The 2012-13 Pakistan Demographic and Health Survey (PDHS) is the third survey conducted so far in Pakistan under the umbrella of the global Demographic and Health Survey (DHS) program. The earlier two surveys were conducted in 1990-91 and 2006-07. The DHS surveys are designed to collect data about demographic and maternal and child health indicators with the purpose of providing reliable and updated information for policymakers and program managers.
The 2012-13 Pakistan Demographic and Health Survey was undertaken to provide current and reliable data on fertility and family planning, childhood mortality, maternal and child health, women’s and children’s nutritional status, women’s empowerment, domestic violence, and knowledge of HIV/AIDS. The survey was designed with the broad objective of providing policymakers with information to monitor and evaluate programmatic interventions based on empirical evidence.
The specific objectives of the survey are to:
• collect high-quality data on topics such as fertility levels and preferences, contraceptive use, maternal and child health, infant (and especially neonatal) mortality levels, awareness regarding HIV/AIDS, and other indicators related to the Millennium Development Goals and the country’s Poverty Reduction Strategy Paper
• investigate factors that affect maternal and neonatal morbidity and mortality (i.e., antenatal, delivery, and postnatal care)
• provide information to address the evaluation needs of health and family planning programs for evidence-based planning
• provide guidelines to program managers and policymakers that will allow them to effectively plan and implement future interventions
Kind of Data
Sample survey data [ssd]
Unit of Analysis
- Ever married women age 15-49
- Ever married men age 15-49
• Alcohol Consumption
• Birth Registration
• Domestic Violence
• Full Pregnancy History
• HIV Behavior
• HIV Knowledge
• Malaria/Bednet Questions
• Men's Survey
• Reproductive Calendar
• Service Availability
• Social Marketing
• Tobacco Use
• Tuberculosis Questions
• Vitamin A Questions
• Women's Status
Producers and sponsors
National Institute of Population Studies (NIPS)
Government of Pakistan
Ministry of National Health Services, Regulations and Coordination (NHSRC)
Government of Pakistan
Pakistan Bureau of Statistics
Government of Pakistan
Provided assistance in the selection of sampling points and household listings
United States Agency for International Development
Provided financial assistance
The primary objective of the 2012-13 PDHS is to provide reliable estimates of key fertility, family planning, maternal, and child health indicators at the national, provincial, and urban and rural levels. NIPS coordinated the design and selection of the sample with the Pakistan Bureau of Statistics. The sample for the 2012-13 PDHS represents the population of Pakistan excluding Azad Jammu and Kashmir, FATA, and restricted military and protected areas. The universe consists of all urban and rural areas of the four provinces of Pakistan and Gilgit Baltistan, defined as such in the 1998 Population Census. PBS developed the urban area frame. All urban cities and towns are divided into mutually exclusive, small areas, known as enumeration blocks, that were identifiable with maps. Each enumeration block consists of about 200 to 250 households on average, and blocks are further grouped into low-, middle-, and high-income categories. The urban area sampling frame consists of 26,543 enumeration blocks, updated through the economic census conducted in 2003. In rural areas, lists of villages/mouzas/dehs developed through the 1998 population census were used as the sample frame. In this frame, each village/mouza/deh is identifiable by its name. In Balochistan, Islamabad, and Gilgit Baltistan, urban areas were oversampled and proportions were adjusted by applying sampling weights during the analysis.
A sample size of 14,000 households was estimated to provide reasonable precision for the survey indicators. NIPS trained 43 PBS staff members to obtain fresh listings from 248 urban and 252 rural survey sample areas across the country. The household listing was carried out from August to December 2012.
The second stage of sampling involved selecting households. At each sampling point, 28 households were selected by applying a systematic sampling technique with a random start. This resulted in 14,000 households being selected (6,944 in urban areas and 7,056 in rural areas). The survey was carried out in a total of 498 areas. Two areas of Balochistan province (Punjgur and Dera Bugti) were dropped because of their deteriorating law and order situations. Overall, 24 areas (mostly in Balochistan) were replaced, mainly because of their adverse law and order situation.
Refer to Appendix B in the final report for details of sample design and implementation.
A total of 13,944 households were selected for the sample, of which 13,464 were found to be occupied at the time of the fieldwork. The shortfall is largely due to household members being absent. Of the occupied households, 12,943 were successfully interviewed, yielding a household response rate of 96 percent.
A total of 14,569 ever-married women age 15-49 were identified in the 12,943 households interviewed (an average of 1.13 women per household). Of the eligible women, 13,558 were successfully interviewed, yielding a response rate of 93 percent. The principal reason for non-response among eligible women was the failure to find individuals at home despite repeated visits to the household. Response rates were lower in urban areas than in rural areas.
