November 1995 - Number 2
More on the IFLS Data
Below are brief descriptions of the IFLS household survey and the IFLS Community-Facility Survey. For more details, please consult the Overview and Field Report for the IFLS listed in the IFLS Documentation section of this newsletter.
IFLS Household and Respondent Sampling
The IFLS household sample design first selected areas, then households for interview, and then selected respondents within the households. The IFLS sampling scheme stratified into provinces, then randomly sampled enumeration areas (EA) within provinces using the 1993 SUSENAS sampling frame designed by the Indonesian Central Bureau of Statistics (CBS). The IFLS covered 13 of the 27 provinces in Indonesia (see the shaded areas on the map). Using the SUSENAS frame, the IFLS randomly selected 321 enumeration areas, oversampling urban EAs and EAs in smaller provinces to facilitate urban-rural and Javanese-non-Javanese comparisons. Sampling weights provided with the data adjust for this oversampling. Within an EA, households were randomly selected by field teams, based upon the 1993 SUSENAS listings obtained from regional offices of the CBS.
For each selected household, a representative member provided household-level demographic and economic information. In addition, within-household respondent selection rules were applied to select at random several household members who were asked to provide detailed individual information. Sampling weights provided in the data adjust for the respondent selection rules. The IFLS conducted detailed interviews with the following household members, if present:
- the household head and his or her spouse;
- two randomly selected children (biological, step, adopted, or fostered) of the head and spouse, aged 0 to 14 (interviewed by proxy);
- an individual aged 50 and above, randomly selected from remaining members, and his or her spouse; and
- for a randomly selected 25 percent of the households, an individual aged 15 to 49 randomly selected from remaining members, and his or her spouse.
The distribution of IFLS respondents (i.e., those administered detailed questionnaires) by characteristics is summarized in Table 1.
IFLS Household Survey Questionnaires
The IFLS household questionnaire consists of seven sections. Three sections of the questionnaire collected information at the household level: a control book (Book K) completed by the interviewer and household head or spouse of head and Books I (Household Roster) and II (Household Economy) administered to a knowledgeable household member (household head or spouse of head). The three remaining questionnaire books collected individual-level data from adult respondents age 15 and above (Book III), ever-married women respondents aged 15 to 49 (Book IV), and, by proxy, young children aged 0 to 14 (Book V). Finally, individual measures of height and weight for interviewed adults and children were recorded in a single anthropometric record for each household (Book CA). The contents of these books are presented in Table 2.
IFLS Community-Facility Survey Sampling
The community questionnaires were administered in each of the 321 enumeration areas that were drawn for the IFLS household survey. The sampling procedure for enumeration areas is discussed under the household survey.
The process by which schools and health facilities were selected for interview is one of the unique elements of the Community-Facility Survey. The sample of facilities was drawn from household respondents' answers to questions about their knowledge of schools and health care facilities. Because the list of relevant facilities is provided by members of the community, the sampling method avoids imposing externally defined, potentially arbitrary boundaries. Interviewers visited an average of 20 facilities per enumeration area for a total number of 6,385 facilities. The number of facilities (by facility type) interviewed is presented below:
IFLS Community-Facility Survey Questionnaires
The Community-Facility Survey consisted of community questionnaires administered to village leaders and facility questionnaires administered at a given facility. Three books constitute the community questionnaire in IFLS. Another set of instruments collects the facility data. These components are described briefly below.
Community questionnaires. Two books (Book I and Book II) contain questions for the village leader(s). These questions are asked in a group interview. An additional questionnaire (Book PKK) was administered to head of the Village Women's Group. Data were collected about both current and past characteristics of villages, including information about the timing of major changes in village infrastructure. All but Section H (see Table 4) were included in the questionnaires administered to the village leader and his staff. The questionnaire for the head of the Village Woman's Group consists of Sections H, I, and J.
Health facility questionnaires. Separate instruments were administered in IFLS for each of the five types of health facilities (listed in Table 3). The instruments were administered to knowledgeable staff at the facility. Table 4 lists the modules typically administered to a health facility, although questionnaire content varied by facility type. For example, the Community Health and Family Planning Posts and the traditional healers were administered only those sections relevant to services they deliver.
School questionnaires. Public, private, and religious schools were interviewed at three levels: primary, junior secondary, and senior secondary. The instruments were quite similar across school levels and consisted of the modules listed in Table 4 above. The instruments were administered to knowledgeable staff at the school.