Demographic and Health Survey 2009
Guyana, 2009
Get MicrodataIdentification
GUY_2009_DHS_v01_M
Demographic and Health Survey 2009
| Name | Country code |
|---|---|
| Guyana | GUY |
Demographic and Health Survey (standard) - DHS V
The 2009 GDHS is the first household-based, comprehensive survey on demographics and health (especially maternal and child health) to be carried out in Guyana.
The primary objective of the 2009 GDHS was to collect information on characteristics of the households and their members, including exposure to malaria and tuberculosis; infant and child mortality; fertility and family planning; pregnancy and postnatal care; childhood immunization, health, and nutrition; marriage and sexual activity; and HIV/AIDS indicators.
Other objectives of the 2009 GDHS included (1) supporting the dissemination and utilization of the results in planning, managing, and improving family planning and health services in the country and (2) enhancing the survey capabilities of the institutions involved to facilitate surveys of this type in the future.
The 2009 GDHS sampled 5,632 households and completed interviews with 4,996 women age 15-49 and 3,522 men age 15-49. Three questionnaires were used for the 2009 GDHS: the Household Questionnaire, the Women's Questionnaire, and the Men's Questionnaire. The content of these questionnaires was based on the model questionnaires developed by the MEASURE DHS program of ICF Macro.
The primary objective of the 2009 GDHS was to collect information on the following topics:
- Characteristics of households and household members
- Fertility and reproductive preferences, infant and child mortality, and family planning
- Health-related matters, such as breastfeeding, antenatal care, children's immunizations, and childhood diseases
- Marriage, sexual activity, and awareness and behavior regarding HIV and other sexually transmitted infections (STIs)
- The nutritional status of mothers and children, including anthropometry measurements and anemia testing Other complementary objectives of the 2009 GDHS were:
- To support dissemination and utilization of the results in planning, managing, and improving family planning and health services in the country
- To enhance the survey capabilities of the institutions involved to facilitate their use of surveys of this type in the future
MAIN RESULTS
FERTILITY
Fertility Levels and Differentials
If fertility were to remain constant in Guyana, women would bear, on average, 2.8 children by the end of their reproductive lifespan. The total fertility rate (TFR) is close to replacement level in urban areas (2.1 children per woman), and higher in the rural areas (3.0 children per woman). The TFR in the Interior area (6.0 children) is more than twice as high as the TFR in the Coastal area (2.4 children per woman) and is three times the fertility in the Georgetown (urban) area (2.0 children). The TFRs for women in the Interior area are significantly higher for all age groups.
Fertility Preferences
Fifty-six percent of currently married women reported that they don't want to have a/another child, and five percent are already sterilized. The figures for men are 51 and 1 percent, respectively. The desire to stop childbearing increases rapidly as the number of children increases. Among respondents with one child, around one in five wants no more children. Among those with three children, about eight in ten women and seven in ten men want no more children.
FAMILY PLANNING
Use of Contraception
Forty-three percent of women who are currently married or in union are currently using a contraceptive method, mainly a modern method (40 percent). The methods most commonly used by currently married women are the male condom (13 percent), the pill (9 percent), and the IUD (7 percent). Female sterilization and injectables are each used by 5 percent of women. The 2009 GDHS prevalence rate of 43 percent represents an increase of 8 percentage points since the 2005 GAIS (35 percent). Most of the increase was in condom use, injectables, and female sterilization.
Unmet Need for Family Planning
Twenty-nine percent of currently married women have an unmet need for family planning, mostly for limiting births (19 percent) compared with spacing (10 percent). Because 43 percent of married women are currently using a contraceptive method (met need), the total demand for family planning is estimated at 71 percent of married women (22 percent for spacing, 49 percent for limiting). As a result, only 60 percent of the total demand for family planning is met.
MATERNAL HEALTH
Antenatal Care
Among women who had a birth in the five years preceding the survey, 92 percent received antenatal care (ANC) from a skilled health provider for their most recent birth (51 percent from a nurse/midwife and 35 percent from a doctor). Older mothers (35-49 years) are less likely to receive antenatal care by a skilled health provider than younger mothers. Eighty-six percent of women with no education received ANC from a skilled health provider compared with 95 percent of women with more than secondary education.
