SlideShare a Scribd company logo
1 of 166
Peter M. Lance, PhD
MEASURE Evaluation
University of North Carolina at
Chapel Hill
MARCH 31, 2016
Fundamentals of Program
Impact Evaluation
Global, five-year, $180M cooperative agreement
Strategic objective:
To strengthen health information systems – the
capacity to gather, interpret, and use data – so
countries can make better decisions and sustain good
health outcomes over time.
Project overview
Improved country capacity to manage health
information systems, resources, and staff
Strengthened collection, analysis, and use of
routine health data
Methods, tools, and approaches improved and
applied to address health information challenges
and gaps
Increased capacity for rigorous evaluation
Phase IV Results Framework
Global footprint (more than 25 countries)
How Do We Know IfAProgram MadeADifference?
ABrief Helicopter Tour of Methods for Estimating Program Impact
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
• The Program Impact Evaluation Challenge
• Randomization
• Selection on observables
• Within estimators
• Instrumental variables
Newton’s “Laws” of Motion
𝐹𝑜𝑟𝑐𝑒 = 𝑀𝑎𝑠𝑠 ∙ 𝐴𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛
Did the program make a
difference?
Did the program cause a change in an
outcome of interest Y ?
(Causality)
Our outcome of Interest
What happens if an individual does
not participate in a program
What happens if that individual does
participate in a program
Potential Outcomes
𝑌 :
𝑌0
:
𝑌1
:
Our outcome of interest
What happens if an individual does
not participate in a program
What happens if that individual does
participate in a program
Potential Outcomes
𝑌𝑖 :
𝑌𝑖
0
:
𝑌𝑖
1
:
Our outcome of interest
What happens if an individual does
not participate in a program
What happens if that individual does
participate in a program
Potential Outcomes
𝑌𝑖 :
𝑌𝑖
0
:
𝑌𝑖
1
:
Our outcome of interest
What happens if an individual does
not participate in a program
What happens if that individual does
participate in a program
Potential Outcomes
𝑌𝑖 :
𝑌𝑖
0
:
𝑌𝑖
1
:
What happens
if the individual
participates
{Causal} Program Impact
𝑌𝑖
1
− 𝑌𝑖
0
= Program Impact
What happens
if the individual
does not
participate
What happens
if the individual
participates
{Causal} Program Impact
𝑌𝑖
1
− 𝑌𝑖
0
= Program Impact
What happens
if the individual
does not
participate
What happens
if the individual
participates
{Causal} Program Impact
𝑌𝑖
1
− 𝑌𝑖
0
= Program Impact
What happens
if the individual
does not
participate
What happens
if the individual
participates
{Causal} Program Impact
𝑌𝑖
1
− 𝑌𝑖
0
= Program Impact
What happens
if the individual
does not
participate
What happens
if the individual
participates
{Causal} Program Impact
𝑌𝑖
1
− 𝑌𝑖
0
= Program Impact
What happens
if the individual
does not
participate
𝑃𝑖 =
1ifindividual 𝑖 participates
0if individual 𝑖 does not participate
Program Participation
𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖
1
+ 1 − 𝑃𝑖 ∙ 𝑌𝑖
0
Observed Outcome
𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖
1
+ 1 − 𝑃𝑖 ∙ 𝑌𝑖
0
Observed Outcome
𝑃𝑖 = 1
𝑌𝑖 = 1 ∙ 𝑌𝑖
1
+ 1 − 1 ∙ 𝑌𝑖
0
Observed Outcome
𝑃𝑖 = 1
𝑌𝑖 = 𝑌𝑖
