years, 19,285 were eligible, for whom responses were obtained from their mother or caregiver in 18,981 complete interviews, giving a response rate of 98 percent.

Sampling error estimates

The sample of respondents selected in the Malawi MDG Endline Survey (MES) is only one of the samples that could have been selected from the same population, using the same design and size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.

The following sampling error measures are presented in this appendix for each of the selected indicators: ? Standard error (se): Standard error is the square root of the variance of the estimate. For survey indicators that are means, proportions or ratios, the Taylor series linearization method is used for the estimation of standard errors. For more complex statistics, such as fertility and mortality rates, the Jackknife repeated replication method is used for standard error estimation. ? Coefficient of variation (se/r) is the ratio of the standard error to the value (r) of the indicator, and is a measure of the relative sampling error. ? Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling based on the same sample size. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design of the survey is as efficient as a simple random sample for a particular indicator, while a deft value above 1.0 indicates an increase in the standard error due to the use of a more complex sample design. ? Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2.se or r – 2.se) of the statistic in 95 percent of all possible samples of identical size and design. For the calculation of sampling errors from MICS data, programs developed in CSPro Version 5.0, SPSS Version 21 Complex Samples module and CMRJack70 have been used.

Python and R scripts to replicate the analyses in this paper.

Abstract

The Gambia Multiple Indicator Cluster Survey 2010 is a nationally representative survey of households, children and women. The main objectives of the survey was to provide up-to-date information for assessing the situation of children and women in The Gambia. Another objective was to furnish data needed for monitoring progress towards the goals established at the World Summit for Children and the Millennium Development Goals (MDGs) as a basis for future action. The findings of this survey would also be utilized by government and development partners in planning and monitoring program implementation.

The module development for the survey captured data on households characteristics, education, water and sanitation, insecticides treated nets, indoor residual spraying, salt iodization, handwashing, birth registration, early childhood development, Breastfeeding, care of illness, malaria, immunization, anthropometry, child mortality, desire for last birth, illness symptoms, maternal and newborn health, rehydration solutions, contraception, unmet need, female genital mutilation, attitudes toward domestic violence, marriage/ union, sexual behavior, and HIV/AIDS. The survey was conducted through inter-agency collaboration with The Gambia Bureau of Statistics (GBoS), acting as the lead agency.

The Gambia's Multiple Indicator Cluster Survey 2010 has the following primary objectives: 1. To provide up-to-date information for assessing the situation of children and women in The Gambia. 2. To furnish data needed for monitoring progress towards the goals established in the Millennium Declaration, the goals of A World Fit for Children (WFFC) and other internationally agreed upon goals as a basis for future action. 3. To contribute to the improvement of data and monitoring systems in The Gambia and to strengthen technical expertise in the design, implementation and analysis of such systems. 4. To generate data on the situation of children and women, including the identification of vulnerable groups and of disparities, to inform policies and interventions.

Geographic coverage

National

Analysis unit

• Households (defined as a group of persons who usually live and eat together)
• Women aged 15-49
• Children aged 0-4

Universe

The survey covered all de jure household members (usual residents), all women aged 15-49 years living in the household, and all children aged 0-4 years (under age 5) living in the household.

Kind of data

Sample survey data [ssd]

Sampling procedure

The sample for The Gambia Multiple Indicator Cluster Survey (MICS4) was designed to provide estimates on a large number of indicators on the situation of children and women at the national level, for urban and rural areas, and for the eight Local Government Areas (LGAs): Banjul, Kanifing, Brikama, Mansakonko, Kerewan, Kuntaur, Janjanbureh and Basse. Other than Banjul and Kanifing which are entirely urban settlements, urban and rural areas within each LGA were identified as the main sampling domains and the sample was selected in two stages. Within each LGA, at least 44 and at most 60 census enumeration areas, (EA's) or clusters were selected systematically with Probability Proportional to Size (PPS).

Sampling deviation

No major deviations from the original sample design were made. All sample enumeration areas were accessed and successfully interviewed with good response rates.

Mode of data collection

Face-to-face [f2f]

Research instrument

The questionnaires are based on the MICS4 Model Questionnaire III. Given that the MICS4 model questionnaires were in an English version, the questionnaires were not translated into the local languages for the training part. The training program for staff conducting or supervising the interviews included detailed discussions of the contents of the questionnaires, how to complete the questionnaires, and interviewing techniques. In addition to taking the trainees through the questionnaires in English, the questions were also verbally translated into the three main local languages of The Gambia (Wollof, Mandinka and Fula). A participatory approach was adopted during these translation sessions to ensure that all participants had common understanding of the translation of all the questions. The questionnaires were pre-tested in few selected EAs in the Greater Banjul in April, 2010. Based on the results of the pre-test, modifications were made to the wording and translation of the questionnaires.