Delivery Care
Overall, 92 percent of births in the five years preceding the survey were assisted by a skilled birth provider, mainly by a nurse or midwife (56 percent), followed by a doctor (31 percent). Births to mothers under age 35 and lower order births are more likely to have assistance at delivery by a skilled provider than births to older mothers and higher order births. By residence, births in Urban areas are more likely than those in Rural areas, and births in the Coastal area are more likely than births in the Interior area, to be assisted by a skilled health provider. The percentage of births assisted by a skilled provider ranges from a low of 57 percent in Region 9 to a high of 98 percent in Region 4. Births to mothers who have more education and births in the higher wealth quintiles are more likely to be assisted by a skilled provider than other births. Almost all births to mothers with more than secondary education (98 percent) are assisted by a skilled provider compared with 71 percent of births to mothers with no education.
Caesarean section
One in eight births (13 percent) in the five years preceding the survey was delivered by caesarean section. The prevalence of C-section delivery increases steadily with mother's age and decreases with birth order. Regions 1, 6, 7, and 9 have the lowest levels of deliveries by C-section (2-5 percent) and Region 3 has the highest level (23 percent). The percentage of births delivered by C-section increases with a mother's education and generally increases with her wealth.
CHILD HEALTH
Infant and Child Mortality
Childhood mortality rates in Guyana are relatively low. For every 1,000 live births, 38 children die during the first year of life (infant mortality), and 40 children die during the first five years (under-age 5 mortality). Almost two-thirds of deaths in the first five years (25 deaths per 1,000 live births) take place during the neonatal period (the first month of life). The mortality rate after the first year of life up to age 5 (child mortality) is also very low at 3 deaths per 1,000 live births. The 2009 GDHS mortality data do not show any clear trends over time. However, mortality data have to be interpreted with caution because sampling errors associated with mortality estimates are large.
Vaccination Coverage
Overall, 63 percent of Guyanese children age 18-29 months are fully immunized, and only 5 percent of the children received no vaccinations at all. Looking at coverage for specific vaccines, 94 percent of children received the BCG vaccination, 92 percent received the first dose of pentavalent vaccine, and 78 percent received the first polio dose (Polio 1). Coverage for the pentavalent and polio vaccinations declines with subsequent doses; 85 percent of children received the recommended three doses of pentavalent vaccine, and 70 percent received three doses of polio. These figures reflect dropout rates of 8 percent for the pentavalent vaccine and 11 percent for polio; the dropout rate represents the proportion of children who received the first dose of a vaccine but who did not get the third dose. Eighty-two percent of children are vaccinated against measles, and 79 percent of children have been vaccinated against yellow fever.
Illnesses and Treatment
Acute Respiratory Infections (ARI)
Five percent of children under age 5 had symptoms of acute respiratory infection (ARI) in the two weeks preceding the survey. Among children with symptoms of ARI, advice or treatment was sought from a health facility or provider for 65 percent, and antibiotics were prescribed as treatment for 18 percent (data not shown).
Fever
Fever was found to be moderately frequent in children under age 5 in Guyana (20 percent), ranging from 17 percent in children under 6 months to about 26 percent in children 12-17 months.. Most of the children under age 5 with fever (59 percent) were taken to a health facility or a health provider for their most recent episode of fever. Overall, about one in five children with fever (21 percent) received antibiotics, and 6 percent received antimalarial drugs.
Diarrhea
Overall, about 10 percent of children were reported to have diarrhea in the two weeks immediately before the survey, with just 1 percent reporting bloody diarrhea. Overall, about six in ten children under age 5 with diarrhea (59 percent) were taken to a health facility or health provider for advice or treatment. Male children (55 percent) are less likely than female children (63 percent) to be taken for treatment or advice to a health facility or provider. Additionally, children living in the Coastal area are much less likely to be taken for treatment or advice (50 percent) than children in the Interior area (79 percent).