1
+ 0 ∙ 𝑌𝑖
0
Observed Outcome
𝑃𝑖 = 1
𝑌𝑖 = 𝑌𝑖
1
Observed Outcome
𝑃𝑖 = 1
𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖
1
+ 1 − 𝑃𝑖 ∙ 𝑌𝑖
0
Observed Outcome
𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖
1
+ 1 − 𝑃𝑖 ∙ 𝑌𝑖
0
Observed Outcome
𝑃𝑖 = 0
𝑌𝑖 = 0 ∙ 𝑌𝑖
1
+ 1 − 0 ∙ 𝑌𝑖
0
Observed Outcome
𝑃𝑖 = 0
𝑌𝑖 = 𝑌𝑖
0
Observed Outcome
𝑃𝑖 = 0
𝑌𝑖
1
, 𝑌𝑖
0
Observed Outcome
𝑌𝑖
1
, 𝑌𝑖
0
Observed Outcome
𝑌𝑖
1
, 𝑌𝑖
0
Observed Outcome
𝑌𝑖
1
, 𝑌𝑖
0
Observed Outcome
𝑌𝑖
1
, 𝑌𝑖
0
Observed Outcome
Fundamental Identification
Problem of Program Impact
Evaluation
𝑌𝑖
1
, 𝑌𝑖
0
Observed Outcome
Fundamental Identification
Problem of Program Impact
Evaluation
Individual Population
Individual Population
Hi. They call me
individual i
Individual Population
?!?
𝑌𝑖
1
, 𝑌𝑖
0
𝑌𝑖
1
, 𝑌𝑖
0
An expected value for a random variable is the
average value from a large number of repetitions
of the experiment that random variable represents
An expected value is the true average of a random
variable across a population
Expected Value
An expected value for a random variable is the
average value from a large number of repetitions
of the experiment that random variable represents
An expected value is the true average of a random
variable across a population
Expected Value
An expected value is the true average of a random
variable across a population
𝐸 𝑋 = sometruevalue
Expected Value
𝐸 𝑐 = 𝑐
𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊
𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍
𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍
𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍
Expectations: Properties
𝑬 𝒄 = 𝒄
𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊
𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍
𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍
𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍
Expectations: Properties
𝐸 𝑐 = 𝑐
𝑬 𝒄 ∙ 𝑾 = 𝒄 ∙ 𝑬 𝑾
𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍
𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍
𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍
Expectations: Properties
𝐸 𝑐 = 𝑐
𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊
𝑬 𝑾 + 𝒁 = 𝑬 𝑾 + 𝑬 𝒁
𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍
𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍
Expectations: Properties
𝐸 𝑐 = 𝑐
𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊
𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍
𝑬 𝑾 − 𝒁 = 𝑬 𝑾 − 𝑬 𝒁
𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍
Expectations: Properties
𝐸 𝑐 = 𝑐
𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊
𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍
𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍
𝑬 𝒂 ∙ 𝑾 ± 𝒃 ∙ 𝒁 = 𝒂 ∙ 𝑬 𝑾 ± 𝒃 ∙ 𝑬 𝒁
Expectations: Properties
𝐸 𝑊 ∙ 𝑍 ≠ 𝐸 𝑊 ∙ 𝐸 𝑍
𝐸
𝑊
𝑍
≠
𝐸 𝑊
𝐸 𝑍
𝐸 𝑓 𝑊 ≠ 𝑓 𝐸 𝑊
Expectations: Properties
𝑬 𝑾 ∙ 𝒁 ≠ 𝑬 𝑾 ∙ 𝑬 𝒁
𝐸
𝑊
𝑍
≠
𝐸 𝑊
𝐸 𝑍
𝐸 𝑓 𝑊 ≠ 𝑓 𝐸 𝑊
Expectations: Properties
𝐸 𝑊 ∙ 𝑍 ≠ 𝐸 𝑊 ∙ 𝐸 𝑍
𝑬
𝑾
𝒁
≠
𝑬 𝑾
𝑬 𝒁
𝐸 𝑓 𝑊 ≠ 𝑓 𝐸 𝑊
Expectations: Properties
𝐸 𝑊 ∙ 𝑍 ≠ 𝐸 𝑊 ∙ 𝐸 𝑍
𝐸
𝑊
𝑍
≠
𝐸 𝑊
𝐸 𝑍
𝑬 𝒇 𝑾 ≠ 𝒇 𝑬 𝑾
Expectations: Properties
𝑌𝑖
1
− 𝑌𝑖
0
Average Treatment Effect (ATE)
𝐸 𝑌1 − 𝑌0