Cleaning operations

Data were entered into 20 microcomputers using the Census and Surveys Processing System (CSPro) software package. Data entry was carried out by forty entry operators and four supervisors. For quality assurance purposes, all questionnaires were double-entered and internal consistency checks performed. Procedures and standard programs developed under the global MICS program and adapted to The Gambia questionnaire were used throughout. Data processing began simultaneously with data collection in April 2010 and was completed in August 2010. Data were analyzed using the Statistical Package for Social Sciences (SPSS) software program, Version 18. Model syntax and tabulation plans developed by UNICEF were customized and used for this purpose.

Response rate

Of the 7,800 households selected for the sample survey, 7,799 households were found to be occupied. Of these 7,791 were successfully interviewed for a household response rate of 99.9 percent. In the interviewed households, the survey identified 15,138 women (age 15-49 years). Of these 14,685 were successfully interviewed, resulting to a response rate of 97.0 percent within interviewed households. In addition 11,807 children under age five were listed. Questionnaires were completed for 11,637 of these children, which corresponds to a response rate of 98.6 percent within interviewed households.

Sampling error estimates

The sample of respondents selected in the Gambia Multiple Indicator Cluster Survey is only one of the samples that could have been selected from the same population, using the same design and size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.

The following sampling error measures are presented in this appendix for each of the selected indicators: 1. Standard error (se): Sampling errors are usually measured in terms of standard errors for particular indicators (means, proportions, etc.). Standard error is the square root of the variance of the estimate. The Taylor linearization method is used for the estimation of standard errors. 2. Coefficient of variation (se/r) is the ratio of the standard error to the value of the indicator, and is a measure of the relative sampling error. 3. Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a deft value above 1.0 indicates the increase in the standard error due to the use of a more complex sample design. 4. Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2.se or r - 2.se) of the statistic in 95 percent of all possible samples of identical size and design.

For the calculation of sampling errors from MICS data, SPSS Version 18 Complex Samples module has been used. The results are shown in the tables that follow. In addition to the sampling error measures described above, the tables also include weighted and unweighted counts of denominators for each indicator. Sampling errors are calculated for indicators of primary interest, for the national level, for the regions, and for urban and rural areas. Three of the selected indicators are based on households, 8 are based on household members, 13 are based on women, and 15 are based on children under 5. All indicators presented here are in the form of proportions. Table SE.1 shows the list of indicators for which sampling errors are calculated, including the base population (denominator) for EAC indicator. Tables SE.2 to SE.12 show the calculated sampling errors for selected domains.

Abstract

The survey is designed to provide statistically sound and internationally comparable data essential for developing evidence-based policies and programmes and for monitoring progress towards national goals and global commitments. Among these global commitments are those emanating from the World Fit for Children Declaration and Plan of Action, the goals of the United Nations General Assembly Special Session on HIV/AIDS, the Education for All Declaration and the Millennium Development Goals (MDGs). The Zimbabwe MICS 2014 results are critical for final MDG reporting in 2015 and are expected to form part of the baseline data for the post-2015 era. The survey results are expected to contribute to the evidence base of several other important initiatives, including Committing to Child Survival: A Promise Renewed 2012.

The Zimbabwe MICS 2014 primary objectives were: - To collect information critical to the monitoring and reporting on selected indicators for all the 8 MDGs, - To assist in monitoring of Government of Zimbabwe (GoZ) ZimASSET national priorities focusing on basic social services, - To assist monitoring the Zimbabwe United Nations Development Assistance Framework (ZUNDAF) 2012 to 2015 and individual GoZ/United Nations programme social outcome indicators including transition funds, namely, the Health Transition Fund (HTF), Education Transition Fund (ETF), Child Protection Fund (CPF), and Water, Sanitation and Hygiene (WASH) programme - To provide decision makers with evidence on the situation of children’s and women’s welfare and rights and other vulnerable groups in Zimbabwe.

Geographic coverage

The survey was designed to provide estimates at national, provincial and urban/rural levels.

Analysis unit

Household Women Men Children under 5

Universe

the survey cover: - All households type - women (15-49 years) - Men (15-49 years) - children under five years - Children (5-17 years)

Kind of data

Données échantillonées [ssd]

Sampling procedure

The sample for the Zimbabwe Multiple Indicator Cluster Survey was designed to provide estimates for a large number of indicators on the situation of children and women at the national, provincial and urban/rural levels. The ten provinces of the country are Manicaland, Mashonaland Central, Mashonaland East, Mashonaland West, Matabeleland North, Matabeleland South, Midlands, Masvingo, Harare and Bulawayo. With the exception of Bulawayo, the other nine provinces were stratified into urban and rural areas. The sample was selected in two stages with the selection of census enumeration areas/clusters in the first stage and selection of households in the second stage. Within each stratum, a specified number of clusters were selected systematically with probability proportional to size. At the second sampling stage, 25 households were selected from each cluster using systematic random sampling.

