A Multi-faceted Program Causes Lasting Progress for the Very Poor: Evidence from Six Countries
Science, 2015 · paper · Economics
Six randomized trials in Ethiopia, Ghana, Honduras, India, Pakistan and Peru test a multifaceted program for the extreme poor, including an asset grant, training and support. Gains in consumption and income persisted a year after it ended, and benefits exceeded costs in five of six sites.
Checked: PDF from economics.mit.edu, fetched 5 Oct 2026. Page numbers are those of that PDF.
13 findings
Substance "In all sites but Ethiopia ... between 2,402 and 5,142 calories per day", but 2,402 is Ethiopia's value in Table 1. The non-Ethiopia minimum is 3,149 (Pakistan).
- p. 3 Consumption support
In all sites but Ethiopia (where the consumption sup- port was part of an existing program), the trans- fer corresponds roughly to the monetary equivalent of between 2,402 and 5,142 calories per day
- p. 6 Table 1 continued, calories row
2,402
- p. 3 Consumption support
In all sites but Ethiopia ... of between 2,402 and 5,142 calories per day
- p. 6 Table 1, consumption support calories
2,402 5,142 3,854 3,928 3,149 4,993
Judge's note: Table 1 (continued) on p.6 gives the calorie row as Ethiopia 2,402, Ghana 5,142, Honduras 3,854, India 3,928, Pakistan 3,149, Peru 4,993. With Ethiopia excluded, as the sentence on p.3 says it is, the range is 3,149–5,142. The printed lower bound is the excluded site's value.
Second check: another model agreed
Report a problem with this finding
Substance Text says consumption support "lasted between 4 and 13 months, depending on the site", but note 11 says Ghana's "total duration of support received was 14 months". Note 11 also describes two 6-month lean seasons.
- p. 3 Consumption support
4 and 13 months, depending on the site
- p. 16 Note 11
the total duration of support received was 14 months
- p. 3 Consumption support
lasted between 4 and 13 months, depending on the site
- p. 3 Consumption support
lasted between 4 and 13 months
Only the authors can say what is right here.
Judge's note: Read p.3 and note 11 (p.16). Ghana's 14 months lies outside the stated 4–13 month range, and the text does not exclude Ghana. Note 11's own description (the 6-month lean season in each of two years) gives 12 months, not 14. Only the authors can say which figure is right.
Second check: another model agreed
Report a problem with this finding
Substance In Table 4 Panel A, Pakistan 'Direct transfer costs' is printed as 2048, but its components 'Asset cost' 1043 and 'Food stipend' 911 sum to 1954; in every other column the components sum exactly to the total.
- p. 12 Table 4 Panel A, Pakistan
2048
- p. 12 Table 4 Panel A, direct transfer costs / asset cost / food stipend, Pakistan column
2048 ... 1043 ... 911
Only the authors can say what is right here.
Judge's note: Checked the Table 4 image on p.12. Every other site satisfies the identity (Ethiopia 1228+0, Ghana 451+229, Honduras 537+187, India 437+263, Peru 854+241). Pakistan's supervision costs are shown as dashes, so the total direct cost of 4680 cannot show which cell is wrong. Note 1 (five partners, costs averaged) does not explain a gap of 94. Only the authors can say which entry is wrong.
Second check: another model agreed · checked by hand
Report a problem with this finding
Substance The text says "the 10th percentile of consumption (income and revenue index) increases by 0.027 SD (0.005 SD), whereas the 90th percentile increases by 0.491 SD (0.079 SD)". In Table 5 at endline 1, consumption is 0.065 (10th) and 0.140 (90th). The values 0.027 and 0.491 are the income-and-revenues index estimates, and 0.005 and 0.079 are its standard errors.