NUTRITION OF CHILDREN
Height and Weight
Almost one in five children (18 percent) under age 5 is short for age or stunted, and one in twenty (5 percent) is severely stunted. As expected, stunting, which reflects chronic malnutrition, rises with age during the first year. Stunting is lower among children whose mothers have more than secondary education (16 percent). Children in Rural areas are almost twice as likely to be stunted as children in Urban areas (20 and 11 percent, respectively). The highest levels of stunting are found among children in the Interior area (35 percent).
Anemia
Overall, about four in ten (39 percent) children age 6-59 months have some level of anemia, including 23 percent of children who are mildly anemic, 15 percent who are moderately anemic, and less than 1 percent with severe anemia. Prevalence of any anemia is highest for children 9-11 months (74 percent) and lowest for those 36-59 months (25 to 28 percent). More than half of children in Region 1 are anemic (51 percent) compared with three in ten (30 percent) in Region 8. The percentage of children with anemia is lowest among children of mothers with secondary or higher education (38-40 percent) and among children of mothers in the highest wealth quintile (32 percent).
Malaria
Eighty-nine percent of households own a mosquito net, whether treated or untreated, and 66 percent of households own more than one net. Rural households are more likely to own at least one net than urban households (90 percent versus 85 percent). About nine in ten households (89 percent) in the malaria-endemic regions (Regions 1, 7, 8, and 9) have at least one mosquito net.
HIV/AIDS
Knowledge of HIV Prevention Methods
Knowledge of AIDS is almost universal in Guyana-97 percent of women and men have heard of AIDS. There are minor variations in knowledge of AIDS by age, marital status, or residence. The only exception is the level of knowledge in the Interior area, which is the lowest for both women (89 percent) and men (95 percent).
Beliefs about AIDS
About nine in ten Guyanese adults know that a healthy-looking person can have the AIDS virus (87 percent of women and men) or that AIDS cannot be transmitted by supernatural means (87 percent of women and 88 percent of men). About three-quarters of women (73 percent) and two-thirds of men (65 percent) are aware that the AIDS virus cannot be transmitted through mosquito bites. Furthermore, 84 percent of women and 79 percent of men know that the AIDS virus cannot be contracted by sharing food with a person who has AIDS. These findings show that the two most common local misconceptions are that the HIV virus can be transmitted (1) by mosquito bites and (2) by sharing food with someone with AIDS.
Mother-to-Child Transmission
About eight in ten women (79 percent) and seven in ten men (67 percent) know that HIV can be transmitted by breastfeeding. Sixty-eight percent of women and 54 percent of men are aware that the risk of mother-to-child transmission (MTCT) can be reduced by the mother taking drugs during pregnancy.
Attitudes toward Negotiating Safer Sex
Almost nine in ten respondents (89 percent of women and 88 percent of men) feel that a wife is justified in refusing to have sexual intercourse with her husband if she knows that he has a sexually transmitted disease. Ninety-six percent of women and men agree that a woman is justified in either refusing sexual intercourse with her husband or in asking him to use a condom if she knows that he has an STI.
Attitudes toward Educating Children on Condom Use
Overall, more than eight in ten women (81 percent) and men (86 percent) age 18-49 agree that children age 12-14 should be taught to use condoms to avoid AIDS. Older respondents age 40-49 are slightly less likely than younger respondents to support education of children age 12-14 about condom use to prevent AIDS. Women and men living in the Coastal area (82 and 86 percent, respectively) are more likely than women and men living in the Interior area (73 and 82 percent, respectively) to agree about education on condom use of children age 12-14.
Higher-risk Sex
A larger proportion of men (10 percent) than women (1 percent) reported having had more than one sexual partner in the 12 months preceding the survey. Additionally, a higher percentage of men (30 percent) than women (17 percent) reported having had sex with a person who was neither their spouse nor their cohabiting partner (higher-risk sex) in the year before the survey.