After a household listing was carried out within the selected enumeration areas, a representative sample of 17 075 households was drawn from 683 clusters. One cluster in Masvingo Province (Tokwe- Mukosi) was not enumerated due to flooding as affected households were re-located. The sample was stratified by province, urban and rural areas and is not self-weighting. For reporting national level results, sample weights are used. A more detailed description of the sample design can be found in Appendix C, Sample Design. in the report.

Mode of data collection

Interview face à face [f2f]

Research instrument

A set of four questionnaires was used in the survey. These questionnaires were adapted and customized from standard MICS5 questionnaires. All questionnaires were translated from English to two main vernacular languages in Zimbabwe, i.e. Shona and Ndebele.

The questionnaires are described below:

• A household questionnaire was used to collect basic demographic information on all de jure household members (usual residents), the household, and the dwelling. This questionnaire was responded to by the head of household or a chief respondent covered the household information panel, listing of household members, education, child discipline for children 1-14 years of age, household characteristics, water and sanitation, handwashing, indoor residual spraying, use of Insect Treated Nets (ITNs), and salt iodisation.

• A Woman’s questionnaire was administered to all women in the 15 to 49 year age group from each selected household, encompassed the woman’s information panel, her background characteristics, fertility, birth history, desire for last birth, maternal and newborn health, maternal mortality, postnatal care, marriage/union, illness symptoms, attitudes towards domestic violence, access to mass media and use of information communication technology, tobacco and alcohol use, contraception, unmet need, sexual behaviour, and knowledge on HIV and AIDS.

• A Man’s questionnaire for the 15 to 54 year age group was administered in every third household selected. The man’s information panel, his background characteristics, fertility, marriage/union, attitudes towards domestic violence, access to mass media and use of information communication technology, tobacco and alcohol use, sexual behaviour, circumcision and knowledge on HIV and AIDS.

• The under-five questionnaire was administered to mothers (or primary caregivers) of children under 5 years of age7 living in the households. Normally, the questionnaire was administered to mothers of under-5 children; in cases when the mother was not listed in the household listing panel, a primary caregiver for the child was identified and interviewed. The questionnaire covered children’s characteristics, birth registration, early childhood development, breastfeeding and dietary intake, care of illness, immunisation and anthropometry.

Cleaning operations

Data were entered into the computers using the Census and Surveys Processing System (CSPro) software package, Version 5.0. The data were entered on 32 desktop computers by 42 data entry operators and nine data entry supervisors. For quality assurance purposes, all questionnaires were double-entered and internal consistency checks were performed. Procedures and standard programmes developed under the global MICS programme and adapted to the Zimbabwe questionnaire were used throughout. Data entry started two weeks into data collection in March 2014 and was completed in May 2014. Data were analysed using the Statistical Package for Social Sciences (SPSS) software, Version 21. Model syntax and tabulation plans developed by the Global MICS team were customized and used for this purpose.

Response rate

Households questionnaire: The sample size was estimated at 17047 households of which 15686 households are interviewed ( 92.1% as response rate) Women (age 15-49) : out of 15376 women who are eligible 14408 are interviewed ( 93.7% as response rate) Men (age 15-49) : out of 9008 men who are eligible 7914 are interviewed ( 87.9% as response rate) Children under five: out of 10223 children who are eligible 9884 are interviewed ( 96.7% as response rate)

Sampling error estimates

for more information view :Appendix E. Estimates of Sampling Errors

The sample of respondents selected in the Zimbabwe Multiple Indicator Cluster Survey is only one of the samples that could have been selected from the same population, using the same design and size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.

The following sampling error measures are presented in this appendix for each of the selected indicators: Standard error (se): Standard error is the square root of the variance of the estimate. For survey indicators that are means, proportions or ratios, the Taylor series linearization method is used for the estimation of standard errors. For more complex statistics, such as fertility and mortality rates, the Jackknife repeated replication method is used for standard error estimation.

Coefficient of variation (se/r) is the ratio of the standard error to the value (r) of the indicator, and is a measure of the relative sampling error.

Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling based on the same sample size. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design of the survey is as efficient as a simple random sample for a particular indicator, while a deft value above 1.0 indicates an increase in the standard error due to the use of a more complex sample design.

Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2.se or r – 2.se) of the statistic in 95 percent of all possible samples of identical size and design.

The Talyor series variance estimation method was used in the calculation of sampling errors. The variance estimator takes into account the different aspects of the sample design, such as the stratification and

Abstract

the Ghana Multiple Indicator Cluster Survey, conducted in 2011 by the Ghana Statistical Service (GSS) provides valuable information on the situation of children, women and men in Ghana, and was based, in large part, on the need to monitor progress towards goals and targets emanating from recent international agreements: The Millennium Declaration, adopted by all 191 United Nations Member States in September 2000, and the Plan of Action of A World Fit For Children, adopted by 189 Member States at the United Nations Special Session on Children in May 2002. Both of these commitments build upon promises made by the international community at the 1990 World Summit for Children.

