The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » DHS-7 Questionnaire Review - Archived » Other, including sampling and methodology » Migration
Re: Migration [message #1981 is a reply to message #1826] Fri, 11 April 2014 16:58 Go to previous message
MarkRMont is currently offline  MarkRMont
Messages: 1
Registered: March 2013
Location: Population Council, New Y...
This past year, Deborah Balk and I (with other colleagues) carried out a large demographic analysis of internal migration in poor countries, which formed the evidence base for a new report on adolescent girls titled Girls on the Move: Adolescent Girls and Migration in the Developing World (New York: Population Council, 2013). Using all DHS surveys that included the core migration questions, as well as dozens of census micro-samples from IPUMS with similar information, we found that in many countries significant percentages of urban adolescent girls are recent migrants, who have arrived in their new urban homes in the 5 years before the survey. The percentage of urban girls aged 15-19 who are migrants ranged as high as 60 percent in some DHS surveys, and in quite a number of countries we found that 1 in 5 or more of all urban girls of this age are newly-arrived migrants.

Why mention this particular piece of research? It demonstrates that for adolescent girls, one of the most important population groups that the DHS program monitors, urban migration is a very common experience. For girls as well as older migrants, a move to a new city is likely to come with a period of disruption and dislocation as the new arrival strives to adjust to unfamiliar and disorienting circumstances. Some young girls manage to adjust successfully to city life, but as we know, far too many others fail to cope and fall victim to a range of risks.

I cannot imagine that in an urbanizing world, as adolescent girls and others seek to take advantage of new urban opportunities while evading the many risks of city life, the DHS program would choose to let migrants and their circumstances simply go unrecorded. Surely that is unimaginable---and yet in the past round of the DHS surveys, that is exactly what happened: the key migration questions were dropped from the core questionnaire.

Our experience in analyzing migration from the DHS and IPUMS demonstrated the value of the simple question "how long have you lived here?" and (for movers) the follow-up question "in what type of place did you live before? Was it a large city, a small city or town, or a rural village?" Population censuses also ask the first of these questions, typically in the form "where did you live 5 years ago?", and in coding the answer most censuses collect the name of the geographic region from which the migrant came (a first-level administrative region, such as a province, and often more detail). Unlike the DHS, population censuses rarely inquire into whether the previous residence was urban or rural. Unlike population censuses, the DHS did not inquire into the *geographic location* of the previous residence. Why not, then, combine the best of both approaches?

If the DHS were to re-instate the migration questions it had asked for decades before Round 6, and supplement those time-tested questions with another well-tested (by censuses) question on the geographic location of the former place of residence, that alone would make for a small but powerfully informative package of questions that would add considerable value, helping us to better understand the urban risks and opportunities facing adolescent girls as well as other important groups of young adults.

The geography of migration---addressed by the census question on the location of previous residence---will be of increasing interest as extreme-weather events, disasters of natural or other origin, and the early manifestations of climate change begin to reshape and redirect migration streams across a range of poor countries. No doubt there will be profound alterations in rural-to-rural migration as well as rural-to-urban moves in the coming years. The DHS program should be congratulated for what it has already done to add geographical specificity to its measure of current residence (via sampling cluster coordinates), and it would be great to see the program now take steps to improve the measurement of the geographic aspects of migration for the future.

Read Message
Read Message
Read Message
Read Message
Previous Topic: Disability comments and issues
Next Topic: World Bank Group input on Household questionnaire
Goto Forum:

Current Time: Wed Oct 20 14:49:39 Coordinated Universal Time 2021