- p. 13 Effects on distribution of outcomes, first column
the 10th percentile of con- sumption (income and revenue index) increases by 0.027 SD (0.005 SD), whereas the 90th per- centile increases by 0.491 SD (0.079 SD)
- p. 14 Table 5, endline 1 rows
0.027*** 0.053*** 0.098*** 0.314*** 0.491***
- p. 14 Table 5, endline 1
0.065*** 0.067*** 0.045** 0.099*** 0.140**
- p. 14 Table 5, endline 1, columns 1 and 5
0.065*** ... 0.140** (consumption); 0.027*** ... 0.491*** (income)
Judge's note: Checked p.13 text and the Table 5 image on p.14. Endline 1 consumption row reads 0.065***, 0.067***, 0.045**, 0.099***, 0.140**. Income row reads 0.027*** (0.005) ... 0.491*** (0.079). The quoted numbers are the income estimate followed by the income SE, not consumption with income in parentheses. Apparent intended text: 0.065 SD (0.027 SD) and 0.140 SD (0.491 SD). Also, the consumption row does not increase at every quantile, so 'all increasing with the quantiles' does not hold for consumption.
Second check: another model agreed
Report a problem with this finding
Reference or label The Summary figure caption reads "This figure summarizes the average treatment effects in each country for the 10 primary outcomes. All treatment effects are presented as standardized z-score indices", but the plot it describes is titled 'Pooled average intent-to-treat effects, endline 2' and shows pooled variable-level estimates.
- p. 1 Summary figure caption
This figure summarizes the average treatment effects in each country for the 10 primary outcomes. All treatment effects are presented as standardized z-score indices
Judge's note: On page 1 the plot has no country labels. Its rows are pooled individual variables such as Per Capita Food Consumption, Livestock Revenue and Total Savings, grouped by family, matching Fig. 2. The caption wording matches the country-index format of Fig. 3, so the caption is mislabelled.
Second check: another model agreed
Report a problem with this finding
Reference or label Table 1, Pakistan eligibility: "income less than INR 25 (PPP US$1.74) per day". The currency should be PKR, not INR.
- p. 4 Table 1, Pakistan eligibility
income less than INR 25 (PPP US$1.74) per day
Judge's note: The page image (p.4) puts this entry in the Pakistan column. Using note 14's rates, PKR 25 at the 2008 rate (15.84) with the CPI adjustment gives 1.74, matching the printed figure. INR 25 at India's 2007 rate (11.76) would give 2.43. So the conversion is a PKR conversion and only the currency label is wrong.
Second check: another model agreed
Report a problem with this finding
Reference or label "using the Benjamini-Hochberg step-up method (21)": ref. 21 is Morduch, Ravi, Bauchet, "Failure vs. Displacement: Why an Innovative Anti-Poverty Program Showed No Net Impact".
- p. 9 Analysis methods
using the Benjamini-Hochberg step-up method (21)
- p. 16 Ref 21
Failure vs. Displacement: Why an Innovative Anti-Poverty Program Showed No Net Impact
- p. 16 Reference 21
J. Morduch, S. Ravi, J. Bauchet, “Failure vs. Displacement
- p. 9 Analysis methods
Benjamini-Hochberg step-up method (21)
Only the authors can say what is right here.
Judge's note: Reference list on p.16: entry 21 is the SKS working paper, which note 10 cites appropriately for that study. Entry 22 is Anderson (2008), cited separately in the next sentence. No Benjamini–Hochberg entry appears in the list, so the intended source is unclear and only the authors can supply it.
Second check: another model agreed
Report a problem with this finding
Reference or label The Fig. 3 caption ("treatment effects presented in tables S3a to S3f") and note 24 ("indexed family outcomes in tables S3a to S3f") point to S3a–S3f for the country-level index results. The text says "Tables S4a through S4f present the impacts on the 10 indexed family outcomes, one table per country" and uses table S3 for the attrition bounds.
- p. 13 Fig. 3 caption
treatment effects presented in tables S3a to S3f
- p. 10 Country-by-country variation
Tables S4a through S4f present the impacts on the 10 in- dexed family outcomes, one table per country
- p. 10 Pooled results
In table S3, we present bounds
- p. 10 Country-by-country variation
Tables S4a through S4f present the impacts on the 10 in- dexed family outcomes, one table per country.
Judge's note: Checked p.10 (text: table S3 holds the bounds; S4a–S4f hold the country index tables; also 'tables S4a to S4f'), the p.13 Fig. 3 caption and note 24 on p.16. The same content is cited under two table numbers. The supplement is not in the files.
Second check: another model agreed
Report a problem with this finding
Reference or label The article cites "supplementary text 5" for the cost-benefit calculations (Table 4 note 4; p.14 text), but the Supplementary Materials list gives "Supplementary Text 1 to 4".
