Table 4 Summary of ITT estimates of the impacts of the BFY high-cash gift on family well-being and family processes: economic resources and economic pressure

From: Effects of unconditional cash transfers on family processes and wellbeing among mothers with low incomes

Family

Outcome

Hypoth.

 

Age 1

Age 2

Age 3

Pooled Sample

Panel 1: Economic Resources

1

Income-to-needs ratio (including the BFY gift)

+

Effect (con. interval)

 

0.10 (0.00, 0.20)

0.11 (0.00, 0.22)

0.11 (0.02, 0.19)

Std. effect

 

0.12

0.13

0.13

N (deg. freedom)

 

922 (876)

922 (876)

1844 (956)

p-value (WY p-value)

 

0.04 (0.06)

0.04 (0.06)

0.01 (0.02)

1

Household income with gift ($1000 s in 2019 dollars)

+

Effect (con. interval)

 

2.79 (−0.09, 5.68)

2.76 (−0.32, 5.83)

2.86 (0.37, 5.35)

Std. effect

 

0.11

0.11

0.11

N (deg. freedom)

 

922 (876)

922 (876)

1844 (956)

p-value (WY p-value)

 

0.06 (0.06)

0.08 (0.08)

0.03 (0.03)

Panel 2: Economic Pressure

2

Food insecurity index

-

Effect (con. interval)

0.23 (−0.00, 0.46)

−0.00 (−0.24, 0.23)

0.05 (−0.17, 0.27)

0.10 (−0.08, 0.28)

Std. effect

0.14

−0.00

0.03

0.06

N (deg. freedom)

929 (882)

921 (875)

920 (874)

2770 (972)

p-value (WY p-value)

0.05 (0.14)

0.98 (0.98)

0.68 (0.88)

0.27 (0.44)

2

Non-food economic hardship index

-

Effect (con. interval)

0.04 (−0.12, 0.20)

0.07 (−0.09, 0.23)

0.02 (−0.09, 0.14)

0.05 (−0.06, 0.16)

Std. effect

0.04

0.06

0.03

0.04

N (deg. freedom)

930 (883)

921 (875)

922 (876)

2773 (972)

p-value (WY p-value)

0.62 (0.62)

0.38 (0.72)

0.68 (0.88)

0.41 (0.44)

2

Expense worry

-

Effect (con. interval)

0.17 (−0.04, 0.38)

0.08 (−0.14, 0.29)

0.11 (−0.10, 0.31)

0.12 (−0.04, 0.28)

Std. effect

0.10

0.05

0.06

0.07

N (deg. freedom)

930 (883)

919 (873)

919 (873)

2768 (972)

p-value (WY p-value)

0.12 (0.23)

0.48 (0.72)

0.31 (0.63)

0.14 (0.34)

  1. Data collection occurred in July 2019 to June 2020 for age 1, July 2020 to July 2021 for age 2, and July 2021 to July 2022 for age 3. Each block of rows presents, for each outcome, the raw treatment effect with confidence intervals in parentheses; the standardized treatment effect size; number of observations with degrees of freedom in parentheses; and the p-value with Westfall-Young (WY) adjusted p-value in parentheses. The ITT estimates come from two-sided regressions with site fixed effects, controlling for baseline covariates, child age at interview, and phone interview status. Outcomes were standardized using the standard deviation of the low-cash gift within each age. We report the degrees of freedom computed as the sample size minus the number of parameters estimated in the model. This statistic is complicated in the pooled sample because we cluster the standard error to adjust for non-independence. For simplicity, we report the default degrees of freedom reported in most software, which is the number of clusters minus one. The p-value comes from analyses that do not correct for multiple outcomes, while the WY p-value is based on Westfall and Young72 step-down resampling methods of addressing multiple hypothesis testing, where outcomes are grouped in families (following Fig. 1) and their p-values adjusted within each family. The “Pooled Sample” column presents estimates from analyses that pool observations across ages, adjust for age indicators, and cluster the standard error at the individual level. Preregistered, hypothesized directions of the intervention effects are presented with “+” or “-” for directional increase or decrease in the outcome, respectively. Household incomes across all years are inflation-adjusted to 2019 dollars, and the poverty line is based on the 2019 U.S. Census poverty threshold. Income-to-needs is the household income divided by the poverty line for a given family size and composition. Income and income-to-needs have been truncated at the 99th percentile.
Back to article page