Average Effect of Treatment on the Treated (ATT)
𝐸 𝑌1 − 𝑌0|𝑃 = 1
Hi
there
Individual Impact
𝑌𝑖
1
− 𝑌𝑖
0
𝐸 𝑌𝑖
1
− 𝑌𝑖
0
Average Treatment Effect (ATE)
𝐸 𝑌1 − 𝑌0
Average Effect of Treatment on the Treated (ATT)
𝐸 𝑌1 − 𝑌0|𝑃 = 1
Treatment Effects
Suppose that we have a sample of 𝑖 = 1,…, 𝑛
individuals….
…but for each individual 𝑖 we observe either
𝑌𝑖
1
or 𝑌𝑖
0
…
…but not both
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Suppose that we have a sample of 𝑖 = 1,…, 𝑛
individuals….
…but for each individual 𝑖 we observe either
𝑌𝑖
1
or 𝑌𝑖
0
…
…but not both
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Remember, however, a key property of expectations:
𝐸 𝑌1
− 𝑌0
= 𝐸 𝑌1
− 𝐸 𝑌0
…but this means that in principle we could
estimate E 𝑌1
and E 𝑌0
separately
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Remember, however, a key property of expectations:
𝐸 𝑌1
− 𝑌0
= 𝐸 𝑌1
− 𝐸 𝑌0
…but this means that in principle we could
estimate E 𝑌1
and E 𝑌0
separately
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
For instance, suppose that in our sample we have:
𝑛 𝑃
participants(𝑃𝑖 = 1)
and
𝑛 𝑁
non-participants(𝑃𝑖 = 0)
(hence 𝑛 𝑃
+ 𝑛 𝑁
= 𝑛)
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Then an estimator of 𝐸 𝑌1
is
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
calculated with the 𝑛 𝑃
participants out of the
sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Then an estimator of 𝐸 𝑌1
is
𝒀 𝟏 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
calculated with the 𝑛 𝑃
participants out of the
sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Then an estimator of 𝐸 𝑌1
is
𝑌1 =
𝑗=1
𝒏 𝑷
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
calculated with the 𝑛 𝑃
participants out of the
sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Then an estimator of 𝐸 𝑌1
is
𝑌1 =
𝑗=1
𝑛 𝑃
𝒀𝒋
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
calculated with the 𝑛 𝑃
participants out of the
sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Then an estimator of 𝐸 𝑌1
is
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝒏 𝑷
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
calculated with the 𝑛 𝑃
participants out of the
sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Then an estimator of 𝐸 𝑌1
is
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝒀𝒋
𝟏
𝑛 𝑃
calculated with the 𝑛 𝑃
participants out of the
sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
Similarly, an estimator of 𝐸 𝑌0
is
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
calculated with the 𝑛 𝑁
non-participants out of
the sample of 𝑛 individuals
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
So then an estimate of
𝐸 𝑌1
− 𝑌0
= 𝐸 𝑌1
− 𝐸 𝑌0
is
𝑌1 − 𝑌0 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
−
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
−
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
So how do we estimate
𝑬 𝒀 𝟏
− 𝒀 𝟎
??
But is it a good estimate??
So we have two samples of size 𝒏
By random chance, between the two samples we almost surely have
1. A different precise mix of individuals
2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁)