HIV/AIDS-Related Knowledge and Sexual Behavior among Young Adults
About half of respondents age 15-24 (54 percent of women and 47 percent of men) have a comprehensive knowledge of AIDS (i.e., they know that people can reduce their chances of getting the AIDS virus by having sex with only one uninfected, faithful partner and by using condoms consistently; know that a healthy-looking person can have the AIDS virus; and know that HIV cannot be transmitted by mosquito bites or by supernatural means).
Sample survey data
- Household
- Children under five years
- Women age 15-49
- Men age 15-49
Scope
The Guyana Demographic and Health Survey 2009 covers the following topics:
- Alcohol Consumption
- Anemia Questions
- Anemia Testing
- Anthropometry
- Birth Registration
- Early Childhood Education
- GPS/Georeferenced
- HIV Behavior
- HIV Knowledge
- Iodine salt test
- Malaria Module –Includes bednet inventory
- Men's Survey
- Micronutrients
- Reproductive Calendar
- Social Marketing
- TB Questions
- Tobacco Use
- Vitamin A Questions
- Women's Status
- Youth
Coverage
- National
- Urban areas (Georgetown urban and other urban, separately) and Rural areas of Guyana
- Coastal areas (Coastal urban and Coastal rural, separately) and Interior areas of Guyana
- Ten regions of Guyana
The population covered by the 2009 GDHS is defined as the universe of all women and men age 15-49 in Guyana. A probability sample of households was selected, and all women and men age 15-49 identified in the households were eligible to be interviewed.
Producers and sponsors
| Name |
|---|
| Bureau of Statistics (BOS) |
| Ministry of Health (MOH) |
| Name | Role |
|---|---|
| ICF Macro | Technical assistance |
| Name | Role |
|---|---|
| United States Agency for International Development | Financial support |
Sampling
SAMPLE FRAME
Administratively, Guyana is divided into 10 regions. For census purposes, each region is divided into enumeration areas (EAs), either urban or rural. The available list of EAs has information on the population size and number of households for each EA. This information and the available demarcated cartographic material from the last census for each EA constitute an adequate sample frame for the selection of EA as the primary sampling units for the 2009 GDHS.
SAMPLE SELECTION
The 2002 Population and Housing Census constituted the frame for the 2009 GDHS sample design. The 2009 GDHS sample of households was selected using a stratified, two-stage cluster design consisting of 330 clusters, which are required for a sample of about 6,590 households. The first-stage units (primary sampling units or PSUs) are the enumeration areas (EAs) used for the 2002 Population and Housing Census. After allocating the total sample among the major regions in the most optimal way possible, the number of EAs (clusters) in each domain region was calculated by dividing its total allocated number of households by the sample take (25 households for selection per EA).
For the first stage, in each major domain, clusters are selected systematically, with probability proportional to size. The selection is done using the following formula:
P1i = (b mi / S mi)
where
- b : number of EAs in the 2009 GDHS assigned to a given domain region mi: measure of size ( number of households ) of the ith EA S
- mi : total measure of size (total number of households) for the corresponding domain region
In each selected EA, a household listing operation was carried out prior to fieldwork, and households were selected to achieve a fixed sample take per cluster.
For the second stage, in the ith cluster in a given area combination (location by residence), a fixed number of households (c) were selected out of the total households (Li) found in the 2009 GDHS listing process.
Then the household probability in the selected ith cluster can be expressed as P2i = (c /Li)
The overall probability of the households in the ith cluster could be calculated as fi = P1i*P2i
The sampling design weight for the ith cluster is given as 1/fi = 1/(P1i*P2i)
SAMPLE ALLOCATION
Result shows the percent distribution of the population of Guyana by urban-rural residence for each of the 10 regions and the sample allocation for the 2009 GDHS: the number of households and the resulting number of primary sampling units.
