Bricks to Blocks: Information and Coordination Challenges for Transitioning to a Cleaner Building Technology

The study evaluates the transition from traditional fired clay bricks to cleaner concrete/soil blocks in the construction sector through a clustered randomized controlled trial with three arms:

Control Group: Participants in these Upazilas receive no intervention and continue with business-as-usual operations.

Treatment 1 (Information & Facilitation): Contractors and relevant stakeholders receive an information and facilitation package delivered through in-person workshops. This includes a concise booklet summarizing the policy context (government shift towards blocks), comparative benefits and costs, a directory of local block supplier contacts, and basic workmanship guidance.

Treatment 2 (Information & Facilitation, Worker's training): Along with the Treatment 1 package, one randomly selected worker from each contractor is invited to the workshop and to a hands-on training session on the use of blocks.

2025-11-03

2026-02-22

The study has three confirmatory primary outcomes, all measures of block adoption. Each is tested once, at the contrast that matches the group receiving the intervention.

Contractors (N=792), T1 and T2 combined versus Control: (1) ever used eco-blocks (hollow concrete or compressed stabilized soil blocks instead of fired clay bricks).

Construction workers (N=1,584), T2 versus Control: (2) ever worked with eco-blocks.

Private clients (N=528), T1 and T2 combined versus Control: (3) used eco-blocks in their own construction.

A primary outcome is a confirmatory finding when its sharpened false-discovery-rate q-value and its randomization-inference p-value are both below 0.05 within the family of the three outcomes.

Adoption (outcomes 1 to 3) is a binary indicator of ever having used eco-blocks; for workers, of ever having worked on a project that used them. A baseline adopter should still report adoption at endline, so the endline outcome is coded history-consistently: the maximum of the baseline value and the endline answer given after a verification prompt to any baseline adopter who reports non-adoption. For workers, a yes given only for a training session is coded 0, because the outcome is work on a construction project. A respondent who is not re-interviewed counts as attrited. The raw endline answer is a robustness check, and the rate of baseline-yes, endline-no answers is reported by arm. Baseline means are 27.1% (contractors), 31.9% (workers) and 3.0% (clients).

Adoption is analyzed three ways. The full-sample ANCOVA of the history-consistent outcome is the confirmatory test. The same model among baseline non-adopters (about 577 contractors and 512 clients) is a featured supplementary analysis of new conversion, reported beside it. An endline-only indicator of first use since baseline is an exploratory outcome. The confirmatory finding rests on the full-sample estimate.

Secondary outcomes, reported with 95% confidence intervals and unadjusted p-values and not used for confirmatory claims:

Contractors (N=792): awareness index (0-2; baseline 0.81); knows a local block supplier (28.7%); ease of contacting suppliers (52.5%); advised clients to use blocks (38.9%); proposed blocks in a government tender (3.0%).
Construction workers: knows a nearby worker who works using blocks (N=1,562; 26.7%); block-laying skill test (endline only, all arms).
Private clients (N=528): awareness index (0.53); knows a local block supplier (11.2%).
Contractors and private clients, all arms: knowledge test, endline only (contractors 0-4, clients 0-2).
Exploratory outcomes, procurement officers (N=152): evaluated a tender that included eco-blocks (49.3%); organized training for contractors on eco-block use (9.9%).

Exploratory endline-only outcomes: first use of eco-blocks since baseline (contractors, clients); number of block projects and total projects since baseline (contractors); belief about the cost of blocks against bricks, with "don't know" as its own indicator (contractors, clients); suppliers contacted (contractors, clients); people told about blocks (contractors, clients); a government tender requiring or encouraging blocks (contractors, officers); poor-quality blocks seen in use (contractors, workers).

Compliance checks, reported without a tier: received training on blocks (workers; 2.3% among T2 workers at baseline); received training to assess block quality (officers; 21.1%); workshop attendance, from the workshop records with an endline cross-check; hands-on training take-up by the invited worker; use of the supplier directory. Contamination checks: printed block materials or detailed block information received from outside the upazila since baseline.

Reported descriptively only: satisfaction after using blocks, client project timing and walling material, the baseline barrier perception, market access and self-rated knowledge indices, and whether officers rank environmental impact among their top bid criteria.

The awareness index is the sum of two binary items: awareness that the government calls for concrete or soil blocks instead of fired clay bricks in all public construction, and receipt of any information or experience related to eco-blocks. The information item covers the past 6 months at baseline and the period since the baseline interview at endline, so that it spans the intervention (November 2025 to January 2026). Baseline means are 0.81 (contractors) and 0.53 (clients). Both items are reported beside each index and are not separate outcomes. Workshop attendees report receiving information almost by definition, so a significant index with a non-significant policy item is described as consistent with recall of treatment exposure, not as a gain in policy knowledge. The knowledge test counts correct answers to multiple-choice questions: for contractors four (0-4), on what blocks are made of, how their quality is checked, whether a hollow-block wall costs less than a brick wall, and how much cement and sand it needs; for clients the first and third (0-2). "Don't know" counts as incorrect, and the questions come before the awareness items.