The Ghana Multiple Indicator Cluster Survey (MICS) 2011, the fourth of its kind, is a nationally representative sample survey of households, women aged 15-49 years, children aged 0-5 years and men aged 15-59 years. In addition to applying the customized version of the MICS4 Questionnaires, an enhanced Malaria Module and Biomarker (for Anaemia and parasitemia in children aged 6-59 months) was included.

The 2011 Ghana Multiple Indicator Cluster Survey has as its primary objectives the following: • To provide more current information for assessing the situation of children, women and men, and reporting on country progress in achieving the GSGDA goals/targets and the MDGs, meet the reporting requirements of other local and international development declarations and agenda, and form the basis for future action; • To provide much-needed data on practices used to treat malaria among children under-five and the use of specific anti-malarial medications, bednet coverage and use, coverage of Intermittent Preventive Treatment for pregnant women, treatment practices for childhood fever, and prevalence of malaria and anaemia among children aged 6-59 months; • To present the current level of knowledge and behavioral indicators regarding HIV and AIDS in Ghana; • To provide a mid-term snapshot on progress on key Health Sector Medium-term Development Plan (HSMTDP) 2010-2013 strategic objectives, and provide nationally and regionally representative data that can inform the development of the next Health Sector Medium-term Plan; • To contribute to the improvement of data and monitoring systems in Ghana and to strengthen technical expertise in the design, implementation, and analysis of such systems; and • To generate data on the situation of children, women and men, including the identification of vulnerable groups and of disparities, which will inform social inclusion and poverty reduction policies and interventions.

Geographic coverage

National coverage

Analysis unit

Households Women 15-49 years old Men 15-59 years old Children under five years

Universe

the survey covered: - all householdes - All women 15-49 years - All children under 5 - All men 15-59 years living in each third household

Kind of data

Sample survey data [ssd]

Sampling procedure

The sample for the Ghana Multiple Indicator Cluster Survey (MICS) was designed to provide estimates for a large number of indicators on the situation of children and women at the national level, for urban and rural areas, and for 10 regions: Western, Central, Greater Accra, Volta, Ashanti, Brong Ahafo, Northern, Eastern, Upper East and Upper West regions. The urban and rural areas within each region were identified as the main sampling strata and the sample was selected in two stages. Within each stratum, a specified number of census enumeration areas were selected systematically with probability proportional to size. Since the sampling frame (the 2000 Ghana Population Census) was up-to-date, a new listing of households was not conducted in all the sample enumeration areas prior to a systematic sample selection of 15 households in each selected cluster. The sample was stratified by region, urban and rural areas, and is not self-weighting since Central, Northern, Upper East and Upper West regions were over-sampled. For reporting national level results, sample weights are used. A more detailed description of the sample design can be found in Appendix A.

Mode of data collection

Face-to-face [f2f]

Research instrument

Four sets of questionnaires were used in the survey: 1) a household questionnaire which was used to collect information on all de jure household members (usual residents), the household, and the dwelling; 2) a women’s questionnaire administered in each household to all women aged 15-49 years; 3) an under-5 questionnaire administered to mothers or caretakers for all children under 5 living in the household; and 4) a men’s questionnaire administered in each third household to all men aged 15-59 years

Cleaning operations

Data were entered using the CSPro software. The data were entered on 20 microcomputers and carried out by 20 data entry operators and 3 data entry supervisors. In order to ensure quality control, all questionnaires were double entered and internal consistency checks were performed. Procedures and standard programs developed under the global MICS4 programme and adapted to the Ghana questionnaire were used throughout. Data capture began in October 2011 and was completed in January 2012. Data were analysed using the Statistical Package for Social Sciences (SPSS) software program, Version 18, and the model syntax and tabulation plans developed by UNICEF were used for this purpose.

Response rate

Of the 12,150 households selected for the sample, 11,970 were found to be occupied. Of these, 11,925 were successfully interviewed for a household response rate of about 100 percent.

In the interviewed households, 10,963 women (aged 15-49 years) were identified. Of these, 10,627 were successfully interviewed, yielding a response rate of 97 percent within interviewed households.

Also 7,626 children under age five were listed in the household questionnaire. Questionnaires were completed for 7,550 of these children, which corresponds to a response rate of 99 percent within interviewed households.

In addition, a men’s questionnaire was used in every third household of the selected sample. For the male survey, 3,511 men aged 15-59 years were identified. Of these, 3,321 were successfully interviewed, yielding a response rate of 94 percent within interviewed households for the male survey

Sampling error estimates

The sample of respondents selected in the Ghana Multiple Indicator Cluster Survey is only one of the samples that could have been selected from the same population, using the same design and size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability between the estimates from all possible samples. The extent of variability is not known exactly, but can be estimated statistically from the survey data.