Seventy-one percent of the population in Guyana resides in Rural areas. The Rural areas of Regions 3, 4, and 6 are the most densely populated. Regions 1, 7, 8, 9, and 10 each account for less than 4 percent of the rural population, and Region 5 accounts for 10 percent. On the other hand, Region 4 includes almost two-thirds (64 percent) of the urban population. Because of these differences, the sample was not allocated by region according to the actual distribution of the population. A minimum of 400 households were allocated to each region. The largest number of households was allocated to Region 4 (1,600) and Region 6 (1,000). Around 600-650 households were allocated to each of Regions 2, 3, and 10 for a total of 6,590 households. Out of this total, around 3,000 households were allocated to the Coastal (rural) domain.
The allocated number of clusters per domain is calculated in Table A.1 by dividing by 25 the number of households allocated for each domain, given that the sample design calls for the selection of 25 households per cluster in the second stage.
As a result of the non-proportional allocation of the number of EDs for the urban-rural domains, the household sample for the 2009 GDHS is not a self-weighted sample. Weights were calculated to ensure that the distribution of respondents (weighted percent and weighted number) reflects the actual representation at the national level.
In the second stage, 25 households were selected by systematic random sampling from the full updated list of households for each of the selected EDs for a total of 6,590 households. All women and men age 15-49 who were either permanent residents of the households in the 2009 GDHS sample or visitors who were present in the household on the night before the survey were eligible to be interviewed in the survey.
Of the 6,376 selected households, 6,042 households were occupied, and a total of 5,632 households were interviewed, yielding a household response rate of 93 percent. By residence, the household response rate is lowest in urban areas (91 percent), especially in Georgetown (86 percent), and highest in Interior areas of the country (96 percent). By region, the household response rate ranges from 89 percent in Region 4 to 99 percent in Region 8.
In the households interviewed, a total of 5,547 eligible women were identified. Interviews were completed with 4,996 of these women, yielding a response rate for women of 90 percent. The women's response rates were lowest in the Interior areas (86 percent) and in Region 1 (83 percent) and highest in the Coastal areas (92 percent) and Region 2 (95 percent).
Of the 4,553 eligible men identified in the same interviewed households, a total of 4,553 men were identified. Interviews were conducted with only 3,522 men, yielding a response rate for men of 77 percent. Men from the Interior area (70 percent) and from Region 1 (62 percent) have the lowest response rates, while men in Urban and Coastal (urban) areas (82 percent, each) have the highest response rates.
The primary reason for non-response among eligible women and men was the failure to find individuals at home despite repeated visits to the household. The substantially lower response rate for men reflects the more frequent and longer absences of men from the household, principally related to employment and lifestyle activities (data not shown).
The weighted numbers are shown because weighting is necessary for the calculation of most indicators-percent distributions, percentages, and rates. This is because the sample was not allocated by region according to the actual distribution of the population. Instead, the sample was allocated to provide a sufficient number of respondents for each region to allow calculation of most survey variables at the regional level. The unweighted numbers are the actual numbers of interviews. Some subgroups shown may include comparatively small numbers of respondents (e.g., respondents with no education and those in some religious and ethnic groups). In some tables in this report, estimates for these subgroups are not shown if the unweighted number of cases is fewer than 25. Also, estimates based on 25 to 49 unweighted cases are shown enclosed in parentheses.
Although only 1,179 women were interviewed in Region 4 (24 percent of the total unweighted number of all women), the weighted number is 2,168 women (43 percent of the total weighted number of women). On the other hand, 280 women were interviewed in Region 9 (6 percent of the total unweighted number of all women), and the weighted number is 78 women (2 percent of the total weighted number of women).
The regional distribution of the population shows no marked differences by sex, with around three in ten women (30 percent) and men (27 percent) living in Urban areas, with two-thirds of these living in Georgetown. Approximately nine in ten respondents of both sexes (90 percent of women and 89 percent of men) live in the Coastal areas, with the majority (60 percent of women and 62 percent of men) living in the Coastal (rural) areas. Only one-tenth of the respondents (10 percent of women and 11 percent of men) live in the Interior areas of the country.