A sample of 3,991 men was identified as eligible to be interviewed. Of these men, 3,134 were successfully interviewed, yielding a response rate of 79 percent. As expected, the response rate for men was lower in urban areas than in rural areas, mainly because men in urban areas are often away from their households for work. In many instances, the interviewers could not contact them even after several visits in the late evenings and, in some cases, efforts to interview them at their place of work.
Refer to Table 1.2 in the final report for more detailed summarized results of the of the 2012-13 PDHS fieldwork for both the household and individual interviews, by urban-rural residence.
Dates of Data Collection
Data Collection Mode
Field Supervision and Monitoring
Data quality was ensured through the inclusion of different levels of supervisory staff who monitored the fieldwork. In addition to the team supervisors, four quality control teams (each comprising one male member and two female members) were deployed to monitor fieldwork. Each quality control team visited field teams for two to three days and was responsible for observing interviews, reviewing the completed questionnaires to ensure that information was recorded correctly, verifying information by revisiting and reinterviewing respondents, observing height and weight measurements, and completing assignment sheets. The quality control teams were in the field from the beginning of the fieldwork to the end of the survey. Each team was provided with a separate vehicle to allow quick mobility. After each visit, the reports submitted by quality control teams to NIPS were examined, and feedback to the field teams was conveyed when necessary.
NIPS also designated three professionals from its research staff to act as field coordinators. They visited the teams assigned to them frequently to check on household selection procedures, the interviewer assignment process, questionnaire editing, team coordination, and time management. These field coordinators, usually accompanied by the quality control interviewers, observed interviews, conducted reinterviews, edited completed questionnaires, reviewed any errors with team members, and provided onthe-job training to weaker field staff.
In addition, monitoring was undertaken by NIPS senior staff, the survey advisor, and the principal investigator to check the quality of the data and other field procedures. Any deviations from set procedures by any member of the field team were pointed out and immediately rectified. Independent monitoring was also undertaken by the staff of USAID and ICF International. In view of the adverse law and order situation, particularly in Balochistan, help in field monitoring was also sought from community-based organizations and provincial population welfare departments. Data quality was monitored as well through the field check tables generated concurrently with data processing activities. Immediate feedback (by phone and through visiting and sharing with interviewers) was given. The interviewers were also cautioned
not to repeat mistakes.
Data Collection Notes
NIPS staff responsible for the survey made considerable efforts to recruit people with the requisite skills to work as field staff. Advertisements were placed in national and local newspapers across the country, and, after screening the applicants, NIPS staff visited various provincial headquarters and large cities to administer tests and interviews before selecting the final candidates. Almost all of those recruited were university graduates; three-quarters had a master’s degree. A few had been involved in work for the 2006-07 PDHS. They came from 57 districts of Pakistan, including Gilgit Baltistan. NIPS organized a three weeks long training course (during September and October 2012) in Islamabad for the 144 participants.
The training was conducted following the standard DHS procedures, which included class presentations, daily reviews, mock interviews, class exercises, and a written test at the end of the training. A few individuals who were unable to pass the test were excluded. The trainers consisted mainly of ICF International and NIPS staff. For the first time, the PDHS used the computer-assisted field editing (CAFE) system in the survey; specialized training was carried out for the participants selected to be field editors.
Toward the end of the training, three days were set aside for field practice in Islamabad and Rawalpindi. Each day after the field practice, the completed questionnaires were reviewed by senior staff, and the problems identified were discussed in the morning plenary sessions. These questionnaires were also entered in the CAFE system to allow practice among the field editors.
A total of 20 teams were organized to collect data; each consisted of a supervisor, a field editor, one male interviewer, and three female interviewers. The teams were initially deployed around Islamabad and Rawalpindi to enable intense supervision and technical backstopping at an early stage. All of the teams completed one field cluster and electronically transferred the data to the central office. Each day, a review session was organized to share the experiences of the teams. The trainers provided necessary feedback on all aspects of the fieldwork, including field management and rapport building with respondents. The fieldwork was carried out from October 2012 to March 2013, with the exception of one team in Balochistan that completed its fieldwork in the third week of April.
National Institute of Population Studies
Government of Pakistan
The 2012-13 PDHS used four types of questionnaires: Household Questionnaire, Woman’s Questionnaire, Man’s Questionnaire, and Community Questionnaire. The contents of the Household, Woman’s, and Man’s Questionnaires were based on model questionnaires developed by the MEASURE DHS program. However, the questionnaires were modified, in consultation with a broad spectrum of research institutions, government departments, and local and international organizations, to reflect issues relevant to the Pakistani population, including migration status, family planning, domestic violence, HIV/AIDS, and maternal and child health. A series of questionnaire design meetings were organized by NIPS, and discussions from these meetings were used to finalize the survey questionnaires. The questionnaires were then translated into Urdu and Sindhi and pretested, after which they were further refined. The questionnaires were presented to the Technical Advisory Committee for final approval.
