Login
Login
    Home / Central Data Catalog / DDI-UGA-UBOS-UDHS-2006-V01
central

UGANDA DEMOGRAPHIC AND HEALTH SURVEY 2006

Uganda
Get Microdata
Reference ID
DDI-UGA-UBOS-UDHS-2006-v01
Producer(s)
UGANDA BUREAU OF STATISTICS
Metadata
DDI/XML JSON
Created on
Aug 17, 2026
Last modified
Aug 17, 2026
Page views
176068
Downloads
132
  • Study Description
  • Data Description
  • Downloads
  • Identification
  • Version
  • Coverage
  • Producers and sponsors
  • Sampling
  • Data Collection
  • Metadata production

Identification

Survey ID Number
DDI-UGA-UBOS-UDHS-2006-v01
Title
UGANDA DEMOGRAPHIC AND HEALTH SURVEY 2006
Country
Name Country code
UGANDA UGA
Abstract
The 2006 Uganda Demographic and Health Survey (UDHS) was the fourth in Uganda's UDHS series (following 1988/89, 1995, and 2000–2001), implemented by the Uganda Bureau of Statistics (UBOS) with technical support from Macro International Inc. under MEASURE DHS (a USAID-funded project). Financial support came from a broad coalition: USAID, the U.S. President's Emergency Plan for AIDS Relief (PEPFAR), DFID, UNFPA, UNICEF, the Government of Japan, and the Health Partnership Fund. Uganda's Institute of Statistics and Applied Economics (ISAE, Makerere University) conducted an independent quality-control study, and the Department of Bio-chemistry (Makerere University) carried out vitamin A deficiency laboratory testing. Critically, the 2006 UDHS is the first round in the series to cover the entire country — all previous rounds (1988/89, 1995, 2000–2001) had excluded some districts because of insecurity.

The survey interviewed a nationally representative sample of 8,531 women aged 15–49 and 2,503 men aged 15–54, substantially larger samples than any prior round. It continued the series' core focus on fertility, family planning, infant/child/adult/maternal mortality, maternal and child health, and nutrition, while further expanding content to include disability and gender-based violence modules — new additions relative to 2000–2001. As in the 2000–2001 round, biomarker data were collected in a subsample (one in three households): women, men, and children under five were weighed and measured for height, women/men/children aged 6–59 months were tested for anaemia, and women and children were tested for vitamin A deficiency.

Key findings show the total fertility rate at 6.7 births per woman nationally (6.5 when restricted to the geographic area comparable to the 2000–2001 round, versus 6.9 recorded then) — indicating a modest fertility decline once coverage differences are controlled for. Large disparities in fertility, contraceptive use, and health outcomes persisted by residence, region, education, and wealth quintile — the report's first systematic use of a wealth-quintile measure in the UDHS series. Overall unmet need for family planning stood at 41 percent among currently married women. As with prior rounds, the survey was conducted as part of the international DHS Program, enabling both national trend analysis and cross-country comparison — with this round's full national coverage finally allowing an unqualified national-level trend comparison against later rounds.
Unit of Analysis
Individuals — the primary unit, with two eligible interview sub-populations:
Women aged 15–49, covering background characteristics, birth history and childhood mortality, family planning knowledge/use, fertility preferences, antenatal/childbirth care, breastfeeding, child vaccinations/illnesses, marriage and sexual activity, employment, husband's characteristics, HIV/AIDS and STI awareness/behaviour, maternal mortality, and domestic violence (new in this round) — 8,531 women interviewed.

Men aged 15–54, interviewed in a 1-in-3 household subsample, using a shorter instrument omitting detailed reproductive history, maternal/child health/nutrition, and maternal mortality content — 2,503 men interviewed.

