The cohort participants who are tracked in this 15-year study, hereinafter referred to as the index children (IC), and their households are the study’s main units of analysis. The survey sample was selected to be nationally representative of 10-year old Filipinos from the country’s three main island groups: Luzon, Visayas, and Mindanao (sampling domains).
Study Design and Sampling Methodology
The LCSFC aims to retain an endline sample of about 2,000 from among those recruited at baseline. Given projected attrition rates at each survey round (increasing as the cohort reaches college age), we estimate that we need to initially recruit 5,000 10-year old girls and boys, equally distributed across the three island groups or domains.
The sampling frame was based on the 2010 Census (PSA 2010)10 using data on 4-year old children expected to be age 10 in 2016. We used a two-stage sample selection scheme. Barangays (or the IC communities of residence) are the primary sampling units and were selected using probability proportional to size systematic sampling with the number of 10-year old children per barangay as the size measure.
In each sample barangay, sample children were selected using equal probability systematic sampling. Implicit stratification was used to ensure selection of urban-rural sample barangays with children considered as vulnerable [indigenous peoples (IP) and children with disabilities]. The final sampling draw yielded 345 barangays. We aimed to enroll 15 households per barangay, or a maximum of 5,175 households to provide enough margin to get at the desired sample size of 5,000 across all domains. Of the 5,175 households recruited at baseline, we interviewed a total of 4,952 households with eligible 10-year old children, giving a response rate of 95.7%.
Data collection for each full survey round consists of a household survey and a community (barangay) survey. Household surveys are done through in-person home interviews. Until the cohort participants reach age 18, the household survey respondents are their mothers (or main caregivers if the mothers are not home at time of visit). Written consent forms are read to and signed by the household respondents and verbal assents are obtained from the IC prior to the IC interviews. At age 18 and thereafter, the IC will be the main respondents (barring changes in data collection procedures), and will be consented directly. At the completion of the home visit, the respondents are given standardized tokens of appreciation for allocating time for the interview and assessments.
For the community survey, the respondents are mainly the Barangay Captain or any knowledgeable barangay official. For specific questionnaire modules (i.e., on community facilities), relevant key informants are interviewed.
Data collection for Wave 4 in early 2020 had to be terminated ahead of schedule as communities started to close their borders due to increasing COVID-19 cases. As a result, we missed 38% of the target sample for that wave. In November 2020, a brief tracking survey (Wave 4A) was done by phone to check on the status and whereabouts of the cohort. For Wave 5 (2021), we were initially given clearance by the government to resume home visits. However, right before data collection was set to begin, there was another surge in COVID-19 cases in the country and data collection was once again done by phone for the safety of both field researchers and respondents. The accompanying Community Survey (Wave 5A) for this wave was deferred for a few months (2022) until it was deemed safe to conduct in-person interviews. While collecting community data, we took the opportunity to visit the households we failed to capture in past phone surveys and update our phone contact/address masterlist.
By Wave 6 (2022), it was once again safe to conduct home visits for the household survey. No tandem Community Survey was done given that the last community survey was done just a few months before the Wave 6 household survey. Please refer to Table 1 below for more data collection and IC details.
TABLE 1. SURVEYS/DATA COLLECTION ACTIVITIES CONDUCTED:
| Study Component | Date Collection Duration | Sample households (Retention rate) | Mean IC age/Grade levels (majority of IC) |
|---|---|---|---|
| Wave 1 (home visit and community survey) | Nov 2016 to Jan 2017 | 4,952 | 10.5/Gr 4-5(98% enrolled) |
| Wave 2 (home visit and community survey) | Feb to Apr 2018 | 4,734 (95.6%) | 11.8/Gr 5-6 |
| Wave 3 (home visit and community survey) | Jan to Jun 2019 | 4,662 (94.1%) | 12.8/Gr 6-7 |
| Wave 4 (home visit and community survey) | Jan to Mar 2020 | 3,079 (62.2%) | 13.7/Gr 7-8 |
| Wave 4A (supplemental phone survey; household respondent interview) | Nov 2020 | 3,182 (64.3%) | 14.4/Gr 8-9 |
| Wave 5 (phone survey; household respondent and IC interviews) | Mar to May 2022 | ||
| Wave 5A (community survey and tracking of households lost to phone survey follow-up) | Nov 2016 to Jan 2017 | 4,952 | 10.5/Gr 4-5(98% enrolled) |
| Wave 6 (home visit) | Oct to Dec 2022 | 4,487 (90.6%) | 16.4/Gr 9-10 |
| Other data collection activities: | |||
| Baseline Qualitative Study(non-cohort sample) | Aug to Oct 2017 | 80 | 10/Gr 4-515-19 (FGDs) |
| Upcoming data collection activities: | |||
| Qualitative Study Follow-up (of Baseline In-depth interviewees) | 2023 |