Review Papers Claim by Claim

Key program metrics (from paper)

• Total SHGs assisted: 2352 (reported)
• Total families covered: 24,055
• SHGs covered in Table 1 (district-count, 101 clusters; 2352 SHGs) — data visualised below
• Reported average annual income from duck flocks per farmer: Rs. 15,000 (range Rs.10,000–20,000)
• Reported additional egg production: 1.39 lakh eggs/day (from 17,360 families)
• Project poultry cost: Rs. 8.025 Crores; reported annual income of 1600 SHGs: Rs. 9.60 Crores

Source: program evaluation / project report in the World's Poultry Science Journal (authors: Deka et al., 2014) — empirical data are aggregated program figures; raw household-level data and statistical analyses are not published with the article.
Chara-Chemballi Assam project report

Interpretation, strengths, and critical limitations

What the paper reliably documents

1. Large-scale program rollout: detailed counts of SHGs, districts and families served are provided and consistent across the paper (2352 SHGs; 24,055 families) — useful program-scale evidence for policy and extension practitioners (Project coverage and metrics).
2. Reported socioeconomic benefits: the paper reports average household annual income gains, examples and multiple qualitative testimonies from beneficiaries that document improved access to bank credit, school fees paid, and new micro-enterprises (Qualitative beneficiary evidence).
3. Operational model documented: supply chain (hatcheries, hatchery eggs from Kerala), credit linkages, buy-back schemes and training via SHGs are clearly described and replicable as an implementation model (Operational model description).

Major scientific and evidentiary limitations (what weakens causal claims)

• No control or comparison group — reported income increases and egg production gains are program aggregates without counterfactuals; causal inference that ducks alone caused the income gains is therefore uncertain (Lack of control groups).
• Aggregated and self-reported economic figures — mean income and ranges are given without distributional data, standard errors, time-series or household-level datasets; risk of recall bias, reporting bias or selection bias exists.
• Short follow-up and sustainability unknown — the core results are early/medium-term program outputs (2003–2006 period and reporting to 2014) but the paper does not provide longitudinal analysis of flock health, mortality, disease outbreaks, feed-cost volatility, market-price shocks or ecological impacts.
• Missing statistical analysis — no formal statistical hypothesis tests, regression adjustments for confounders, or formal cost–benefit sensitivity analyses are provided; reported project financials are program aggregates rather than net-present-value or per-household ROI over time.
• Potential biases not quantified — selection of SHGs, facilitator/NGO effects, credit access, and other concurrent interventions could confound outcomes; the paper acknowledges the need for systematic evaluation but does not quantify these biases (Authors note need for systematic evaluation).

Strengths

• Scale: the program covered thousands of households across many districts — valuable for implementation science and policy transferability.
• Operational detail: supply chains, hatcheries, bank linkages and buy-back schemes are explicitly described and can guide replication.
• Mixed evidence: both quantitative aggregates and qualitative beneficiary statements are presented, which is useful for mixed-methods program assessment.

What would strengthen the paper (specific, testable improvements)

1. Release anonymised household-level datasets (income, eggs produced, flock sizes, mortality, input costs, dates) to enable independent reanalysis and rigorous statistical evaluation.
2. Implement matched control villages or stepped-wedge/randomized rollout to estimate causal impacts and heterogeneity (by household wealth, female-headed households, market access).
3. Provide time-series of egg prices, feed costs, mortality/disease rates and net household profit (income minus input costs) with uncertainty measures (SD, CI).
4. Perform sensitivity analyses for market shocks and disease outbreaks to assess long-term sustainability and resilience.

Confidence statement and falsifiability

Given the program-style evidence presented (aggregated administrative figures + qualitative vignettes) my confidence that the program reached the scale reported (SHG counts, hatchery establishment, reported incomes) is high because counts and operational details are verifiable administrative facts; however, confidence that Chara–Chemballi ducks alone caused sustained household-level income increases is moderate-to-low without counterfactuals and household-level statistical analysis. The claims would be falsified if independent household-level surveys with comparable non-participant controls showed no difference (or worse outcomes) in income, nutritional indicators, or asset accumulation after adjusting for confounders.

Blindspots / domains not addressed

• Animal health/disease surveillance (mortality, vaccination, biosecurity) — not quantified.
• Environmental impacts (waste, water use, local pathogen dynamics) — not discussed.
• Gender dynamics beyond income — intra-household allocation and long-term empowerment measured qualitatively but not with validated indices.
• Market resilience and price-variability effects on net incomes — absent.

Author reviews