Demographic and Health Survey 2017-2018 - IPUMS Subset
Pakistan, 2017 - 2018
National Institute of Population Studies (NIPS) [Pakistan], and ICF., Minnesota Population Center
Created on
May 14, 2020
Last modified
May 14, 2020
Page views
4643
Documentation in PDF
Metadata
DDI/XML
JSON
Sample-specific cluster number (H_CLUSTERNOALL)
Data file: PAK2017-H.dat
Overview
Type:
Continuous
Decimal:
0
Width:
8
Range:
-
Format:
Numeric
Questions and instructions
Literal question
IDENTIFICATION
PROVINCE/REGION
PUNJAB 1
SINDH 2
KPK 3
BALOCHISTAN 4
GB 5
ICT 6
AJK 7
FATA 8
DISTRICT ______
TEHSIL ________
NAME OF HOUSEHOLD HEAD _______
CLUSTER NUMBER _______
HOUSEHOLD NUMBER _______
HOUSEHOLD SELECTED FOR MAN'S SURVEY?
YES 1
NO 2
HOUSEHOLD SELECTED FOR DV?
YES 1
NO 2
PROVINCE/REGION
PUNJAB 1
SINDH 2
KPK 3
BALOCHISTAN 4
GB 5
ICT 6
AJK 7
FATA 8
DISTRICT ______
TEHSIL ________
NAME OF HOUSEHOLD HEAD _______
CLUSTER NUMBER _______
HOUSEHOLD NUMBER _______
HOUSEHOLD SELECTED FOR MAN'S SURVEY?
YES 1
NO 2
HOUSEHOLD SELECTED FOR DV?
YES 1
NO 2
Description
Definition
CLUSTERNOALL (HV001) reports the cluster number for the household member's de facto residence.
According to the DHS Sampling Manual, "A cluster is the smallest geographical survey statistical unit for DHS surveys. It consists of a number of adjacent households in a geographical area. For DHS surveys, a cluster corresponds either to an [enumeration area], or a segment of a large [enumeration area]." The most recent census for a country, which divided all territory into enumeration areas to ensure full coverage of the population, often provides the sampling frame for a DHS survey.
CLUSTERNOALL is the variable indicating the primary sampling unit or PSU. CLUSTERNOALL should be used along with the variable STRATAHH (HV022) to account for the impact of sample design clustering on the estimates of variance and standard errors.
See the notes on DHS webpage "Use sample weights" [URL omitted from DDI.] on variance estimation for additional information.
CLUSTERNOALL (HV001) and HHNUMALL (HV002) can be used to merge the household member recode file to the women's individual recode file. See the DHS webpage Merging Datasets [URL omitted from DDI.] on variables to be used for linking data sets.
According to the DHS Sampling Manual, "A cluster is the smallest geographical survey statistical unit for DHS surveys. It consists of a number of adjacent households in a geographical area. For DHS surveys, a cluster corresponds either to an [enumeration area], or a segment of a large [enumeration area]." The most recent census for a country, which divided all territory into enumeration areas to ensure full coverage of the population, often provides the sampling frame for a DHS survey.
CLUSTERNOALL is the variable indicating the primary sampling unit or PSU. CLUSTERNOALL should be used along with the variable STRATAHH (HV022) to account for the impact of sample design clustering on the estimates of variance and standard errors.
See the notes on DHS webpage "Use sample weights" [URL omitted from DDI.] on variance estimation for additional information.
CLUSTERNOALL (HV001) and HHNUMALL (HV002) can be used to merge the household member recode file to the women's individual recode file. See the DHS webpage Merging Datasets [URL omitted from DDI.] on variables to be used for linking data sets.
concept
Concept
| var_concept.title | Vocabulary |
|---|---|
| Identifiers Variables -- TOPICS | IPUMS |
Imputation and derivation
Derivation
CLUSTERNOALL is an 8 digit variable.