Constarium
← Search

Data · dataset · 2008

National election study 2006 -CSES Module 3

Listed in Harvard Dataverse

The sample is selected by a multistage procedure.

Description

The universe was divided in four regions: states with PAN Governor, Northern States with a PRI governor, Center-South states with PRI governors, and states with a PRD governor. An independent sample was drawn within each region.

In each region precincts were ordered by the vote for PRI in the last election, and divided in four groups of the same number of precincts. Within each group, precincts are clustered by county. So you have groups with similar vote for PRI and the same county.

Read the rest (2 more)

The first selection stage is done with this list, clustering precincts within each group with probability proportional to size (PPS), being turnout the size of the cluster. In the second stage precincts are selected with PPS. Fieldwork teams receive the sample of precincts.

In the field, the third stage are blocks randomly selected in the precinct area. In each block, houses are selected following systematic methods of random start. In the households, respondents are selected according to quotas of sex and age.

Links

Where it is published

Catalogue records · 1
Provenance · 1 source records, 6 field assertions
SourceKeyLast seenRaw
Harvard Dataversedoi:10.7910/DVN/Z8KO1010 d agoJSON v1
FieldAssertionExtractorEvidence
created_datesource · Harvard Dataverseconnector:dataverse@1.0.0
descriptionsource · Harvard Dataverseconnector:dataverse@1.0.0/description
publication_datesource · Harvard Dataverseconnector:dataverse@1.0.0
titlesource · Harvard Dataverseconnector:dataverse@1.0.0/name
updated_datesource · Harvard Dataverseconnector:dataverse@1.0.0
version_labelsource · Harvard Dataverseconnector:dataverse@1.0.0