- p. 14 Cost-benefit text
See supplementary text 5 for details
- p. 17 Supplementary Materials list
Supplementary Text 1 to 4
- p. 14 Cost-benefit discussion
See supplementary text 5 for details on the cost-benefit analysis calculations.
- p. 12 Table 4 note 4
See supplementary text 5
Only the authors can say what is right here.
Judge's note: Checked the Table 4 notes on p.12, the p.14 text and the p.17 Supplementary Materials list, which reads 'Supplementary Text 1 to 4; Figs. S1 to S3; Tables S1 to S8'. The supplement is not in the files, so it is not possible to tell whether the list or the pointer is wrong.
Second check: another model agreed
Report a problem with this finding
Reference or label "Finally, in Bangladesh (3), households were followed for 2 more years" cites ref. 3 (Ashley, Holden, Bazeley, Livestock in Poverty-Focused Development, 1999). The Bangladesh BRAC RCT is ref. 6 (Bandiera et al.).
- p. 16 Discussion, persistence
Finally, in Bangladesh (3), households were followed for 2 more years
- p. 16 Ref 3
Livestock in Poverty-Focused Development
- p. 16 Reference 3
S. D. Ashley, S. J. Holden, P. B. S. Bazeley
- p. 16 Discussion, persistence
in Bangladesh (3), households were followed for 2 more years
Judge's note: Checked p.16: reference 3 is the 1999 livestock report and reference 6 is Bandiera et al., 'Can entrepreneurship programs transform the economic lives of the poor?'. Ref. 6 is cited for the BRAC Bangladesh evaluation on pp.11 and 14. A 1999 report cannot describe a follow-up of that RCT.
Second check: another model agreed
Report a problem with this finding
Reference or label Note 24 refers to country-specific indexed family outcomes "in tables S3a to S3f". The body text places these in tables S4a–S4f and calls table S3 the attrition bounds.
- p. 16 Note 24
indexed family outcomes in tables S3a to S3f
Only the authors can say what is right here.
Judge's note: Page 10 says "In table S3, we present bounds" and "Tables S4a through S4f present the impacts on the 10 indexed family outcomes". The Fig. 3 caption also cites S3a–S3f. The supplement is not provided, so the authors must say which label is right.
Second check: another model agreed
Report a problem with this finding
Wording "All but one site (Peru) transferred productive assets worth between four to eight goats", but Table 1 gives Pakistan "3.75 goats", which is below four.
- p. 3 Productive-asset transfer
All but one site (Peru) transferred productive assets worth between four to eight goats
- p. 5 Table 1, value in goats
3.75 goats
- p. 3 Productive-asset transfer
transferred productive assets worth between four to eight goats
Judge's note: Table 1 on p.5 lists 7.98 (Ethiopia), 6.00, 4.75, 6.53, 3.75 (Pakistan) and 17.14 (Peru) goats. The p.11 text also gives Pakistan as 3.75 goat equivalents. So two sites fall outside 4–8, not one. The fix would be to name Pakistan or give the range as 3.75–8.
Second check: another model agreed
Report a problem with this finding
Wording "positive and significant impacts on income, consumption, and assets, at all tested quantiles", but Table 5 gives the endline 2 10th-percentile income and revenues effect as 0.001 (0.006) with no stars.
- p. 11 Effects on distribution of outcomes
we see positive and significant impacts on income, consumption, and assets, at all tested quantiles
- p. 14 Table 5, Incomes and revenues, column 6
0.001
- p. 15 Discussion, third explanation
though positive every-where, are lower at the bottom
- p. 14 Table 5, income row, endline 2
0.001 0.029*** 0.171***
Judge's note: In Table 5 (p.14), column 6 of the income and revenues row is 0.001 (0.006) with no significance marker. Every other income, consumption and asset cell is starred. The p.11 sentence carries no endline restriction, so the claim overstates the table for this one cell. Restricting it to endline 1, or excepting this cell, would fix it.
Second check: another model agreed
Report a problem with this finding
29 minor (typos and the like; most checked by the judge only)
Wording The introduction describes "regular interactions with the households over the course of a year", whereas the program section says visits ran "throughout the 2-year program" and Table 1 lists 18–24 months.
