Experiments on survey measurement error and high frequency data collection in Bangladesh

This study embeds a set of randomized experiments within a multi-round household panel survey in flood-prone districts of Bangladesh. The experiments are designed to identify the sources and mechanisms of measurement error in household surveys.

2026-04-25

2026-10-31

Data quality indicators:
Data accuracy (match rates between enumerator-observed and respondent-reported dwelling characteristics and observation items); internal consistency (FIES attention check, PHQ-8 attention check, locus of control consistency, climate adaptation question pair consistency, income aggregate vs. itemized discrepancy); enumerator-assessed interview quality (1–10 ladder); paradata-based measures (interview duration, response heaping, straightlining, acquiescence bias, RISSK Unit Risk Score).

Substantive welfare outcomes:
PHQ-8 score and share with any depressive symptoms (score ≥ 4); FIES raw score; total household labor income (aggregated from sources); child school attendance and absenteeism; any work in last week and hours worked; road conditions (water/mud visible); flood likelihood perception.

PHQ-8 score: Sum of eight items (Q2–Q9), each scored 0–3 (range 0–24). Excludes planted attention-check item (Q10). Share with any depressive symptoms defined as score ≥ 4.
FIES raw score: Count of “Yes” responses across eight standard items (Q2–Q9, range 0–8). Excludes planted attention-check item (Q10).
Income: Sum of itemized labor income sources (wages/salary, business profits, farm profits, other labor income) as reported by respondent.
Data accuracy index: Inverse-covariance-weighted index (Anderson 2008) combining match rates across dwelling characteristic and observation item benchmarks.
Consistency check index: Inverse-covariance-weighted index combining binary indicators for correct responses to FIES, PHQ-8, locus of control, and climate adaptation consistency checks.

Depressive symptom severity categories (mild, moderate, moderately severe, severe); binary food insecurity indicators; income by source; days of school absence; work by type; job search; WASH indicators; climate adaptation perceptions (inverse framing); within-household outcome agreement; enumerator feedback index.

Severity categories for PHQ-8: mild (5–9), moderate (10–14), moderately severe (15–19), severe (20–24). Within-household agreement computed as the degree of concordance between the two respondents’ reports on household-level outcomes.

A multi-round panel survey experiment with approximately 1,500 households in flood-prone districts of Bangladesh, surveyed four times between April and October 2026. Two respondents per household are interviewed with identical instruments. The study embeds several randomized experiments within the survey to identify sources of measurement error. The main randomizations vary survey mode (phone vs. in-person), respondent incentive levels, enumerator assignment, and the ordering of modules within the questionnaire. A digital monitoring component operates between survey rounds.

Not available

Computer-generated randomization using Stata, stratified as described below.

Multiple levels: - EA level: Incentive amounts, Resonance group assignment. - Household level (within EA): Survey mode (stratified by EA), recording visibility (household level for in-person interviews). - Respondent level (within household): Module ordering.

Yes

50 enumeration areas (EAs).

Approximately 1,500 households with 2 respondents each (approximately 3,000 respondents).

Survey mode: ~750 households phone, ~750 households in-person (15 per arm per EA × 50 EAs).
Incentives: ~17 EAs low, ~17 EAs medium, ~16 EAs high (~510/510/480 households).
Module order: ~500 respondents per version (6 versions, respondent-level).
Recording visibility: Round 0: ~2,700 respondents with visible external microphone, ~300 respondents tablet-only;
High frequency monitoring system: ~20 EAs information + monitoring (~600 HH), ~20 EAs monitoring only (~600 HH), ~10 EAs pure control (~300 HH).