Survey instrument
Three questionnaires were used for the 2009 GDHS: the Household Questionnaire, the Women's Questionnaire, and the Men's Questionnaire. The contents of these questionnaires were based on the model questionnaires developed by the MEASURE DHS program. In consultation with USAID/Guyana, technical institutions, and local and international organizations, the contents of the model questionnaires were modified to reflect relevant issues in population, family planning, and other health issues in Guyana.
a) The Household Questionnaire was used to list all the usual members and visitors in the selected households. The following basic information was collected:
- Characteristics of each person listed, including age, sex, education, and relationship to the head of the household. As a result, women and men who would be eligible for a subsequent individual interview could be identified.
- Characteristics of the household's dwelling unit, such as the source of water, type of toilet facilities, materials used for the floor of the house, ownership of various durable goods, and ownership and use of mosquito nets.
- Height and weight measurements of women age 15-49 and children under age 6, as well as the results of anemia testing.
b) The Women's Questionnaire was used to collect information from all women age 15-49. Eligible women were asked questions on the following topics:
- Background characteristics (e.g., education, residential history, media exposure)
- Birth history and childhood mortality
- Knowledge and use of family planning methods
- Fertility preferences
- Antenatal and delivery care for children born after January 2004
- Breastfeeding and infant feeding practices
- Vaccinations and illnesses for children born after January 2004
- Marriage and sexual activity
- Woman's work and husband's background characteristics
- Awareness and behavior regarding AIDS and other STIs
c) The Men's Questionnaire was administered to all men age 15-49 living in households included in the 2009 GDHS sample. The Men's Questionnaire collected information similar to that of the Women's Questionnaire but was shorter because it did not include a reproductive history or questions on maternal and child health and nutrition. The following topics were addressed:
- Background characteristics (e.g., education, residential history, media exposure)
- Reproductive history and basic health questions about last birth
- Knowledge and use of family planning methods
- Fertility preferences
- Marriage and sexual activity
- Employment and gender roles
- Awareness and behavior regarding AIDS and other sexually transmitted infections (STIs)
Data collection
| Start | End |
|---|---|
| 2009-03 | 2009-07 |
Staff from the Bureau of Statistics was responsible for coordinating and supervising fieldwork activities. Two nurses supervised the anthropometry and anemia testing. ICF Macro staff participated in the survey, assisting with questionnaire design, training for data collection, data processing and tabulation, field supervision of interviews, and training in anthropometry and anemia testing.
A training of trainers took place in early December 2008. One hundred and twelve candidates (50 men and 62 women) participated in the main survey training of interviewers, supervisors, and field editors, which took place on December 9-19, 2008, and January 12-22, 2009. Special parallel training sessions for supervisors and editors were conducted on January 19-21. All participants received training in interviewing techniques and became acquainted with the contents of the survey questionnaires. The training was conducted following standard DHS procedures and included class presentations, mock interviews, and tests in which the actual 2009 GDHS questionnaires were used. During the last week of January 2009, the editors of each team received training and instruction on how to use measuring boards and scales to conduct anthropometric measurements (height and weight) of women and young children and on how to conduct anemia testing. Because the beginning of fieldwork was delayed, a refresher training course was conducted on February 24-26, 2009.
Data collection for the 2009 GDHS took place over a five-month period from March 1 through late July 2009 and was carried out by 16 interviewing teams. Each team consisted of one team supervisor, one field editor, two female and two male interviewers, and one driver. In total, 96 fieldworkers completed work first in the Coastal enumeration districts (Regions 2, 3, 4, 5, and 10). At a later stage, 60 fieldworkers were selected to work in the Interior districts (Regions 1, 7, 8, 9, and 10).
Data processing
The processing of the 2009 GDHS questionnaires began on March 16, 2009, shortly after fieldwork commenced. Completed questionnaires were submitted periodically to BOS offices in Georgetown, where they were edited by data processing personnel who had been trained specifically for this task by ICF Macro staff. Data processing was done concurrently with fieldwork using CSPro, a program specially developed for use in complex surveys. The concurrent processing of the data was an advantage because field check tables were produced periodically to advise field teams of any problems that were detected during data processing. Data processing was completed in late August 2009.
