Common use of Empirical Model for Determining Clause in Contracts

Empirical Model for Determining. PFS Impact Estimates Analysis of the Preterm Birth, Child Injury, and the Healthy Birth Interval metrics will compare outcomes for Sample Members who were Randomized to the Intervention Group to those who were Randomized to the Control Group. To account for potential non-compliance, the Independent Evaluator will calculate the PFS Impact Estimates using an Instrumental Variable approach. Consider an outcome, ;, such as an indicator for Preterm Birth. For subject i, the estimating equation is: ; Ü L è4 E è5 s:(QUROOHG□□□L□□Q□□□□ ZKHUH ³(QUROOHG LQ 1)3´ PHDQV KDYLQJ UHFHL NFP for service delivery. This model will be estimated using Two-Stage Least Squares (2SLS), where the first stage is: s:(QUROOHG□□□L□Q□□□T□7UHDW□P□□H□Q□□W□□□□ where s:7UHDW□P□isHaQn Windicator variable equal to one if the subject was Randomized to the Intervention Group and zero if the subject was Randomized to the Control Group; : Üis a vector of covariates. These covariates should in theory be uncorrelated with the treatment indicator, but they can aid in the precision of the estimate. The Independent Evaluator will define the exact set of covariates in the Evaluation Plan. Potential candidates include standard demographic variables based on previous NFP trials, such as age (<17), socio- economic status (self-reported income, race/ethnicity, education), sex of the child, smoking status. The exact set of control variables will be determined in the Evaluation Plan in the timeline outlined in Article III of this Annex F. Whether the set of control variables outlined in the Evaluation Plan can be used in the PFS Outcomes Metrics Results Reports depends on the data quality of these variables for the Sample Members. This model estimates the effect of NFP relative to the services consumed by the Control Group. The only source of non-compliance that it explicitly captures is that some Randomized into the Intervention Group may never receive NFP VHUYLFHV WKH ³HQUROOPIHf sQomWe SUamDplWe MHe´mbeLrsVin thVe PDOOHU Control Group receive services from similar home visiting programs that may also affect outcomes, this model estimates the effect of NFP relative to the mix of other home-visiting programs that the control group receives, rather than relative to no home-visiting service at all. The Independent Evaluator will report the share of Control group and Intervention Group members receiving other home visiting services (to the extent that those programs are captured in the data provided to the Independent Evaluator), which the Operations Committee may use as context in interpreting and disseminating the NFP Program impact findings. Analysis of the Coverage of LIZCs will report outcomes for Sample Members in the Intervention Group.

Empirical Model for Determining. PFS Impact Estimates Analysis of the Preterm Birth, Child Injury, and the Healthy Birth Interval metrics will compare outcomes for Sample Members who were Randomized to the Intervention Group to those who were Randomized to the Control Group. To account for potential non-compliance, the Independent Evaluator will calculate the PFS Impact Estimates using an Instrumental Variable approach. Consider an outcome, ;, such as an indicator for Preterm Birth. For subject i, the estimating equation is: ; Ü L è4 E è5 s:(QUROOHG□□□L□□Q□□□□ ZKHUH ³(QUROOHG LQ 1)3´ PHDQV KDYLQJ UHFHL s:Enrolled in .&0 ) where “Enrolled in NFP” means having received at least one home visit from NFP for service delivery. This model will be estimated using Two-Stage Least Squares (2SLS), where the first stage is: s:(QUROOHG□□□L□Q□□□T□7UHDW□P□□H□Q□□W□□□□ s:Enrolled in .&0 ) Treatment where s:7UHDW□P□isHaQn Windicator s:Treatment is an indicator variable equal to one if the subject was Randomized to the Intervention Group and zero if the subject was Randomized to the Control Group; : Üis a vector of covariates. These covariates should in theory be uncorrelated with the treatment indicator, but they can aid in the precision of the estimate. The Independent Evaluator will define the exact set of covariates in the Evaluation Plan. Potential candidates include standard demographic variables based on previous NFP trials, such as age (<17), socio- economic status (self-reported income, race/ethnicity, education), sex of the child, smoking status. The exact set of control variables will be determined in the Evaluation Plan in the timeline outlined in Article III of this Annex F. Whether the set of control variables outlined in the Evaluation Plan can be used in the PFS Outcomes Metrics Results Reports depends on the data quality of these variables for the Sample Members. This model estimates the effect of NFP relative to the services consumed by the Control Group. The only source of non-compliance that it explicitly captures is that some Randomized into the Intervention Group may never receive NFP VHUYLFHV WKH ³HQUROOPIHf sQomWe SUamDplWe MHe´mbeLrsVin thVe PDOOHU services (the “enrollment rate” is smaller than 1). If some Sample Members in the Control Group receive services from similar home visiting programs that may also affect outcomes, this model estimates the effect of NFP relative to the mix of other home-visiting programs that the control group receives, rather than relative to no home-visiting service at all. The Independent Evaluator will report the share of Control group and Intervention Group members receiving other home visiting services (to the extent that those programs are captured in the data provided to the Independent Evaluator), which the Operations Committee may use as context in interpreting and disseminating the NFP Program impact findings. Analysis of the Coverage of LIZCs will report outcomes for Sample Members in the Intervention Group.