- p. 2 Introduction
plus regular interactions with the house-holds over the course of a year
- p. 3 High-frequency home visits
throughout the 2-year program
Judge's note: The sentence may describe the generic BRAC model, but as written it describes the program studied here. The duration it gives conflicts with page 3 and with Table 1.
Second check: another model agreed
Report a problem with this finding
Wording Peru conversions "PEN 100 (PPP US$71.96)" and "PEN 200 (PPP US$143.33)" are not consistent with "PEN 1200 (PPP US$853.97)" or with each other.
- p. 5 Table 1, Peru, value of asset transfer and consumption support
PEN 1200 (PPP US$853.97) ... PEN 100 (PPP US$71.96)
- p. 3 Consumption support text
PEN 200 (PPP US$143.33)
Only the authors can say what is right here.
Judge's note: Note 14 gives the PEN rate for 2011 as 1.48 and the CPI for 2011 as 224.939. These reproduce PEN 1200 = 853.97 exactly, but give 71.16 for PEN 100 and 142.33 for PEN 200. The 2013 and 2014 rates match neither printed value. 143.33 is also not twice 71.96. Each printed figure differs from the 2011 conversion in a single digit, which suggests transcription slips, but only the authors can confirm. Values checked on the p.3 and p.5 images.
Second check: another model agreed
Report a problem with this finding
Wording The Ghana location is "Northern and Upper East regions" in Table 1 but "Northern Region" in Table 2.
- p. 4 Table 1, Ghana location
Northern and Upper East regions
- p. 7 Table 2, Ghana location
Northern Region
Judge's note: Table 2 may simply be abbreviated, but the two tables are inconsistent.
Second check: another model agreed
Report a problem with this finding
Wording Savings text says "In India and Ghana, individuals were able to save at program meetings ... but in the other four sites, households had to make deposits at the financial institution," yet the other four include Pakistan, where Table 1 lists the financial institution partner as "None" and households saved at home or in ROSCAs.
- p. 6 Savings
but in the other four sites, households had to make deposits at the financial institution
- p. 4 Table 1, Pakistan financial partner
None
Only the authors can say what is right here.
Judge's note: Page 6: the remaining four sites are Ethiopia, Honduras, Pakistan and Peru. Page 4 Table 1 gives Pakistan's partner as "None", and page 5 says Pakistan households were encouraged to save at home, in boxes or with ROSCAs. The paper gives no explanation for the difference.
Second check: another model agreed
Report a problem with this finding
Wording Table 2 says for Ethiopia "One public lottery in each of 10 tabias ... In each tabia, 50 were selected for treatment status, 50 for control," which gives 500 treatment and 1,000 households, but the same table lists 458 chosen in randomization and a sample size of 925.
- p. 7 Table 2, Ethiopia
In each tabia, 50 were selected for treatment status, 50 for control
- p. 7 Table 2, Ethiopia chosen
458
Only the authors can say what is right here.
Judge's note: Checked Table 2 on page 7 and the Ethiopia text on page 6 (458 participants). I found no explanation, such as post-lottery exclusions or unequal tabia sizes. Only the authors can reconcile the figures.
Second check: another model agreed
Report a problem with this finding
Wording "all the effects on economic variables are still significant, and usually similar to or larger than after endline 1". Table 3 shows declines in 3 of the 6 economic indices.
- p. 9 Pooled sample
all the effects on economic variables are still significant, and usual- ly similar to or larger than after endline 1
- p. 9 next sentences
The gains in financial inclusion, total time spent working, income and revenue, and mental health have declined
Judge's note: Table 3 (p.10), endline 1 to endline 2: consumption 0.122 to 0.120; food security 0.107 to 0.113; assets 0.258 to 0.249; financial inclusion 0.367 to 0.212; time working 0.090 to 0.054; income and revenues 0.383 to 0.273. Only half are similar or larger, so 'usually' overstates. The very next sentence does acknowledge the declines, so this is a wording point only.
Second check: another model agreed
Report a problem with this finding
Wording "PPP US$4.55 per capita per month, or roughly 5% of control group mean of PPP US$78.80": the ratio is about 5.8%.