3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 :
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
So we have two samples of size 𝒏
By random chance, between the two samples we almost surely have
1. A different precise mix of individuals
2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁)
3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 :
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
So we have two samples of size 𝒏
By random chance, between the two samples we almost surely have
1. A different precise mix of individuals
2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁)
3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 :
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
So we have two samples of size 𝒏
By random chance, between the two samples we almost surely have
1. A different precise mix of individuals
2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁)
3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 :
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
So we have two samples of size 𝒏
By random chance, between the two samples we almost surely have
1. A different precise mix of individuals
2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁)
3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 :
𝑌1 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
=
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
So we have two samples of size 𝒏
By random chance, between the two samples we almost surely have
1. A different precise mix of individuals
2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁)
3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 :
𝒀 𝟏 =
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝒏 𝑷
=
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑌0 =
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
𝒀 𝟏 𝑬 𝒀 𝟏
𝒀 𝟏 =
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝒏 𝑷
=
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝒀 𝟏 =
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝒏 𝑷
=
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝒀 𝟏 =
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝒏 𝑷
=
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝒀 𝟏 =
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝒏 𝑷
=
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 = 𝒏 𝑷 ∙ 𝑬
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝑬
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁
𝑬
𝒊=𝟏
𝒏 𝑷
𝑿𝒊 =
𝒊=𝟏
𝒏 𝑷
𝑬 𝑿𝒊
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 = 𝒏 𝑷 ∙ 𝑬
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝑬
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁
𝑬
𝒊=𝟏
𝒏 𝑷
𝑿𝒊 =
𝒊=𝟏
𝒏 𝑷
𝑬 𝑿𝒊
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁
𝑬
𝒊=𝟏
𝒏 𝑷
𝑿𝒊 =
𝒊=𝟏
𝒏 𝑷
𝑬 𝑿𝒊
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁
𝑬
𝒊=𝟏
𝒏 𝑷
𝑿𝒊 =
𝒊=𝟏
𝒏 𝑷
𝑬 𝑿𝒊
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝒏 𝑷
∙ 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝒏 𝑷
∙ 𝑬 𝒀𝒋
𝟏
1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿
𝑬
𝟏
𝒏 𝑷
∙ 𝑿 =
𝟏
𝒏 𝑷
∙ 𝑬 𝑿
𝑬 𝒀 𝟏 = 𝑬
𝒋=𝟏
𝒏 𝑷
𝒀𝒋
𝟏
𝒏 𝑷
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝑬
𝒋=𝟏
𝒏 𝑷
𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 =
𝟏
𝒏 𝑷
∙ 𝒏 𝑷
∙ 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀𝒋
𝟏
= 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀𝒋
𝟏
= 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋
𝟏
𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
𝑬 𝒀𝒋
𝟏
= 𝑬 𝒀 𝟏
𝑬 𝒀𝒋
𝟏
= 𝑬 𝒀 𝟏
𝑬 𝒀 𝟏
𝑃 = 0
𝑃 = 0
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝑃 = 0
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝒀 𝟏
𝑃 = 0
𝑃 = 0
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 0
𝑃 = 1
𝑃 = 1
𝑃 = 0
𝑃 = 0
𝑃 = 1
𝑃 = 0
𝒀 𝟏
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝒀 𝟏
Z W
“Z Causes W”
𝑬 𝑾|𝒁 ≠ 𝑬 𝑾
Z W
“Z causes W”
𝑬 𝑾|𝒁 ≠ 𝑬 𝑾
Z W
“Z causes W”
𝑬 𝑾|𝒁 ≠ 𝑬 𝑾
X Y1
X
Y
P
X
Y
P
0
X
Y
P
X
Y
P
𝐸 𝑋|𝑃 = 1 ≠ 𝐸 𝑋
𝐸 𝑌1
|𝑃 = 1 ≠ 𝐸 𝑌1
X Y1
𝐸 𝑋|𝑃 = 1 ≠ 𝐸 𝑋
𝐸 𝑌1
|𝑃 = 1 ≠ 𝐸 𝑌1
X Y1
𝐸 𝑋|𝑃 = 1 ≠ 𝐸 𝑋
𝐸 𝑌1
|𝑃 = 1 ≠ 𝐸 𝑌1
X Y1
X
Y
P
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝑃 = 1
𝐸 𝑌0
|𝑃 = 0 ≠ 𝐸 𝑌0
𝐸 𝑌0
|𝑃 = 1 ≠ 𝐸 𝑌0
𝐸 𝑌0
|𝑃 = 0 ≠ 𝐸 𝑌0
|𝑃 = 0
𝐸 𝑌1
|𝑃 = 0 ≠ 𝐸 𝑌1
𝐸 𝑌1
|𝑃 = 1 ≠ 𝐸 𝑌1
𝐸 𝑌1
|𝑃 = 0 ≠ 𝐸 𝑌1
|𝑃 = 0
𝐸 𝑌0
|𝑃 = 0 ≠ 𝐸 𝑌0
𝐸 𝑌0
|𝑃 = 1 ≠ 𝐸 𝑌0
𝐸 𝑌0
|𝑃 = 0 ≠ 𝐸 𝑌0
|𝑃 = 0
𝐸 𝑌1
|𝑃 = 0 ≠ 𝐸 𝑌1
𝐸 𝑌1
|𝑃 = 1 ≠ 𝐸 𝑌1
𝐸 𝑌1
|𝑃 = 0 ≠ 𝐸 𝑌1
|𝑃 = 0
The estimator
𝑌1 − 𝑌0 =
𝑗=1
𝑛 𝑃
𝑌𝑗
𝑛 𝑃
−
𝑘=1
𝑛 𝑁
𝑌𝑘
𝑛 𝑁
=
𝑘=1
𝑛 𝑁
𝑌𝑘
0
𝑛 𝑁
−
𝑗=1
𝑛 𝑃
𝑌𝑗
1
𝑛 𝑃
of
𝐸 𝑌1
− 𝑌0
would be biased if some individuals occurred only
among participants or non-participants
Or
more often among one of the two groups
X
Y
P
X
Y
P
Sir Austin Bradford Hill
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
Strength: How strong is the relationship?
Consistency: How consistently is link found?
Specificity: How specific is the setting or disease?
Temporality: Does the cause precede the effect?
Gradient: Does more cause lead to more effect?
Analogy: Do similar “causes” have similar effect?
Coherence: Are field and laboratory findings similar?
Experiment: Was variation in the cause random?
Plausibility: Does theory agree?
Bradford Hill Criteria
We are presented with data
in the form of a sample:
Causality: Our Approach
𝒀𝒊, 𝑷𝒊, 𝑿𝒊 ,
𝒊 = 𝟏, . . , 𝒏
We are presented with data
in the form of a sample:
Causality: Our Approach
𝒀𝒊, 𝑷𝒊, 𝑿𝒊 ,
𝒊 = 𝟏, . . , 𝒏
Assumptions Model
E(Y1-Y0),
E(Y1-Y0|P=1),
Etc.
We are presented with data
in the form of a sample:
Causality: Our Approach
𝒀𝒊, 𝑷𝒊, 𝑿𝒊 ,
𝒊 = 𝟏, . . , 𝒏
Assumptions Model
E(Y1-Y0),
E(Y1-Y0|P=1),
Etc.
Conclusion
Links:
The manual:
http://www.measureevaluation.org/resources/publications/ms-
14-87-en
The webinar introducing the manual:
http://www.measureevaluation.org/resources/webinars/metho
ds-for-program-impact-evaluation
My email:
pmlance@email.unc.edu
MEASURE Evaluation is funded by the U.S. Agency
for International Development (USAID) under terms
of Cooperative Agreement AID-OAA-L-14-00004 and
implemented by the Carolina Population Center, University
of North Carolina at Chapel Hill in partnership with ICF
International, John Snow, Inc., Management Sciences for
Health, Palladium Group, and Tulane University. The views
expressed in this presentation do not necessarily reflect
the views of USAID or the United States government.
www.measureevaluation.org

