Post Enumeration Surveys Pres. 2

1
Post Enumeration SurveysPres. 2
2
Post Enumeration Survey

• Consists of two separate coverage studies
• A survey conducted using a sample frame
  independent of the census. Persons from this
  survey are matched to the census to estimate the
  number of persons missed in the census
• A survey conducted using a sample drawn from
  persons enumerated in the census. This sample is
  re-enumerated to determine if the sample person
  or unit was erroneously enumerated (inc.
  erroneously located)

3
Objectives of the PES

• The PES can indicate to users where specific
  coverage problems occur in the census data and
  quantify these errors
• The PES can provide guidance to census planners
  in designing future censuses, ie efforts can be
  made to improve difficult-to-enumerate sub-groups
  of population
• The PES can identify problems or erroneous
  procedures used in the census.

4
Post Enumeration Survey (contd.)

• If the problems are sufficiently severe to
  warrant an adjustment of the census, the PES can
  provide information to be used in adjustments
  models.
• The PES can provide measures levels of agreement
  on responses on selected characteristics such as
  sex, age, marital status, relationship to
  reference person or head of the household.

5
Post Enumeration Survey

• Concept of independence between the Census and
  PES
• To the extent possible the sampling frame should
  be independent from that used in the census.
  However, in practice, the census EAs are commonly
  used sampling units in a PES
• Planning and Operations It is essential that the
  planning of the census should be independent to
  that of the PES, for strategic reasons, PES
  financial resources are planned as part of the
  census programme

6
Requirement for a good PES

• The following steps are usually taken
• Setting up a separate PES planning unit from the
  census planning unit
• Use of better qualified field staff in the PES
  relative to the census
• If census enumerators are used, the better
  qualified should be involved and deployed in
  different areas for the PES
• In some countries (e.g. Australia in 2006) the
  use of an independent frame from that of the
  census is practiced
• PES questionnaire designed to have more probing
  questions

7
Requirement for a good PES

• Develop an efficient sample design
• The timing of PES should be soon after census
• Maintain operational independence
• Ensure quality assurance in all the stages of the
  PES, thus from planning to data processing
• Control of non-sampling errors is essential like
  in any other surveys, as unlike sampling errors,
  they are difficult to measure, therefore one is
  better off controlling them
• Credible estimation methodologies and procedures

8
Type of data coverage evaluation

• Non-movers persons who were in a particular
  household as of the census and PES date
• Out-movers persons who were in a particular
  household at the census date but moved out or
  were not part of the household at the time of the
  PES
• In-movers Persons who were enumerated in a
  particular household during the PES but where not
  there during the census date

9
Type of data coverage evaluation

• Erroneous inclusions persons enumerated in the
  census as usual members of a household on census
  date either due to fabrication or duplication
• Census correctly enumerated persons who were
  enumerated as usual members of a household on
  census date and PES date

10
Some items for content evaluation

• Age
• Sex
• Relationship to reference person or head of
  household
• Marital status
• Educational level
• Type of housing unit, etc.
• Countries choose preferred items to include in
  the evaluation

11

• Thank You!
• Muchas gracias !