- p. 9 Pooled sample, endline 1
PPP US$4.55 per capita per month, or roughly 5% of control group mean of PPP US$78.80
- p. 9 Results, pooled sample
equivalent to PPP US$4.55 per capita per month, or roughly 5% of control group mean of PPP US$78.80
Judge's note: Checked on the p.9 image. 4.55/78.80 = 5.77%, which would normally round to 6%. 'Roughly' allows some slack, so this is a minor approximation point.
Second check: another model agreed
Report a problem with this finding
Wording The text says "the increase in assets held by the households is lower than the value of the asset in all countries but Ethiopia", but Table 4 row 12 gives Ethiopia 97%, also below 100%. The p.9 text ("The impact of the program on asset values is lower than the cost of the assets") makes no exception.
- p. 13 Cost-benefit analysis
lower than the value of the asset in all countries but Ethiopia
- p. 12 Table 4 row 12
97%
- p. 13 Cost-benefit analysis, third column
the increase in assets held by the households is lower than the value of the asset in all countries but Ethiopia
- p. 12 Table 4, row 12
97% 32% 8% 43% 17% 16%
Only the authors can say what is right here.
Judge's note: Checked Table 4 on p.12 and the text on pp.9 and 13. Row 12 is 97/32/8/43/17/16%, all below 100%. Ethiopia exceeds 100% only at the 75th and 90th percentiles (rows 16-17), and India does too at the 90th. The exception is not supported by the average measure the text refers back to ('As noted previously'). Ethiopia is close to 100%, so this may be loose wording.
Second check: another model agreed
Report a problem with this finding
Wording The cost-benefit text says households drew down assets "in the first year, but there is no further decline between year 1 and year 2". The paper defines endline 1 as year 2 and endline 2 as year 3, and on p.9 says asset impacts are "stable from year 2 to year 3".
- p. 13 Cost-benefit analysis
there is no further decline between year 1 and year 2
- p. 9 Pooled results
stable from year 2 to year 3 (Table 3)
- p. 9 Pooled sample
the pro- gram impact on asset holdings is stable from year 2 to year 3 (Table 3)
Only the authors can say what is right here.
Judge's note: Checked p.9 (endline 1 = 'year 2 of the study', endline 2 = 'year 3'; 'stable from year 2 to year 3 (Table 3)') and p.13. Table 4 also labels the endlines as years 2 and 3. No assets were measured in year 1, so the year labels in the Discussion sentence do not match the paper's own convention.
Second check: another model agreed
Report a problem with this finding
Wording Honduras paragraph says "In both endlines we do see an increase in revenue coming from chickens, as well as a significant increase in food consumption", but the text reports a significant food-consumption increase for Honduras only "in endline 1" and says households "were not consuming more" by endline 2.
- p. 11 Second point (Honduras)
In both endlines we do see an in- crease in revenue coming from chickens, as well as a significant increase in food consumption.
- p. 11 Turning to the key variables
statistically significant increase in food con- sumption for Honduras in endline 1
- p. 11 Same Honduras paragraph
and were not consum- ing more
- p. 11
In both endlines we do see an in-crease in revenue coming from chickens
Only the authors can say what is right here.
Judge's note: Checked p.11 image. The left column says the food consumption gain is significant for Honduras in endline 1 (tables S5a-1, S5a-2). The right column puts the food consumption increase in the same 'In both endlines' clause, then says households 'were not consuming more' by endline 2. As written the sentence reads as claiming a significant increase at both endlines. It may be meant to cover only chicken revenue. Only the authors can say which reading is intended.
Second check: another model agreed
Report a problem with this finding
Wording The text says the effects on consumption and the income index "are all increasing with the quantiles," but endline 1 consumption in Table 5 runs 0.065, 0.067, 0.045, 0.099, 0.140, so it is not monotone.
- p. 13 Effects on distribution, Third point
are all increasing with the quantiles
- p. 14 Table 5, consumption endline 1
0.065*** 0.067*** 0.045** 0.099*** 0.140**
Judge's note: The income index is monotone at both endlines and consumption is monotone at endline 2. Only endline 1 consumption dips at the median, and that is the endline the sentence uses as its example.