More Related Content

What's hot

Monitoring evaluation
Monitoring evaluationMonitoring evaluation
Monitoring evaluation
Carlo Magno
 
Project m&e & logframe
Project m&e & logframeProject m&e & logframe
Project m&e & logframe
Wesley Opaki
 
Monitoring and evaluation (2)
Monitoring and evaluation (2)Monitoring and evaluation (2)
Monitoring and evaluation (2)
Dr.RAJEEV KASHYAP
 
Difference between monitoring and evaluation
Difference between monitoring and evaluationDifference between monitoring and evaluation
Difference between monitoring and evaluation
Doreen Ty
 
Workshop: Monitoring, evaluation and impact assessment
Workshop: Monitoring, evaluation and impact assessmentWorkshop: Monitoring, evaluation and impact assessment
Workshop: Monitoring, evaluation and impact assessment
WorldFish
 

What's hot (20)

Basics of Monitoring & Evaluation
Basics of Monitoring & EvaluationBasics of Monitoring & Evaluation
Basics of Monitoring & Evaluation
 
Monitoring evaluation
Monitoring evaluationMonitoring evaluation
Monitoring evaluation
 
6 M&E - Monitoring and Evaluation of Aid Projects
6 M&E - Monitoring and Evaluation of Aid Projects6 M&E - Monitoring and Evaluation of Aid Projects
6 M&E - Monitoring and Evaluation of Aid Projects
 
Presentation Training on Result Based Management (RBM) for M&E Staff
Presentation Training on Result Based Management (RBM) for M&E StaffPresentation Training on Result Based Management (RBM) for M&E Staff
Presentation Training on Result Based Management (RBM) for M&E Staff
 
Monitoring indicators
Monitoring indicatorsMonitoring indicators
Monitoring indicators
 
Impact evaluation methods: Qualitative Methods
Impact evaluation methods: Qualitative MethodsImpact evaluation methods: Qualitative Methods
Impact evaluation methods: Qualitative Methods
 
Project m&e & logframe
Project m&e & logframeProject m&e & logframe
Project m&e & logframe
 
Monitoring and evaluation (2)
Monitoring and evaluation (2)Monitoring and evaluation (2)
Monitoring and evaluation (2)
 
Capacity Development For Monitoring And Evaluation
Capacity Development For Monitoring And EvaluationCapacity Development For Monitoring And Evaluation
Capacity Development For Monitoring And Evaluation
 
Difference between monitoring and evaluation
Difference between monitoring and evaluationDifference between monitoring and evaluation
Difference between monitoring and evaluation
 
Introduction to the Logical Framework Approach
Introduction to the Logical Framework ApproachIntroduction to the Logical Framework Approach
Introduction to the Logical Framework Approach
 
Impact Evaluation Overview
Impact Evaluation OverviewImpact Evaluation Overview
Impact Evaluation Overview
 
Monitoring and Evaluation: Lesson 2
Monitoring and Evaluation: Lesson 2Monitoring and Evaluation: Lesson 2
Monitoring and Evaluation: Lesson 2
 
Logical framework
Logical  frameworkLogical  framework
Logical framework
 
Workshop: Monitoring, evaluation and impact assessment
Workshop: Monitoring, evaluation and impact assessmentWorkshop: Monitoring, evaluation and impact assessment
Workshop: Monitoring, evaluation and impact assessment
 
Evaluation Capacity Development
Evaluation Capacity DevelopmentEvaluation Capacity Development
Evaluation Capacity Development
 
Introduction to Stata
Introduction to StataIntroduction to Stata
Introduction to Stata
 
Logical framework
Logical frameworkLogical framework
Logical framework
 
Introduction to monitoring and evaluation
Introduction to monitoring and evaluationIntroduction to monitoring and evaluation
Introduction to monitoring and evaluation
 
Monitoring and Evaluation Framework
Monitoring and Evaluation FrameworkMonitoring and Evaluation Framework
Monitoring and Evaluation Framework
 

Viewers also liked

Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...
Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...
Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...
LIDC
 

Viewers also liked (20)

Measuring Impact Qualitatively
Measuring Impact QualitativelyMeasuring Impact Qualitatively
Measuring Impact Qualitatively
 
An Introduction to the Manual: How Do We Know if a Program Made a Difference?...
An Introduction to the Manual:How Do We Know if a Program Made a Difference?...An Introduction to the Manual:How Do We Know if a Program Made a Difference?...
An Introduction to the Manual: How Do We Know if a Program Made a Difference?...
 
How we can use impact evaluation to assure effective use of resources for dev...
How we can use impact evaluation to assure effective use of resources for dev...How we can use impact evaluation to assure effective use of resources for dev...
How we can use impact evaluation to assure effective use of resources for dev...
 