The Household Questionnaire was used to list the usual members and visitors in the selected households. Basic information was collected on the characteristics of each person listed, including age, sex, marital status, education, and relationship to the head of the household. Data on current school attendance, migration status, and survivorship of parents among those under age 18 were also collected. The questionnaire also provided the opportunity to identify ever-married women and men age 15-49 who were eligible for individual interviews and children age 0-5 eligible for anthropometry measurements. The Household Questionnaire collected information on characteristics of the dwelling unit as well, such as the source of drinking water; type of toilet facilities; type of cooking fuel; materials used for the floor, roof, and walls of the house; and ownership of durable goods, agricultural land, livestock/farm animals/poultry, and mosquito nets.
The Woman’s Questionnaire was used to collect information from ever-married women age 15-49 on the following topics:
• Background characteristics (education, literacy, native tongue, marital status, etc.)
• Reproductive history
• Knowledge and use of family planning methods
• Fertility preferences
• Antenatal, delivery, and postnatal care
• Breastfeeding and infant feeding practices
• Vaccinations and childhood illnesses
• Woman’s work and husband’s background characteristics
• Infant and childhood mortality
• Women’s decision making
• Awareness about AIDS and other sexually transmitted infections
• Other health issues (e.g., knowledge of tuberculosis and hepatitis, injection safety)
• Domestic violence
Similarly, the Man’s Questionnaire, used to collect information from ever-married men age 15-49, covered the following topics:
• Background characteristics
• Knowledge and use of family planning methods
• Fertility preferences
• Employment and gender roles
• Awareness about AIDS and other sexually transmitted infections
• Other health issues
The Community Questionnaire, a brief form completed for each rural sample point, included questions about the availability of various types of health facilities and other services, particularly transportation, education, and communication facilities.
All elements of the PDHS data collection activities were pretested in June 2012. Three teams were formed for the pretest, each consisting of a supervisor, a male interviewer, and three female interviewers. One team worked in the Sukkur and Khairpur districts in the province of Sindh, another in the Peshawar and Charsadda districts in Khyber Pakhtunkhwa, and the third in the district of Rawalpindi in Punjab. Each team covered one rural and one urban non-sample area.
The processing of the 2012-13 PDHS data began simultaneously with the fieldwork. Completed questionnaires were edited and data entry was carried out immediately in the field by the field editors. The data were uploaded on the same day to enable retrieval in the central office at NIPS in Islamabad, and the Internet File Streaming System was used to transfer data from the field to the central office. The completed questionnaires were then returned periodically from the field to the NIPS office in Islamabad through a courier service, where the data were again edited and entered by data processing personnel specially trained for this task. Thus, all data were entered twice for 100 percent verification. Data were entered using the CSPro computer package. The concurrent processing of the data offered a distinct advantage because of the assurance that the data were error-free and authentic. Moreover, the double entry of data enabled easy identification of errors and inconsistencies, which were resolved via comparisons with the paper questionnaire entries. The secondary editing of the data was completed in the first week of May 2013.
As noted, the PDHS used the CAFE system in the field for the first time. This application was developed and fully tested before teams were deployed in the field. Field editors were selected after careful screening from among the participants who attended the main training exercise. Seven-day training was arranged for field editors so that each editor could enter a sample cluster’s data under the supervision of NIPS senior staff, which enabled a better understanding of the CAFE system. The system was deemed efficient in capturing data immediately in the field and providing immediate feedback to the field teams. Early transfer of data back to the central office enabled the generation of field check tables on a regular basis, an efficient tool for monitoring the fieldwork.
Estimates of Sampling Error
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 2012-13 Pakistan Demographic and Health Survey (PDHS) 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 2012-13 PDHS 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.
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 2012-13 PDHS sample is the result of a multi-stage stratified design, and, consequently, it was necessary to use more complex formulae. Sampling errors are computed in either ISSA or SAS, using programs developed by ICF International. These programs use the Taylor linearization method of variance estimation for survey estimates that are means, proportions, or ratios. The Jackknife repeated replication method is used for variance estimation of more complex statistics such as fertility and mortality rates.
The Taylor linearization method treats any percentage or average 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.
Refer to Appendix C in the final report for details of estimates of sampling errors.
The following data quality tables are produced:
- Household age distribution
- Age distribution of eligible and interviewed women
- Age distribution of eligible and interviewed men
- Completeness of reporting
- Births by calendar years
- Reporting of age at death in days
- Reporting of age at death in months
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
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.
DDI Document ID
World Bank, Development Economics Data Group
Date of Metadata Production
DDI Document version
Version 01 (January 2014). Metadata is excerpted from "Pakistan Demographic and Health Survey 2012-13" final report.