Second check: another model agreed
Report a problem with this finding
Wording The Discussion says improvements in self-reported well-being "at endline 1, are visible at all level of the distribution except for the 90th percentile". The Table 5 mental health index at endline 1 is 0.005 (0.025), not significant, at the 25th percentile as well as 0.000 at the 90th. 'level' should also read 'levels'.
- p. 15 Discussion, behavioral theories paragraph
are visible at all level of the distribution except for the 90th percentile
- p. 14 Table 5, Mental health, column 2
0.005
- p. 14 Table 5, mental health index, endline 1
0.108*** 0.005 0.112*** 0.082*** 0.000
- p. 15 Discussion, behavioral theories paragraph
are visible at all level of the distribution
Only the authors can say what is right here.
Judge's note: Checked the Table 5 mental health row on p.14 (0.108***, 0.005, 0.112***, 0.082***, 0.000) and the p.15 text. The text may refer to a single self-reported variable not shown in quantile form. If it refers to the Table 5 index, the 25th percentile is also null. The number-agreement slip is also confirmed.
Second check: another model agreed
Report a problem with this finding
Wording Note 11 says Ghana households received support "during the 6-month lean season for both years" but that "the total duration of support received was 14 months"; two 6-month seasons give 12 months.
- p. 16 Note 11
6-month lean season for both years
Only the authors can say what is right here.
Judge's note: The 14 months also lies outside the main-text range of "between 4 and 13 months" (page 3). Only the authors can say whether the seasons differed in length or a number is wrong.
Second check: another model agreed
Report a problem with this finding
Wording Note 14 says "The U.S. CPIs used for 2007 to 2014 are, respectively, 207.3, 215.303, 214.537, 218.056, 224.939, 232.957, and 236.9111", which gives seven values for eight years.
- p. 16 Note 14
207.3, 215.303, 214.537, 218.056, 224.939, 232.957, and 236.9111
- p. 16 Footnote 14
The U.S. CPIs used for 2007 to 2014 are, respectively, 207.3, 215.303, 214.537, 218.056, 224.939, 232.957, and 236.9111.
Only the authors can say what is right here.
Judge's note: Checked the p.16 image: seven values are printed for 2007–2014 (eight years), so 'respectively' cannot map them to the years. Rests on outside knowledge: the claim that the missing value is 2012 (about 229.6) depends on the published CPI series, which is not in the files.
Second check: another model agreed
Report a problem with this finding
Typo The text gives "Progress out of Poverty (PPI) scorecard"; the expansion lacks 'Index', so it does not match the acronym.
- p. 3 Targeting
Progress out of Poverty (PPI) scorecard
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the full name is 'Progress out of Poverty Index'. Within the paper, the expansion as printed does not yield the acronym 'PPI'.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 (Honduras eligibility): "(iv) not a benificiary" should read "beneficiary".
- p. 4 Table 1
not a benificiary
- p. 4 Table 1, India method
by NGO verificiation
- p. 5 Table 1, Honduras savings
avereage account
- p. 5 Table 1, Honduras consumption
Treatment houeholds
Judge's note: Visible on the p.4 image. The other misspellings listed in this cluster's first reviewer note are judged separately in C012 (verificiation), C014 (avereage) and C015 (houeholds).
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 (India, method of identifying participants): "by NGO verificiation" should read "verification".
- p. 4 Table 1, Method of identifying participants, India
NGO verificiation
Judge's note: Visible on the p.4 image. The Pakistan cell next to it spells 'verification' correctly.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 and the Acknowledgments give the organisation name as "Organización de Desarollo Empresarial Feminino Social".
- p. 4 Table 1, Implementing NGO, Honduras
Organización de Desarollo Empresarial Feminino Social
- p. 17 Acknowledgments
Organización de Desarollo
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the correct Spanish spellings are 'Desarrollo' and 'Femenino', and the organisation's registered name presumably uses them. The same spelling appears consistently on page 4 and page 17.
Second check: none (judge only)
Report a problem with this finding
Typo Misspellings: "Komolgorov-Smirnov" (Kolmogorov) and "partially our entirely" (or) in the Table 5 notes; "benificiary", "verificiation", "avereage" and "houeholds" in Table 1.