Project Management Cycle: Tackling Implementation Challenges
Project Management Cycle: Tackling Implementation ChallengesProject Management Cycle: Tackling Implementation Challenges
Project Management Cycle: Tackling Implementation Challenges
 
Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...
Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...
Impact Evaluation, Policy Making and Academic Research: Some Reflections and ...
 
Evaluations of Gender-Integrated Reproductive Health Interventions: A Review ...
Evaluations of Gender-Integrated Reproductive Health Interventions: A Review ...Evaluations of Gender-Integrated Reproductive Health Interventions: A Review ...
Evaluations of Gender-Integrated Reproductive Health Interventions: A Review ...
 
M&E for Social Service System Strengthening
M&E for Social Service System Strengthening M&E for Social Service System Strengthening
M&E for Social Service System Strengthening
 
Using Maps in Decision Making to Strengthen Programs for Orphans and Vulnerab...
Using Maps in Decision Making to Strengthen Programs for Orphans and Vulnerab...Using Maps in Decision Making to Strengthen Programs for Orphans and Vulnerab...
Using Maps in Decision Making to Strengthen Programs for Orphans and Vulnerab...
 
Strengthening National M&E Systems for Orphans and Vulnerable Children Programs
Strengthening National M&E Systems for Orphans and Vulnerable Children ProgramsStrengthening National M&E Systems for Orphans and Vulnerable Children Programs
Strengthening National M&E Systems for Orphans and Vulnerable Children Programs
 
Monitoring and Evaluating Male Engagement in Family Planning Programs
Monitoring and Evaluating Male Engagement in Family Planning ProgramsMonitoring and Evaluating Male Engagement in Family Planning Programs
Monitoring and Evaluating Male Engagement in Family Planning Programs
 
Measuring National M&E System Strengthening in Nigeria: Application of the Mo...
Measuring National M&E System Strengthening in Nigeria: Application of the Mo...Measuring National M&E System Strengthening in Nigeria: Application of the Mo...
Measuring National M&E System Strengthening in Nigeria: Application of the Mo...
 
Implementing the Population Registration System: Progress Towards a Data Revo...
Implementing the Population Registration System: Progress Towards a Data Revo...Implementing the Population Registration System: Progress Towards a Data Revo...
Implementing the Population Registration System: Progress Towards a Data Revo...
 
Digital Data Ethics: Harnessing without Hurting
Digital Data Ethics: Harnessing without HurtingDigital Data Ethics: Harnessing without Hurting
Digital Data Ethics: Harnessing without Hurting
 
Information Use Map
Information Use MapInformation Use Map
Information Use Map
 
Evaluation of the Impact of Malaria Control Interventions on All-Cause Mortal...
Evaluation of the Impact of Malaria Control Interventions on All-Cause Mortal...Evaluation of the Impact of Malaria Control Interventions on All-Cause Mortal...
Evaluation of the Impact of Malaria Control Interventions on All-Cause Mortal...
 
Lessons Learned Collecting Most Significant Change Stories in an Impact Evalu...
Lessons Learned Collecting Most Significant Change Stories in an Impact Evalu...Lessons Learned Collecting Most Significant Change Stories in an Impact Evalu...
Lessons Learned Collecting Most Significant Change Stories in an Impact Evalu...
 
Strengths and Challenges in the Implementation of Women’s Justice and Empower...
Strengths and Challenges in the Implementation of Women’s Justice and Empower...Strengths and Challenges in the Implementation of Women’s Justice and Empower...
Strengths and Challenges in the Implementation of Women’s Justice and Empower...
 
Within Models
Within ModelsWithin Models
Within Models
 
Data Quality Review (DQR) Methods and Tools: Holistic, Country-Led Data Qual...
Data Quality Review (DQR) Methods and Tools: Holistic, Country-Led Data Qual...Data Quality Review (DQR) Methods and Tools: Holistic, Country-Led Data Qual...
Data Quality Review (DQR) Methods and Tools: Holistic, Country-Led Data Qual...
 
