Transcript Household survey data on remittances in sending

Household Survey Data on
Remittances in Sending Countries
Sampling and Questionnaire Design: Options and Uses
Johan A. Mistiaen
World Bank - Development Data Group
International Technical meeting on Measuring Remittances
Washington DC - January 24-25, 2005
Overview
Why Collect Micro-Data from Remittance Senders?
Sampling Frame Design Options




Why is Sampling a Critical Issue?
Plan A: Build a Representative Sampling Frame
Plan B: Some Micro-Data is Better Than None
On Sample Size
Questionnaire Design and Implementation


A Core Module: Towards Data Consistency
Implementation Challenges
Ideas for a Research Agenda
Sampling Design Options
Why is Sampling a Key Issue?
A representative sampling frame is the cornerstone of sample-based
statistical analysis:
 Without it we cannot obtain sample-based statistics or inferences
that are representative of the population of interest.
 For instance, representative sample data is needed to compute
“propensity to remit” estimates.
Sampling frames of the population sub-groups that send remittances
are non-existing  need to build them
Need to define our target population (domain of analysis)
 All persons above 18 years of age that were born in a foreign
country.
 Unlikely standard frames can be used…
Sampling Design Options
Plan A: Build a Representative Sampling Frame
Option I: Finding All Needles in the Haystack
Current Population Registers
Data systems that record selected info on the de jure population in
a country; including data that identify residents by street address,
age and country of birth.
Construct address referenced listings of all members in
the respective target sub-population groups by
geographical areas (asap) which become the “clusters”
of our sampling frame.
Sampling Design Options
Plan A: Build a Representative Sampling Frame
Option I: Finding All Needles in the Haystack
Can apply standard techniques to select a
representative (stratified) sample of each sub-group
(i.e. by country of birth) with associated sampling
weights (the inverse selection probability).
Work ongoing to implement this approach in some EU
member states.
Already in design phase to draw samples of Africanborn residents in Belgium.
Advantages:
 Representative sample
 Relatively easy to maintain sampling frame
Sampling Design Options
Plan A: Build a Representative Sampling Frame
Option II: Finding the key Haystacks
Population Census Data

Typically collect data on “country of birth”
(sometimes also include street addresses)

Identify all geographical areas (as small as possible)
from the census that contain target sub-population
group members; these become “clusters” in our
sample frame.

Examples: UK 2001 Population Census
US 2000 Population Census
From Population Census data
it is possible to build a “frame”
of Enumeration Areas/Blocs
(100?-150? hhs) in the UK that
contain people born in specific
foreign countries
Data on “country of birth”
was also collected via the
“long form” of the 2000 US
census (1 out of 6 hhs)
Sampling Design Options
Plan A: Build a Representative Sampling Frame
Option II: Finding the key Haystacks
A Two-Step Sampling Approach

Step 1: Draw sample of clusters (can adjust
probability of selection on the proportion of target
sub-population).

Step 2: Conduct a “screening” or “re-listing” exercise
to identify current incidence of the target population.

Draw sample based on screened clusters

If needed, adjust initial cluster sample ex-post (if
step 2 conducted “on-the-go”) either via reweighting methods or with supplementary sampling.
Sampling Design Options
Plan A: Build a Representative Sampling Frame
Options I and II: Limitations and Caveats
Frame Errors: All Needles?…“illegal” immigrants…
 Population registers vs. population census data
 Pilot attempts to supplement main sampling frame by
“snowball” sampling (i.e. referrals), through relevant
organizations, and at key likely contact points
(Groenewold and Bilsborrow, 2004).
Population register approach potentially feasible in most
EU member states; but few useable population registers
elsewhere (Bilsborrow et al., 1997).
Sampling Design Options
Plan A: Build a Representative Sampling Frame
Options I and II: Limitations and Caveats
“sensitive data”: Government cooperation critical
“updating” of population census based frames…
without screening all relevant clusters
 will need to account for modeling errors.
Sampling Design Options
Plan B: Some Micro-Data is Better Than None
Aggregation Point Sampling
Listing of migrant (foreign-born) meeting points

Religious venues, community centers, international phone
businesses, employment offices, etc…
Will capture both legal and undocumented immigrants
Ex-post determination of respondent selection
probabilities

Based on “visit frequency” profiles (e.g., what aggregation
points in the sample are visited, how often, when, etc…)
Can yield a (representative) sample
Applied successfully to interview Ghanaian and Egyptian
born persons in Italy (Groenewold and Bilsborrow,
2004).
Sampling Design Options
On Sample Size
Osili (2004): Sampled 112 Nigerian born residents in the Chicago
area to study remittances
Average annual per capita remittances: $6,000
Standard deviation: $11,250
 95% confidence interval = [$3,750 ; $8,250]
Average annual per capita income: $25,500
 Mean Propensity to Remit = 0.23
 95% confidence interval = [0.15 ; 0.32]
Increasing sample size to 400 would halve the standard error
Optimal sample size will be a function of the distribution
of the variable of interest and the targeted precision
Questionnaire Design and Implementation
A Core Module: Towards Data Consistency



Core data collection
Consistent across countries and within countries
Modular: stand alone or tag-on to other survey
Implementation Challenges

Minimizing Non-Response


Questionnaire design, interviewer selection/training,
collaboration with community groups, etc.
Understanding/Correcting for Non-Response
Ideas for a Research Agenda
Statistical and econometric analysis to obtain better
measures of the “propensity to remit” and its
determinants; both household characteristics and
market variables (e.g., transaction costs…).
Small area estimation of the “propensity to remit” by
combining survey and census data.