The following sampling error measures are presented in this appendix for each of the selected indicators: - Standard error (se): Sampling errors are usually measured in terms of standard errors for particular indicators (means, proportions, etc.). Standard error is the square root of the variance of the estimate. The Taylor linearization method is used for the estimation of standard errors. - Coefficient of variation (se/r) is the ratio of the standard error to the value of the indicator, and is a measure of the relative sampling error. - Design effect (deff) is the ratio of the actual variance of an indicator, under the sampling method used in the survey, to the variance calculated under the assumption of simple random sampling. The square root of the design effect (deft) is used to show the efficiency of the sample design in relation to the precision. A deft value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a deft value above 1.0 indicates the increase in the standard error due to the use of a more complex sample design. ?? Confidence limits are calculated to show the interval within which the true value for the population can be reasonably assumed to fall, with a specified level of confidence. For any given statistic calculated from the survey, the value of that statistic will fall within a range of plus or minus two times the standard error (r + 2.se or r – 2.se) of the statistic in 95 percent of all possible samples of identical size and design.

For the calculation of sampling errors from MICS data, SPSS Version 18 Complex Samples module has been used. The results are shown in the tables that follow. In addition to the sampling error measures described above, the tables also include weighted and unweighted counts of denominators for each indicator.

Sampling errors are calculated for indicators of primary interest, for the national level, for the regions, and for urban and rural areas. Three of the selected indicators are based on households, 8 are based on household members, 13 are based on women, and 15 are based on children under 5. Ten are based on men. All indicators presented here are in the form of proportions. Table SE.1 shows the list of indicators for which sampling errors are calculated, including the base population (denominator) for each indicator. Tables (SE.2 to SE.14) show the calculated sampling errors for selected domains.

Data and R scripts to replicate the statistical analysis of this manuscript, presently submitted to Medical Teacher.

Abstract

The high-frequency phone survey of refugees monitors the economic and social impact of and responses to the COVID-19 pandemic on refugees and nationals, by calling a sample of households every four weeks. The main objective is to inform timely and adequate policy and program responses. Since the outbreak of the COVID-19 pandemic in Ethiopia, two rounds of data collection of refugees were completed between September and November 2020. The first round of the joint national and refugee HFPS was implemented between the 24 September and 17 October 2020 and the second round between 20 October and 20 November 2020.

Analysis unit

Household

Kind of data

Sample survey data [ssd]

Sampling procedure

The sample was drawn using a simple random sample without replacement. Expecting a high non-response rate based on experience from the HFPS-HH, we drew a stratified sample of 3,300 refugee households for the first round. More details on sampling methodology are provided in the Survey Methodology Document available for download as Related Materials.

Mode of data collection

Computer Assisted Telephone Interview [cati]

Research instrument

The Ethiopia COVID-19 High Frequency Phone Survey of Refugee questionnaire consists of the following sections:

• Interview Information
• Household Roster
• Camp Information
• Knowledge Regarding the Spread of COVID-19
• Behaviour and Social Distancing - Access to Basic Services
• Employment
• Income Loss
• Coping/Shocks
• Social Relations
• Food Security
• Aid and Support/ Social Safety Nets.

A more detailed description of the questionnaire is provided in Table 1 of the Survey Methodology Document that is provided as Related Materials. Round 1 and 2 questionnaires available for download.

Cleaning operations

DATA CLEANING At the end of data collection, the raw dataset was cleaned by the Research team. This included formatting, and correcting results based on monitoring issues, enumerator feedback and survey changes. Data cleaning carried out is detailed below.

Variable naming and labeling: • Variable names were changed to reflect the lowercase question name in the paper survey copy, and a word or two related to the question. • Variables were labeled with longer descriptions of their contents and the full question text was stored in Notes for each variable. • “Other, specify” variables were named similarly to their related question, with “_other” appended to the name. • Value labels were assigned where relevant, with options shown in English for all variables, unless preloaded from the roster in Amharic.

Variable formatting: • Variables were formatted as their object type (string, integer, decimal, time, date, or datetime). • Multi-select variables were saved both in space-separated single-variables and as multiple binary variables showing the yes/no value of each possible response. • Time and date variables were stored as POSIX timestamp values and formatted to show Gregorian dates. • Location information was left in separate ID and Name variables, following the format of the incoming roster. IDs were formatted to include only the variable level digits, and not the higher-level prefixes (2-3 digits only.)
• Only consented surveys were kept in the dataset, and all personal information and internal survey variables were dropped from the clean dataset. • Roster data is separated from the main data set and kept in long-form but can be merged on the key variable (key can also be used to merge with the raw data). • The variables were arranged in the same order as the paper instrument, with observations arranged according to their submission time.

Backcheck data review: Results of the backcheck survey are compared against the originally captured survey results using the bcstats command in Stata. This function delivers a comparison of variables and identifies any discrepancies. Any discrepancies identified are then examined individually to determine if they are within reason.