Knows a local block supplier equals 1 if the respondent knows a local eco-block supplier; it is estimated separately for contractors and clients, with no test of a difference between them. Ease of contacting suppliers is asked of all contractors on a 5-point scale; the main measure equals 1 for moderate, easy or very easy, and the 5-point score is a robustness check. Advised clients equals 1 if the contractor has recommended eco-blocks to private clients. Proposed blocks in tender equals 1 if the contractor has ever proposed eco-blocks in a government tender; contractors who report no tender participation are coded 0, because participation could itself respond to treatment. Peer uses blocks equals 1 if the worker knows a nearby worker who works using eco-blocks; "don't know" answers are excluded at endline and their share is reported by arm. The skill test counts correct answers (0-3) on mortar ratio, block orientation and joint thickness; it is asked of all workers, and its headline contrast is T2 versus Control.

Endline-only measures use recall from the baseline interview. Project counts come from a project roster that also records whether each project was public or private and who specified the material. Suppliers contacted is analyzed as any and as two or more; people told as any and as three or more. The cost belief (5-point scale) is estimated on those who answer, beside the "don't know" indicator. The tender item is asked of all contractors, with those who did not bid coded 0.

This is a cluster randomized controlled trial with upazila as the unit of randomization. Upazilas are randomized into three arms (Control, T1, T2) stratified by district, with one upazila per arm within each sampled district. Outcomes are measured via surveys of contractors, procurement officers, construction workers, and private clients sampled within each upazila after the endline survey of the same respondents. Main analysis will estimate intention-to-treat effects on respondent-level outcomes, accounting for clustering at the upazila level and stratification by district.

Not available

Randomization was done in office by computer (Stata 17 MP). A random uniform number was generated to order upazilas within each district, and study arms were assigned within district using a modular rule. A reproducible random seed based on a timestamp was used. (Note: Code set seed 20251026 was used in Stata).

Randomization Unit Upazila (Cluster randomization). Randomization was stratified by District, assigning Upazilas within a district to different arms.
Additional randomization (within one arm): Within-contractor randomization of construction workers in T2, selecting exactly one of two sampled workers per contractor for treatment using a uniform random number.

Yes

66 upazilas (clusters), spread across 22 districts, with 22 upazilas per arm.

3,056 respondents to be tracked as a panel from baseline to endline. Baseline composition: 792 contractors, 152 procurement officers, 1,584 construction workers, and 528 private clients.

Control: 22 Upazilas (~1020 respondents)

Treatment 1 (Information): 22 Upazilas (~1,015 respondents)

Treatment 2 (Info + Worker Training): 22 Upazilas (~1,021 respondents)

Power calculations use the cluster design: 66 upazilas (22 per arm) in 22 district strata, alpha 0.05 (two-sided), 80% power and 5% attrition. Each outcome is adjusted for its baseline value with an assumed baseline-endline correlation (rho) of 0.10 to 0.40 set by outcome type; these are planning assumptions. Intracluster correlations (ICC) come from a random-intercept model for the upazila with district fixed effects, estimated by restricted maximum likelihood: 0.051 for contractor adoption and zero for client adoption. Worker outcomes add a contractor level (ICC 0.0265 for the upazila and 0.229 for the contractor). The design effect allows for unequal cluster sizes; the primary design effects are 1.00 to 1.78. Degrees of freedom are 42 for an arm against Control and 43 for T1 and T2 combined against Control. Minimum detectable effects (MDE) of the primary outcomes at their confirmatory contrasts: Contractors, ever used eco-blocks: 11.7 percentage points (0.26 SD) from a baseline of 27.1%. Workers, ever worked with eco-blocks (T2 versus Control): 10.8 percentage points (0.23 SD) from a baseline of 31.9%. Private clients, used eco-blocks: 4.6 percentage points (0.27 SD) from a baseline of 3.0%, or 5.7 points if the treated-arm variance is taken at the treated rate. At 10% attrition the contractor adoption MDE is 11.9 points and the worker MDE 10.9 points. If only one primary outcome has an effect, its sharpened q-value is below 0.05 only if p is at most 0.016, and the primary MDEs are then 0.27 to 0.31 SD. At the estimated upazila ICC and at twice and three times it (floors of 0.01 and 0.02 for client adoption), the primary MDEs at their confirmatory contrasts lie between 0.23 and 0.34 SD. Other MDEs: secondary outcomes 0.24 to 0.41 SD at the arm-versus-Control contrast; knowledge test (endline only) 0.26 SD (contractors) and 0.30 SD (clients) at the combined contrast; worker skill test (endline only, T2 versus Control) 0.24 SD, using the design effect of worker block usage; procurement officers (N=152) 0.48 SD for organized contractor training and 0.55 SD for evaluated tender; baseline non-adopters 0.31 SD (contractors, N=577) and 0.28 SD (clients, about 512); selected against unselected T2 worker 0.21 SD (238 intact pairs after attrition); unselected T2 workers against Control or T1 workers 0.26 SD; worker training receipt 3.7 points (0.25 SD); officer quality training 22.2 points (0.54 SD).