Indonesian Family Life Survey(IFLS) IFLS2+ Information

In addition to the data collected in conjunction with IFLS1 and IFLS2, we conducted a follow-up survey of a subsample of IFLS households in 1998. This survey is called IFLS2+. Please note that the IFLS2+ data is not publicly available. Questions regarding IFLS2+ data may be directed to Duncan Thomas (d.thomas@duke.edu) and Elizabeth Frankenberg (e.frankenberg@duke.edu).

The purpose of IFLS2+ was to provide insights into the effects of Indonesia's economic crisis by collecting timely data on who was affected by the crisis and on the strategies adopted to mitigate the impact. The IFLS2 was uniquely well-positioned to serve as a baseline for another interview, but we had neither the time nor the resources to mount a survey of the same magnitude as IFLS2 (which took over two years to plan, test, and conduct). Instead, we chose to field a scaled-down survey that retained as much as possible from IFLS2. The IFLS2+ was fielded almost one year after IFLS2, thus permitting us to explore the immediate impacts of the crisis. We interviewed the same respondents during each wave of the survey. This longitudinal design enables us to characterize the changes that Indonesians are initiating and experiencing during economic downturn. For more information on findings from the IFLS2+ and IFLS2 data, please see The Indonesian Crisis: Results from the IFLS2 and IFLS2+.

IFLS2+ was conducted on a 25% subsample of the IFLS communities. The IFLS2+ sample was drawn in two stages. First, to reduce costs we decided to revisit 7 of the 13 IFLS provinces: West Nusa Tengarra, Central Java, Jakarta, West Java, South Kalimantan, South Sumatra, and North Sumatra. Second, within those provinces, we purposively drew 80 Enumeration Areas (EAs) with weighted probabilities in order to match the IFLS sample as closely as possible. In the aggregate, key characteristics of the households selected for IFLS2+ match those of the full IFLS sample. Counting all original households in IFLS1 and the split-offs in IFLS2, there are 2,066 households in the IFLS2+ target sample. From a scientific point of view, it is important to retain all the original households in our target sample, even if they were not interviewed in IFLS2. This means, therefore, that our target sample includes the IFLS1 households that were not interviewed in 1997.

In 1998, we successfully contacted 60% of the households that were interviewed in IFLS1, but not in IFLS2. Restricting ourselves to the 1,934 households that were interviewed in IFLS2, we reinterviewed over 98% of the original households and in one province, West Nusa Tenggara, we re-interviewed every single household. Our completion rates at the individual level are 94% (including individuals not found in IFLS2) and 96% (of the IFLS2 respondents).

In addition to interviews with households and individuals, we repeated the community-facility survey in IFLS2+. Fieldworkers were instructed to reinterview both the community leaders, and all the facilities interviewed in IFLS2. For each community, interviewers were given a specific list of the names and addresses

of the government health centers, private providers, community health posts, and schools from which data were collected in 1997. Of the providers interviewed in 1997, a total of 219 public providers (about 2.8 per community) and 387 private providers (about 4.8 per community) were reinterviewed in 1998.