Data appraisal

The following data quality checks were completed: • Daily SurveyCTO monitoring: This included outlier checks, skipped questions, a review of “Other, specify”, other text responses, and enumerator comments. Enumerator comments were used to suggest new response options or to highlight situations where existing options should be used instead. Monitoring also included a review of variable relationship logic checks and checks of the logic of answers. Finally, outliers in phone variables such as survey duration or the percentage of time audio was at a conversational level were monitored. A survey duration of close to 15 minutes and a conversation-level audio percentage of around 40% was considered normal. • Dashboard review: This included monitoring individual enumerator performance, such as the number of calls logged, duration of calls, percentage of calls responded to and percentage of non-consents. Non-consent reason rates and attempts per household were monitored as well. Duration analysis using R was used to monitor each module's duration and estimate the time required for subsequent rounds. The dashboard was also used to track overall survey completion and preview the results of key questions. • Daily Data Team reporting: The Field Supervisors and the Data Manager reported daily feedback on call progress, enumerator feedback on the survey, and any suggestions to improve the instrument, such as adding options to multiple choice questions or adjusting translations. • Audio audits: Audio recordings were captured during the consent portion of the interview for all completed interviews, for the enumerators' side of the conversation only. The recordings were reviewed for any surveys flagged by enumerators as having data quality concerns and for an additional random sample of 2% of respondents. A range of lengths were selected to observe edge cases. Most consent readings took around one minute, with some longer recordings due to questions on the survey or holding for the respondent. All reviewed audio recordings were completed satisfactorily. • Back-check survey: Field Supervisors made back-check calls to a random sample of 5% of the households that completed a survey in Round 1. Field Supervisors called these households and administered a short survey, including (i) identifying the same respondent; (ii) determining the respondent's position within the household; (iii) confirming that a member of the the data collection team had completed the interview; and (iv) a few questions from the original survey.

R code for simulating sampling strategies. Description: R code that creates an exemplary data set and simulates the sampling strategies. (R 26Â kb)

Abstract

The SHDS is a national sample survey designed to provide information on population, birth spacing, reproductive health, nutrition, maternal and child health, child survival, HIV/AIDS and sexually transmitted infections (STIs), in Somalia.. The main objective of the SHDS was to provide evidence on the health and demographic characteristics of the Somali population that will guide the development of programmes and formulation of effective policies. This information would also help monitor and evaluate national, sub-national and sector development plans, including the Sustainable Development Goals (SDGs), both by the government and development partners. The target population for SHDS was the women between 15 and 49 years of age, and the children less than the age of 5 years

Geographic coverage

The SHDS 2020 was a nationally representative household survey.

Analysis unit

The unit analysis of this survey are households, women aged 15-49 and children aged 0-5

Universe

This sample survey covered Women aged 15-49 and Children aged 0-5 years.

Kind of data

Sample survey data [ssd]

Sampling procedure

Sample Design The sample for the SHDS was designed to provide estimates of key indicators for the country as a whole, for each of the eighteen pre-war geographical regions, which are the country's first-level administrative divisions, as well as separately for urban, rural and nomadic areas. With the exception of Banadir region, which is considered fully urban, each region was stratified into urban, rural and nomadic areas, yielding a total of 55 sampling strata. All three strata of Lower Shabelle and Middle Juba regions, as well as the rural and nomadic strata of Bay region, were completely excluded from the survey due to security reasons. A final total of 47 sampling strata formed the sampling frame. Through the use of up-to-date, high-resolution satellite imagery, as well as on-the-ground knowledge of staff from the respective ministries of planning, all dwelling structures were digitized in urban and rural areas. Enumeration Areas (EAs) were formed onscreen through a spatial count of dwelling structures in a Geographic Information System (GIS) software. Thereafter, a sample ground verification of the digitized structures was carried out for large urban and rural areas and necessary adjustments made to the frame.