- p. 14 Table 5 notes
Komolgorov-Smirnov
- p. 14 Table 5 notes
partially our entirely
- p. 14 Table 5, Notes 2
relied either partially our entirely upon categorical variables
- p. 4 Table 1
benificiary
Judge's note: All six spellings are visible on the page images: p.14 Table 5 notes, and the Table 1 crops from pp.4-5.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 (Honduras savings component): "avereage account balance" should read "average".
- p. 5 Table 1, Savings component, Honduras
avereage account
Judge's note: Visible on the p.5 image.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 (Honduras consumption support): "Treatment houeholds" should read "households".
- p. 5 Table 1, Consumption Support, Honduras
Treatment houeholds
Judge's note: Visible on the p.5 image.
Second check: none (judge only)
Report a problem with this finding
Typo The Honduras savings match is described as "savings matching semi-annually" in the text (p. 6) but "savings matching biannually" in Table 1 (p. 5).
- p. 6 Savings text
savings matching semi-annually
- p. 5 Table 1, Honduras savings
savings matching biannually equal
Judge's note: Both wordings are printed. "Biannually" usually means twice a year, so the two likely agree, but the wording is inconsistent and "biannually" can be ambiguous. This is a consistency point only.
Second check: none (judge only)
Report a problem with this finding
Typo Table 4 note 1 has an unclosed parenthesis: "(identifying the district and poorest individuals, and screening are included in the staff costs."
- p. 12 Table 4, Notes 1
(identifying the district and poorest individuals, and screening are included in the staff costs.
Judge's note: The page image confirms the opening bracket is never closed. In the image there is a space after "costs.", so the missing-space point is an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo In Table 5, the women's empowerment endline 1 75th-percentile standard error is printed as a bare "0.000", without the parentheses used for every other standard error.
- p. 14 Table 5, Women's empowerment, col 4 SE row
0.000
Judge's note: Checked in the page image: the SE row for column 4 reads 0.000 without parentheses.
Second check: none (judge only)
Report a problem with this finding
Typo Table 5 note 3 misspells the test name: "We perform a Komolgorov-Smirnov test" (should be Kolmogorov-Smirnov).
- p. 14 Table 5, Notes 3
We perform a Komolgorov-Smirnov test
Judge's note: The misspelling is visible in the Table 5 notes on page 14.
Second check: none (judge only)
Report a problem with this finding
Typo Table 5, Women's empowerment index, endline 1, 75th percentile: the standard-error cell reads "0.000" without parentheses, unlike every other SE cell in the table.
- p. 14 Table 5, Women's empowerment, Endline 1, 75th percentile (col 4)
(0.008) 0.000 (0.023)
Judge's note: The crop shows the bare 0.000 in the SE row of column 4. This is a formatting slip; it may also be a misplaced value.
Second check: none (judge only)
Report a problem with this finding
Typo The microcredit paragraph has the garbled phrase "there is no effect of an impact on consumption over a 1- to 3-year time period" (likely "no impact on consumption").
- p. 15 Discussion, microcredit paragraph
there is no effect of an impact on consumption
- p. 15
there is no effect of an impact on consumption over a 1- to 3-year time period
Judge's note: Checked the p.15 image. The phrase is printed as quoted; the meaning is clear, so this is a copy-edit slip.
Second check: none (judge only)
Report a problem with this finding
Typo A stray closing parenthesis follows the URL in ref 1 ("org/handle/10986/16091).") and in ref 49 ("Haushofer_Shapiro_UCT_2013.pdf).").
- p. 16 Reference 1
https://openknowledge.worldbank. org/handle/10986/16091).
- p. 17 Reference 49
Haushofer_Shapiro_UCT_2013.pdf).
Judge's note: Neither entry has an unmatched opening parenthesis before the URL.
Second check: none (judge only)
Report a problem with this finding
Typo Ref 37 lists "A. Banerjee, A. V. Banerjee, E. Duflo, Poor Economics", which appears to duplicate Banerjee as an author.
- p. 17 Ref 37
A. Banerjee, A. V. Banerjee, E. Duflo
- p. 17 Reference 37
A. Banerjee, A. V. Banerjee, E. Duflo, Poor Economics
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that Poor Economics has two authors, A. V. Banerjee and E. Duflo. The repeated name is visible on page 17.
Second check: none (judge only)