Empirical Model for Determining. PFS Impact Estimates Analysis of the Preterm Birth, Child Injury, and the Healthy Birth Interval metrics will compare outcomes for Sample Members who were Randomized to the Intervention Group to those who were Randomized to the Control Group. To account for potential non-compliance, the Independent Evaluator will calculate the PFS Impact Estimates using an Instrumental Variable approach. Consider an outcome, ;, such as an indicator for Preterm Birth. For subject i, the estimating equation is: ; Ü L è4 E è5 s:(QUROOHG□□□L□□Q□□□□ ZKHUH ³(QUROOHG LQ 1)3´ PHDQV KDYLQJ UHFHL ! = ! + !1(Enrolled in NFP)! + !! + ! where “Enrolled in NFP” means having received at least one home visit from NFP for service delivery. This model will be estimated using Two-Stage Least Squares (2SLS), where the first stage is: s:(QUROOHG□□□L□Q□□□T□7UHDW□P□□H□Q□□W□□□□ 1(Enrolled in NFP)! = ! + !1(Treatment)! + !! + ! where s:7UHDW□P□isHaQn Windicator 1(Treatment)! is an indicator variable equal to one if the subject was Randomized to the Intervention Group and zero if the subject was Randomized to the Control Group; : Üis ! is a vector of covariates. These covariates should in theory be uncorrelated with the treatment indicator, but they can aid in the precision of the estimate. The Independent Evaluator will define the exact set of covariates in the Evaluation Plan. Potential candidates include standard demographic variables based on previous NFP trials, such as age (<17), socio- economic status (self-reported income, race/ethnicity, education), sex of the child, smoking status. The exact set of control variables will be determined in the Evaluation Plan in the timeline outlined in Article III of this Annex F. Whether the set of control variables outlined in the Evaluation Plan can be used in the PFS Outcomes Metrics Results Reports depends on the data quality of these variables for the Sample Members. This model estimates the effect of NFP relative to the services consumed by the Control Group. The only source of non-compliance that it explicitly captures is that some Randomized into the Intervention Group may never receive NFP VHUYLFHV WKH ³HQUROOPIHf sQomWe SUamDplWe MHe´mbeLrsVin thVe PDOOHU services (the “enrollment rate” is smaller than 1). If some Sample Members in the Control Group receive services from similar home visiting programs that may also affect outcomes, this model estimates the effect of NFP relative to the mix of other home-visiting programs that the control group receives, rather than relative to no home-visiting service at all. The Independent Evaluator will report the share of Control group and Intervention Group members receiving other home visiting services (to the extent that those programs are captured in the data provided to the Independent Evaluator), which the Operations Committee may use as context in interpreting and disseminating the NFP Program impact findings. Analysis of the Coverage of LIZCs will report outcomes for Sample Members in the Intervention Group.

Empirical Model for Determining. PFS Impact Estimates Analysis of the Preterm Birth, Child Injury, and the Healthy Birth Interval metrics will compare outcomes for Sample Members who were Randomized to the Intervention Group to those who were Randomized to the Control Group. To account for potential non-compliance, the Independent Evaluator will calculate the PFS Impact Estimates using an Instrumental Variable approach. Consider an outcome, ;, such as an indicator for Preterm Birth. For subject i, the estimating equation is: ; Ü L è4 E è5 s:(QUROOHG□□□L□□Q□□□□ ZKHUH ³(QUROOHG LQ 1)3´ PHDQV KDYLQJ UHFHL = 0 + 1 1(Enrolled in NFP) + 2 + where “Enrolled in NFP” means having received at least one home visit from NFP for service delivery. This model will be estimated using Two-Stage Least Squares (2SLS), where the first stage is: s:(QUROOHG□□□L□Q□□□T□7UHDW□P□□H□Q□□W□□□□ 1(Enrolled in NFP) = 0 + 11(Treatment) + 2 + where s:7UHDW□P□isHaQn Windicator 1(Treatment) is an indicator variable equal to one if the subject was Randomized to the Intervention Group and zero if the subject was Randomized to the Control Group; : Üis is a vector of covariates. These covariates should in theory be uncorrelated with the treatment indicator, but they can aid in the precision of the estimate. The Independent Evaluator will define the exact set of covariates in the Evaluation Plan. Potential candidates include standard demographic variables based on previous NFP trials, such as age (<17), socio- economic status (self-reported income, race/ethnicity, education), sex of the child, smoking status. The exact set of control variables will be determined in the Evaluation Plan in the timeline outlined in Article III of this Annex F. Whether the set of control variables outlined in the Evaluation Plan can be used in the PFS Outcomes Metrics Results Reports depends on the data quality of these variables for the Sample Members. This model estimates the effect of NFP relative to the services consumed by the Control Group. The only source of non-compliance that it explicitly captures is that some Randomized into the Intervention Group may never receive NFP VHUYLFHV WKH ³HQUROOPIHf sQomWe SUamDplWe MHe´mbeLrsVin thVe PDOOHU services (the “enrollment rate” is smaller than 1). If some Sample Members in the Control Group receive services from similar home visiting programs that may also affect outcomes, this model estimates the effect of NFP relative to the mix of other home-visiting programs that the control group receives, rather than relative to no home-visiting service at all. The Independent Evaluator will report the share of Control group and Intervention Group members receiving other home visiting services (to the extent that those programs are captured in the data provided to the Independent Evaluator), which the Operations Committee may use as context in interpreting and disseminating the NFP Program impact findings. Analysis of the Coverage of LIZCs will report outcomes for Sample Members in the Intervention Group.