Households — the Household Questionnaire enumerated all usual members and visitors, dwelling/housing characteristics (water source, toilet type, floor materials, durable goods, mosquito net ownership/use — new), survival status of parents for children under 18 (orphanhood), care/support services received by orphans and other vulnerable children, and disability status of household members — several of these are new content areas relative to 2000–2001, matching the expanded TOC we saw earlier (Fosterhood/Orphanhood, Disability chapters).
Individual biomarker measurements (refined relative to 2000–2001) — in a 1-in-3 subsample of households (a defined, named subsample this time, rather than "every other household selected for the male survey" as in 2000-01):

Height, weight, and haemoglobin: women aged 15–49, men aged 15–54, and children aged 6–59 months (note: broader age range for anaemia testing than 2000-01's "children under five" for haemoglobin)
Vitamin A deficiency (VAD) testing via blood sample: voluntary — the Household Questionnaire specifically records the respondent's decision on whether to consent to VAD blood sampling, making informed consent an explicitly documented data point in this round.

Version

Version Description
VERSION 1

Coverage

Geographic Coverage
Geographic scope: Full national coverage — the first UDHS round to cover all districts of Uganda, with no districts excluded for security or any other reason (a major discontinuity from all three prior rounds, each of which excluded some districts). At the time of the survey there were 56 districts, though this later increased to 80 (worth noting for anyone matching district codes to later administrative boundaries).

Special oversampled sub-populations within the national frame:

Karamoja area — oversampled specifically to allow Karamoja-specific reporting (5 districts: Kotido, Abim, Kaabong, Moroto, Nakapiripirit)
Internally displaced persons (IDP) camps — 30 camps sampled from a UN OCHA list (as of July 2005), located within Apac, Oyam, Gulu, Amuru, Kitgum, Lira, Amolatar, Dokolo, and Pader districts. This is a genuinely new geographic/population category not present in any prior round, reflecting the conflict-displacement situation in northern Uganda at the time.

Administrative/geographic units referenced:

Country: Uganda
Districts: 56 at time of survey (all 56 covered)
Sampling frame: 2002 Population and Housing Census (a shift from the 1991 Census frame used in 1995 and 2000-01)

Geographic coding levels available in the data/variables:

Sub-national region — a 9-region scheme (a substantial expansion from 2000-01's standard 4-region scheme), reflecting a much finer sub-national breakdown:
Central 1: Kalangala, Masaka, Mpigi, Rakai, Lyantonde, Sembabule, Wakiso
Central 2: Kayunga, Kiboga, Luwero, Nakaseke, Mubende, Mityana, Mukono, Nakasongola
Kampala: Kampala (reported as its own single-district region)
East Central: Bugiri, Busia, Iganga, Namutumba, Jinja, Kamuli, Kaliro, Mayuge
Eastern: Kaberamaido, Kapchorwa, Bukwa, Katakwi, Amuria, Kumi, Bukedea, Mbale, Bududa, Manafwa, Pallisa, Budaka, Sironko, Soroti, Tororo, Butaleja
North: Apac, Oyam, Gulu, Amuru, Kitgum, Lira, Amolatar, Dokolo, Pader, Kotido, Abim, Kaabong, Moroto, Nakapiripirit (estimates for this region combine both settled and IDP populations)
West Nile: Adjumani, Arua, Koboko, Nyadri, Nebbi, Yumbe
Western: Bundibugyo, Hoima, Kabarole, Kamwenge, Kasese, Kibaale, Kyenjojo, Masindi, Buliisa
Southwest: Bushenyi, Kabale, Kanungu, Kisoro, Mbarara, Ibanda, Isingiro, Kiruhura, Ntungamo, Rukungiri
Residence: Urban / Rural
Karamoja sub-region flag (within North region) — separately identifiable for reporting due to the oversample
IDP-camp flag (within North region) — separately identifiable for reporting due to the oversample
Universe
Individual-level universe (Women's Questionnaire):
All women aged 15–49 who were usual residents or visitors present in a sampled household, across all districts of Uganda — the first UDHS round with full national coverage, including populations in Karamoja and internally displaced persons (IDP) camps in the North region. No restriction by marital status.

Individual-level universe (Men's Questionnaire):
All men aged 15–54 who were usual residents or visitors in the subset of households selected for the male subsample (every third household in the sample) — same design as 1995 and 2000-01.