Each EA created had a minimum of 50 and a maximum of 149 dwelling structures. A total of 10,525 EAs were digitized: 7,488 in urban areas and 3,037 in rural areas. However, because of security and accessibility constraints, not all digitized areas were included in the final sampling frame-9,136 EAs (7,308 in urban and 1,828 in rural) formed the final frame. The nomadic frame comprised an updated list of temporary nomadic settlements (TNS) obtained from the nomadic link workers who are tied to these settlements. A total of 2,521 TNS formed the SHDS nomadic sampling frame. The SHDS followed a three-stage stratified cluster sample design in urban and rural strata with a probability proportional to size, for the sampling of Primary Sampling Units (PSU) and Secondary Sampling Units (SSU) (respectively at the first and second stage), and systematic sampling of households at the third stage. For the nomadic stratum, a two-stage stratified cluster sample design was applied with a probability proportional to size for sampling of PSUs at the first stage and systematic sampling of households at the second stage. To ensure that the survey precision is comparable across regions, PSUs were allocated equally to all regions with slight adjustments in two regions. Within each stratum, a sample of 35 EAs was selected independently, with probability proportional to the number of digitized dwelling structures. In this first stage, a total of 1,433 EAs were allocated (to urban - 770 EAs, rural - 488 EAs, and nomadic - 175 EAs) representing about 16 percent of the total frame of EAs. In the urban and rural selected EAs, all households were listed and information on births and deaths was recorded through the maternal mortality questionnaire. The data collected in this first phase was cleaned and a summary of households listed per EA formed the sampling frames for the second phase. In the second stage, 10 EAs were sampled out of the possible 35 that were listed, using probability proportional to the number of households. All households in each of these 10 EAs were serialized based on their location in the EA and 30 of these households sampled for the survey. The serialization was done to ensure distribution of the households interviewed for the survey in the EA sampled. A total of 220 EAs and 150 EAs were allocated to urban and rural strata respectively, while in the third stage, an average of 30 households were selected from the listed households in every EA to yield a total of 16,360 households from 538 EAs covered (220 EAs in urban, 147 EAs in rural and 171 EAs in nomadic) out of the sampled 545 EAs. In nomadic areas, a sample of 10 EAs (in this case TNS) were selected from each nomadic stratum, with probability proportional to the number of estimated households. A complete listing of households was carried out in the selected TNS followed by the selection of 30 households for the main survey interview. In those TNS with less than 30 households, all households were interviewed for the main survey. All eligible ever-married women aged 12 to 49 and never-married women aged 15 to 49 were interviewed in the selected households, while the household questionnaire was administered to all households selected. The maternal mortality questionnaire was administered to all households in each sampled TNS.

Mode of data collection

Face-to-face [f2f]

Response rate

A total of 16,360 households were selected for the sample, of which 15,870 were occupied. Of the occupied households, 15,826 were successfully interviewed, yielding a response rate of 99.7 percent. The SHDS 2020 interviewed 16,486 women-11,876 ever-married women and 4,610 never-married women.

Sampling error estimates

Sampling errors are important data quality parameters which give measure of the precision of the survey estimates. They aid in determining the statistical reliability of survey estimates. The estimates from a sample survey are affected by two types of errors: non-sampling errors and sampling errors. Non-sampling errors are the results of mistakes made in implementing data collection and data processing, such as failure to locate and interview the correct household, misunderstanding of the questions on the part of either the interviewer or the respondent, and data entry errors. Although numerous efforts were made during the implementation of the Somaliland Health and Demographic Survey ( SHDS 2020) to minimise this type of error, non-sampling errors are impossible to avoid and difficult to evaluate statistically. Sampling errors, on the other hand, can be evaluated statistically. The sample of respondents selected in the SHDS 2020 is only one of many samples that could have been selected from the same population, using the same design and sample size. Each of these samples would yield results that differ somewhat from the results of the actual sample selected. Sampling errors are a measure of the variability among all possible samples. Although the degree of variability is not known exactly, it can be estimated from the survey results. Sampling error is usually measured in terms of the standard error for a particular statistic (mean, percentage, etc.), which is the square root of the variance. The standard error can be used to calculate confidence intervals within which the true value for the population can reasonably be assumed to fall. For example, for any given statistic calculated from a sample survey, the value of that statistic will fall within a range of plus or minus two times the standard error of that statistic in 95% of all possible samples of identical size and design. If the sample of respondents had been selected by simple random sampling, it would have been possible to use straightforward formulas for calculating sampling errors. However, the SHDS 2020 sample was the result of a multi-stage stratified design, and, consequently, it was necessary to use more complex formulas. The variance approximation procedure that account for the complex sample design used R program was estimated sampling errors in SHDS which is Taylor series linearization. The non-linear estimates are approximated by linear ones for estimating variance. The linear approximation is derived by taking the first-order Tylor series approximation. Standard variance estimation methods for linear statistics are then used to estimate the variance of the linearized estimator. The Taylor linearisation method treats any linear statistic such as a percentage or mean as a ratio estimate, r = y/x, where y represents the total sample value for variable y and x represents the total number of cases in the group or subgroup under consideration

Data appraisal

• Household age distribution
• Age distribution of eligible and interviewed women
• Pregnancy- related mortality trends Note: See detailed data quality tables in APPENDIX C of the report.

Abstract

The Geneal Household Survey is a brainchild of the National Bureau of Statistics (NBS) and is often referred to as Regular survey carried out on quarterly basis by the NBS over the years. In recent times, starting from 2004 to be precise, there is a collaborative effort between the NBS and the CBN in 2004 and 2005 and in 2006 the collaboration incorporated Nigerian Communications commission (NCC). The main reason of for conducting the survey was to enable the collaborating agencies fulfil their mandate in the production of current and credible statistics, to monitor and evaluate the status of the economy and the various government programmes such as the National Economic Empowerment and Development Strategy (NEEDS) and the Millennium Development Goals (MDGs).
The collaborative survey also assured the elimination of conflicts in data generated by the different agencies and ensured a reliable, authentic national statistics for the country.