Data appraisal
The estimates from a sample survey are affected by two types of errors: (1) non-sampling errors, and (2) 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 2009 Guyana Demographic and Health Survey (GDHS) to minimize 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 2009 GDHS is only one of many samples that could have been selected from the same population, using the same design and expected 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 all possible samples. Although the degree of variability is not known exactly, it can be estimated from the survey results.
A 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 percent of all possible samples of identical size and design.
If the sample of respondents had been selected as a simple random sample, it would have been possible to use straightforward formulas for calculating sampling errors. However, the 2009 GDHS sample is the result of a multi-stage stratified design, and, consequently, it was necessary to use more complex formulas. The computer software used to calculate sampling errors for the 2009 GDHS is the ISSA Sampling Error Module. This module uses the Taylor linearization method of variance estimation for survey estimates that are means or proportions. The Jackknife repeated replication method is used for variance estimation of more complex statistics such as fertility and mortality rates.
The Jackknife repeated replication method derives estimates of complex rates from each of several replications of the parent sample, and calculates standard errors for these estimates using simple formulas. Each replication considers all but one cluster in the calculation of the estimates. Pseudo-independent replications are thus created. In the 2009 GDHS, there were 325 non-empty clusters. Hence, 325 replications were created.
In addition to the standard error, the design effect (DEFT) for each estimate is also calculated. The design effect is defined as the ratio between the standard error using the given sample design and the standard error that would result if a simple random sample had been used. A DEFT value of 1.0 indicates that the sample design is as efficient as a simple random sample, while a value greater than 1.0 indicates the increase in the sampling error due to the use of a more complex and less statistically efficient design. Relative errors and confidence limits for the estimates are also computed.
Sampling errors for the 2009 GDHS are calculated for selected variables considered to be of primary interest for the women's and men's samples. The results are presented in an Appendix to the Survey Final Report for the country as a whole, for urban and rural areas, and for each of the 10 regions. For each variable, the type of statistic (mean, proportion, or rate) and the base population are given in Table B.1. Tables B.2.1 to B.2.14 present the value of the statistic (R), its standard error (SE), the number of unweighted (N) and weighted (WN) cases, the design effect (DEFT), the relative standard error (SE/R), and the 95 percent confidence limits (R±2SE), for all the selected variables, except for fertility and mortality rates. The sampling errors for fertility rates for the three-year period preceding the survey are included in Table B.3. The sampling errors for mortality rates for the five-year period preceding the survey are presented in Table B.4.1 for the total population. Table B.4.1 also includes the sampling errors for the infant mortality rate for several five-year periods preceding the survey. The sampling errors for mortality rates are presented in Table B.4.2 by residence for the ten-year period preceding the survey. The DEFT is considered undefined when the standard error is zero (when the estimate is close to 0 or 1). In the case of the total fertility rate, the number of unweighted cases is not relevant, as there is no known unweighted value for woman-years of exposure to childbearing.
The confidence interval (e.g., as calculated for children ever born to women age 40-49) can be interpreted as follows: the overall average from the national sample is 3.440, and its standard error is 0.091. Therefore, to obtain the 95 percent confidence limits, one adds and subtracts twice the standard error to the sample estimate, i.e., 3.440 ± 2×0.091. There is a high probability (95 percent) that the true average number of children ever born to all women age 40 to 49 is between 3.257 and 3.633.
Data Access
| Name | Affiliation | URL | |
|---|---|---|---|
| MEASURE DHS | ICF International | www.measuredhs.com | archive@measuredhs.com |
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.
Contacts
| Name | URL | |
|---|---|---|
| General Inquiries | info@measuredhs.com | www.measuredhs.com |
| Data and Data Related Resources | archive@measuredhs.com | www.measuredhs.com |
Metadata production
DDI_GUY_2009_DHS_v01_M_WBDG
| Name | Role |
|---|---|
| World Bank, Development Economics Data Group | Generation of DDI documentation |
2012-05-22