Individual-level universe (biomarker measurement — height, weight, haemoglobin):
Women aged 15–49, men aged 15–54, and children aged 6–59 months, in the households selected for these measurements (one in three households).

Individual-level universe (vitamin A deficiency testing):
A subset of the biomarker-eligible women and children who voluntarily consented to provide a blood sample for VAD testing — consent status was itself recorded, making this an explicitly opt-in sub-universe rather than the full eligible group.

Geographic restriction on the universe: None — this is the first UDHS round with no district-level exclusion. The universe covers the entire national population within the eligible age/sex/residency criteria, including the special oversampled Karamoja area and IDP camp populations in the North.

Household-level universe (Household Questionnaire):
All households within the 368 sampled clusters (321 shared with the 2005–2006 UNHS, 17 additional in Karamoja, 30 IDP camp segments), including their usual members and visitors of any age — used to enumerate household composition, dwelling characteristics, mosquito net ownership, orphanhood/parental survival status (for children under 18), care and support for orphans and vulnerable children, and disability status of household members.

New/expanded sub-populations of interest within the household universe (relative to prior rounds):

Children under 18 — parental survival status (orphanhood determination)
Orphans and other vulnerable children — care and support services received
Household members with disabilities

Explicitly excluded from the universe:

Institutional populations (consistent with standard DHS household-based sampling practice across all rounds)
Men outside the 1-in-3 household subsample (for men's interview and men's biomarker measurement)
Women/children who declined consent for VAD blood sampling (excluded from that specific biomarker sub-universe, though still part of the broader interview universe)

Producers and sponsors

Primary investigators
Name Affiliation
UGANDA BUREAU OF STATISTICS MOFPED
Producers
Name Affiliation Role
UGANDA BUREAU OF STATISTICS MOFPED DATA COLLECTION AND ANALYSIS
Funding Agency/Sponsor
Name Abbreviation Role
UGANDA BUREAU OF STATISTICS UBOS FUNDING

Sampling

Sampling Procedure
Sample type: Stratified, multi-stage sample, matched to the concurrent 2005–2006 Uganda National Household Survey (UNHS) cluster sample — a design not seen in any prior UDHS round.

Sampling frame: The 2002 Population and Housing Census (a shift from the 1991 Census frame used in the 1995 and 2000-01 rounds).

Strata / domains:

National level, and urban/rural separately
9 sub-national regions (Central 1, Central 2, Kampala, East Central, Eastern, North, West Nile, Western, Southwest — full district composition given in the geographic-coverage answer above)
Special oversampled areas: Karamoja and IDP camps (within North region)

Two-stage design, with three distinct cluster sources:

Stage 1 (cluster/PSU selection):
321 clusters selected by reusing the cluster sample already drawn for the 2005–2006 UNHS (itself sourced from the 2002 Census frame) — done deliberately to allow linking 2006 UDHS health indicators to UNHS poverty/welfare data.
17 additional clusters selected fresh from the 2002 Census frame specifically in Karamoja, to boost that sub-region's sample size for independent reporting.
30 additional clusters selected as segments of IDP camps, drawn from a UN OCHA list of camps as of July 2005.
Total: 368 primary sampling units (clusters).
Stage 2 (household selection):
In the 321 UNHS-linked clusters: household lists came from the UNHS's own listing operation (April–August 2005). The 10 households already sampled by UNHS in each cluster were automatically/purposively included in the UDHS sample, plus an additional 15–20 households randomly selected per cluster.
In the 17 additional Karamoja clusters: a fresh listing was conducted, and 27 households selected per cluster.
In the 30 IDP camp segments: each camp was divided into segments (due to large size), one segment selected per camp, a listing operation conducted within it, and 30 households selected per camp from the listed segment.

Achieved/target sample: A representative probability sample of 9,864 households was selected — this is the sample-size figure given directly in §1.5.1; response-rate and completed-interview breakdowns are detailed further in Appendix A (Sample Implementation) and Chapter 1.6 (Response Rates) if you want those pulled next.

