 |
Monday, Sep 8 |
|
| ||||
Study Name: Indonesia Family Life Survey
Principal Investigators: Kathleen Beegle, Elizabeth Frankenberg, Paul Gertler, Lynn Karoly, Bondan Sikoki, John Strauss, Duncan Thomas
Host Organization: RAND
Year Initiated: 1993
- Funding sources
The IFLS is supported primarily by the National Institute on Aging and the National Institute for Child Health and Human Development.Additional support has come from USAID, the Ford Foundation, and the World Health Organization, John Snow (OMNI project), the Hewlett Foundation, The Futures Group (the POLICY project) the International Food Policy Research Institute, the World Bank, and the United Nations Population Fund.
- Host organization
The study is conducted by RAND in collaboration with the Demographic Institute of the University of Indonesia, the Population Studies Center of Gadjah Mada University, UCLA, and Michigan State University.
An Advisory Board, whose composition changes over time, has provided input into the design of the IFLS.Advisors to the IFLS include researchers, scholars, and policy makers in the United States and Indonesia.
- Sample selection
Four waves of data have been collected:IFLS1 (1993),IFLS2 (1997), IFLS2+ (1998), and IFLS3 (2000).
The original sampling scheme for IFLS1 stratified on
provinces and urban/rural location, then randomly sampled within these
strata.For mainly cost-effectiveness
reasons, the IFLS is conducted in 13 of Indonesia’s 27 provinces, containing
83% of the population.Within the
provinces 321 enumeration areas (EAs) were randomly chosen from a nationally
representative sample frame used in a socioeconomic survey conducted in 1993 by
Indonesia’s Central Bureau of Statistics.Urban EAs and EAs in less-populated provinces were over-sampled.Within a selected EA, households were chosen
randomly.For IFLS1 a total of 7,730
households were sampled with the expectation of interviewing 7,000 of
them.Interviews were conducted with
7,224 households in IFLS1.Interviews
were conducted at the household and individual level.Within households in IFLS1 a
sampling scheme was used to randomly select several members within a household
to provide detailed individual informationIn all subsequent waves every household member in an origin household is
interviewed and a subset of household members are interviewed in split-off
households.
IFLS1, IFLS2, and IFLS3 interviewed households in all 321 EAs. IFLS2+ interviewed households in 90 EAs.
- Follow rules
In subsequent rounds we have sought to relocate and reinterview the 7,224 households interviewed in 1993.If members of the household have moved but arethought to be within any of the 13 IFLS provinces, the household is tracked to the new location and if possible interviewed there.In IFLS2 94.5% of IFLS1 households were relocated and reinterviewed, in the sense that at least one person from the IFLS1 household was interviewed.We also track members of IFLS1 households who have split-off and formed new households, if they meet certain eligibility criteria (for example in IFLS2 we tracked anyone born in 1968 or earlier and any other person who provided detailed individual-level information in 1993).91% of all eligible IFLS1 respondents were interviewed in IFLS2.Follow-up rates in IFLS2+ were slightly better (over 95% of households and 93% of individuals were reinterviewed).
-Oversampling
The IFLS over-samples urban households and households from provinces off the island of Java.
-Weights and attrition bias
The sample is re-weighted every wave.See Thomas, Frankenberg, and Smith, 2000, for a discussion of attrition.
-Sample “refreshing”
None.
The Indonesia Family Life Survey is designed to provide
data for studying a variety of demographic and economic behaviors and
outcomes.The survey contains a wealth
of information collected at the individual and household levels, including
multiple indicators of economic well-being (consumption, income, and assets);
education, migration, and labor market outcomes; marriage, fertility, and
contraceptive use; health status, use of health care, and health insurance;
relationships among coresident and non-coresident family members; processes
underlying household decision-making; transfers among family members and
inter-generational mobility; and participation in community activities.
Physical assessments of health have been performed in
each round were expanded from anthropometry alone in IFLS1 to include
hemoglobin status, blood pressure, lung capacity, and time to rise from a sit
to a stand (five repetitions) in IFLS2.In addition, in IFLS3 dried blood spots are being collected.
In addition to
individual- and household-level information, the IFLS provides detailed
information from the communities in which IFLS households are located and from
the facilities that serve residents of those communities.These data cover aspects of the physical and
social environment, infrastructure, employment opportunities, food prices,
access to health and educational facilities, and the quality and prices of
services available at those facilities.
The instrument used in each round
of the survey reflects the policy environment with respect to individual- and
household-level questions about program participation.The
Community-Facility
instruments are also updated to reflect policy
change.
By linking data from IFLS households to data from their
communities, the analyst can address many important questions regarding the
impact of policies on the lives of the respondents, as well as document the
effects of social, economic, and environmental change on the population.
Much of the IFLS content is comparable to data available from other RAND surveys conducted in Malaysia (MFLS), and Bangladesh (MHSS), and the LSMS surveys conducted by the World Bank.
In each follow-up wave of the IFLS we have sought to maintain comparability with the first wave wherever possible, while introducing some additional modules and scaling back other modules.For example, IFLS2 added a module on intra-household decision-making and replaced extensive breastfeeding questions with more detailed data collection on use of prenatal care and assistance at delivery.
-Mode
The IFLS is collected in face-to-face interviews using paper and pencil questionnaires.From IFLS2 onwards each interviewing team has been accompanied by a team of data editors, who enter the data on laptops and perform extensive consistency checks on the same day that the interview takes place.Inconsistencies are corrected in the field, when necessary by revisiting respondents.Electronic data are sent to the Principal Investigators to facilitate monitoring of data quality during the fieldwork.
-Instrument design (paper and pencil, CATI, EHC)
The questionnaire is paper and pencil, but extensive compter-assisted field editing (CAFÉ) involving a complex data entry program is performed.
In each follow-up round of the survey we have “preprinted” information about the household’s location, prior composition, and about the demographic characteristics of individual respondents’ non-coresident children and siblings.For example, the name, age, and sex of all household members from the previous round is preprinted.
-Calendar year, survey year and point of survey measures
Most measures reference the survey date and refer back to the previous week, month or year.
- Look Ups, Translations, and Other Answer Coding
In addition to the in-field editing, the data are subjected to an extensive set of consistency checks.Inconsistencies in the electronic files are investigated using the paper questionnaires and interviewers’ and editors’ field notes.Indonesian responses are translated into English and added to the data.Responses of “other” to questions with categorical response categories are reviewed and recoded to existing or new categories as needed.
Specialized files are occasionally produced.Information about them is provided on our website (www.rand.org/FLS/ILFS).
The data are documented in a survey results volume, a users guide, and codebooks.The questionnaires are available as well.Data are available in SAS, STATA, and ASCII formats.
-Dissemination techniques (CD, web)
The data are available from our website (www.rand.org/FLS/IFLS).
Institute for Social Research | University of Michigan | Privacy | Conditions of Use