Geographic coverage

National Zone State Local Government

Analysis unit

Household based

Universe

Household

Kind of data

Sample survey data [ssd]

Sampling procedure

The GHS was implemented as a NISH module. Six replicates were studied per State while three replicates were studied in the FCT, Abuja. With a fixed-take of 10 HUs systematically selected per EA, 600 HUs thus were selected for interview per State and 300 for FCT, Abuja. Hence, nationally, a total of 21,900 HUs drawn from the 2,190 cut across the rural and urban sectors.

Introduction: The sample design for the survey derives from the National Integrated Survey of Household (NISH) developed by National Bureau of Statistics (NBS). The NISH design employed a replicated sampling design that is technique by which many sample (replicates) were selected independently from a population such that each replicate sample represents the population.

Essentially, the NISH sample design is a 2-stage replicated and rotated cluster sample design with Enumeration Areas (EAs) as first stage sampling unit or Primary Sampling Unit (PSU) and Housing Units the second stage sampling units (secondary sampling units). Generally, for each state of the Federation, the NISH Master Sample is made up of 120 EAs drawn in12 replicates. A replicate consists of 10 EAs.

Selection Procedures:

The EAs demarcated by the National Population Commission (NpopC) for the 1991 Population Census served as the primary Sample Frame for the design.

First Stage Selection:

Sixty EAs were selected with equal probability from the list of EAs in each state of the federation and 30 EAs for FCT, Abuja. The selected EAs cuts across rural and urban sectors. The study EAs for the collaborative survey were drawn from replicates 7,8,9,10,11 and 12 of the master sample of each state.

Second Stage Selection: In each selected EA, a listing of housing units was carried out. The result provided the frame for the second stage selection. Ten housing units were selected systematically in each EA after the completion of the listing exercise. Thereafter, all the households within the selected HUs were interviewed using GHS questionnaire. Out of the expected 2,190 EAs, 1,883 were studied. Out of the 21,900 housing units expected to be covered, 18,826 were canvassed.

Sampling deviation

Variance Estimate (Jackknife Method) Estimating variances using the Jackknife method will require forming replicate from the full sample by randomly eliminating one sample cluster [Enumeration Area (EA) at a time from a state containing k EAs, k replicated estimates are formed by eliminating one of these, at a time, and increasing the weight of the remaining (k-1) EAs by a factor of k/(k-1). This process is repeated for each EA.

For a given state or reporting domain, the estimate of the variance of a rate, r, is given by k Var(r ) = (Se)2 = 1 S (ri - r)2 k(k-1) i=1

where (Se) is the standard error, k is the number of EAs in the state or reporting domain.

r is the weighted estimate calculated from the entire sample of EAs in the state or reporting domain.
ri = kr - (k - 1)r(i), where

r(i) is the re-weighted estimate calculated from the reduced sample of k-1 EAs.

To obtain an estimate of the variance at a higher level, say, at the national level, the process is repeated over all states, with k redefined to refer to the total number of EAs (as opposed to the number in the states).

Mode of data collection

Face-to-face [f2f]

Research instrument

The questionnaire for the GHS is a structured questionnaire based on household characteristics with some modifications and additions. The House project module is a new addition and some new questions on ICT. The questionnaires were scaned This section deals with the characteristics of the socio-economic data of Nigerian population, such as demography, education, employment, health, housing condition, fertility, mortality etc. Demographic factors are both determinants and consequences of economic and social development. It has been shown that the study of demographic variables yield important information on the inventories of human resources that are needed for effective development planning.

Cleaning operations

The data editing is in 2 phases namely manual editing before the questionnaires were scanned. This involved using editors at the various zones to manually edit and ensure consistency in the information on the questionnaire. The second editing is the computer editing, this is the cleaning of the already scanned data.

Response rate

On National basis, 85.98 percent response rate was acheived at EA level while 85.96 percent was acheived at housing units level.

Sampling error estimates

No sampling error estimate

Data appraisal

QUALITY CONTROL AND RETRIEVAL OF RECORD
The Quality Control measures were carried out during the survey, essentially to ensure quality of data. There were three levels of supervision involving the supervisors at the first level, CBN staff, NBS State Officers and Zonal Controllers at second level and finally the NBS/NCC Headquarter staff constituting the third level supervision. Field monitoring and quality check exercises were also carried out during the period of data collection as part of the quality control measures.

Number of observations and percent (bracket) correct classified for female and male sample population using discriminant analysis.

Number of observations and percent-classified (in brackets) into the site using a non-parametric discriminant for both male and female sample chicken populations.

Class means on canonical variables of female and male chickens.