Sampling error estimation: As with the 2000-01 round, the ISSA Sampling Error Module was used, applying:

Taylor linearization for variance estimation of means/proportions — the ratio-estimator formula (r = y/x) is given explicitly in Appendix B, now incorporating cluster-level weighted sums (yhi, xhi) rather than the simpler unweighted form shown in the 1988/89 report.
Jackknife repeated replication for complex statistics (fertility and mortality rates) — with 368 replications created (one per non-empty cluster, each replication dropping one cluster), consistent with the 368 total clusters in the design.
Weighting
Why weighting is required: As with all prior rounds, the 2006 UDHS used a multi-stage stratified sample design with deliberate oversampling — urban areas, Karamoja, and IDP camps were all sampled at rates disproportionate to their population share (Karamoja specifically boosted via 17 extra clusters; IDP camps sampled as an entirely separate stratum; UNHS-linked clusters received a mix of "automatically included" and "additionally randomly selected" households, which itself creates unequal selection probabilities within cluster). This structure necessarily requires sample weights to produce nationally representative estimates — the same underlying logic as 1995 and 2000-01, though the 2006 report never states this explicitly in the plain language used in those two prior reports (e.g., no direct sentence equivalent to 2000-01's "the sample is not self-weighting."

Data Collection

Data Collection Mode
Primary mode: Face-to-face structured interviews using paper questionnaires, supplemented by anthropometric measurement and clinical biomarker collection (haemoglobin, vitamin A) — same core mode as 2000–2001, with a more elaborate community-preparation phase.

Instruments used (three questionnaires — same structure as 2000-01, no community-level instrument):

Household Questionnaire — listed usual members/visitors; collected age, sex, education, relationship to household head; dwelling characteristics (water source, toilet type, floor materials, durable goods, mosquito net ownership/use); parental survival status for children under 18; care/support for orphans and vulnerable children; disability status; documented consent decisions for VAD blood sampling; and recorded height, weight, and haemoglobin for women 15–49, men 15–54, and children 6–59 months in the biomarker subsample.
Women's Questionnaire — administered to all women 15–49, covering background, birth history/childhood mortality, family planning, fertility preferences, antenatal/childbirth care, breastfeeding, child vaccinations/illnesses, marriage/sexual activity, employment, husband's characteristics, HIV/AIDS and STI knowledge/behaviour, maternal mortality, and domestic violence (new module).
Men's Questionnaire — administered to men 15–54 in the 1-in-3 household subsample; shorter, omitting detailed reproductive history, maternal/child health, nutrition, and maternal mortality content.

Instrument development: Based on MEASURE DHS model questionnaires, adapted by UBOS in consultation with technical institutions and local organizations to reflect Uganda-specific population and health issues. Full questionnaires reproduced in Appendix F.

Language/translation: Translated from English into six local languages, pretested prior to finalization in January–February 2006 (exact language list not yet confirmed from the excerpt reviewed — let me know if you want me to pull the specific six languages).

Biomarker/clinical testing procedures:

Haemoglobin testing: field-based, via HemoCue-style equipment (consistent with 2000-01 methodology), for women, men, and children 6–59 months in the 1-in-3 biomarker subsample.
Vitamin A deficiency (VAD) testing: via dried blood spot (DBS) samples collected from a finger prick, voluntary and consent-documented. Samples were dried overnight in light-proof boxes, then stored in portable vehicle-battery-powered refrigerators in the field to prevent degradation of retinol binding protein (RBP), before transport to the Biochemistry Department, Makerere University, for storage at -20°C pending laboratory testing — explicitly stated to be unrelated to HIV/AIDS testing (a point deliberately communicated to respondents during mobilization, to avoid confusion/refusal).
Anthropometry: height and weight for women, men, and children 6–59 months.
Iodine testing: salt samples tested for iodine content using UNICEF-provided test kits (new relative to prior rounds).
Mosquito net identification

Metadata production

DDI Document ID
DDI-UGA-UBOS-UDHS-2006-v01
Back to Catalog
Uganda Bureau of Statistics (UBOS)

© Uganda Bureau of Statistics (UBOS), All Rights Reserved.