Two Examples of Program Planning, Monitoring and Evaluation
Two Examples of Program Planning, Monitoring and EvaluationTwo Examples of Program Planning, Monitoring and Evaluation
Two Examples of Program Planning, Monitoring and Evaluation
 

Similar to Fundamentals of Program Impact Evaluation

Score Week 5 Correlation and RegressionCorrelation and Regres.docx
Score Week 5 Correlation and RegressionCorrelation and Regres.docxScore Week 5 Correlation and RegressionCorrelation and Regres.docx
Score Week 5 Correlation and RegressionCorrelation and Regres.docx
kenjordan97598
 
Research methods 2 operationalization & measurement
Research methods 2   operationalization & measurementResearch methods 2   operationalization & measurement
Research methods 2 operationalization & measurement
attique1960
 

Similar to Fundamentals of Program Impact Evaluation (20)

Selection on Observables
Selection on ObservablesSelection on Observables
Selection on Observables
 
Causality in Python PyCon 2021 ISRAEL
Causality in Python PyCon 2021 ISRAELCausality in Python PyCon 2021 ISRAEL
Causality in Python PyCon 2021 ISRAEL
 
Randomization and Its Discontents
Randomization and Its DiscontentsRandomization and Its Discontents
Randomization and Its Discontents
 
Intro to Quant Trading Strategies (Lecture 9 of 10)
Intro to Quant Trading Strategies (Lecture 9 of 10)Intro to Quant Trading Strategies (Lecture 9 of 10)
Intro to Quant Trading Strategies (Lecture 9 of 10)
 
Lecture 4
Lecture 4Lecture 4
Lecture 4
 
PR 113: The Perception Distortion Tradeoff
PR 113: The Perception Distortion TradeoffPR 113: The Perception Distortion Tradeoff
PR 113: The Perception Distortion Tradeoff
 
Hypothesis testing; z test, t-test. f-test
Hypothesis testing; z test, t-test. f-testHypothesis testing; z test, t-test. f-test
Hypothesis testing; z test, t-test. f-test
 
Statistics (recap)
Statistics (recap)Statistics (recap)
Statistics (recap)
 
Score Week 5 Correlation and RegressionCorrelation and Regres.docx
Score Week 5 Correlation and RegressionCorrelation and Regres.docxScore Week 5 Correlation and RegressionCorrelation and Regres.docx
Score Week 5 Correlation and RegressionCorrelation and Regres.docx
 
Inferential Statistics.pdf
Inferential Statistics.pdfInferential Statistics.pdf
Inferential Statistics.pdf
 
Business Optimization via Causal Inference
Business Optimization via Causal InferenceBusiness Optimization via Causal Inference
Business Optimization via Causal Inference
 
Mentor mix review
Mentor mix reviewMentor mix review
Mentor mix review
 
HYPOTHESIS TESTS.pptx
HYPOTHESIS TESTS.pptxHYPOTHESIS TESTS.pptx
HYPOTHESIS TESTS.pptx
 
Observational studies in social media
Observational studies in social mediaObservational studies in social media
Observational studies in social media
 
Bio-statistics definitions and misconceptions
Bio-statistics definitions and misconceptionsBio-statistics definitions and misconceptions
Bio-statistics definitions and misconceptions
 
Measures of fertility and mortality
Measures of fertility and mortalityMeasures of fertility and mortality
Measures of fertility and mortality
 
[GAN by Hung-yi Lee]Part 2: The application of GAN to speech and text processing
[GAN by Hung-yi Lee]Part 2: The application of GAN to speech and text processing[GAN by Hung-yi Lee]Part 2: The application of GAN to speech and text processing
[GAN by Hung-yi Lee]Part 2: The application of GAN to speech and text processing
 
Introduction of “Fairness in Learning: Classic and Contextual Bandits”
Introduction of “Fairness in Learning: Classic and Contextual Bandits”Introduction of “Fairness in Learning: Classic and Contextual Bandits”
Introduction of “Fairness in Learning: Classic and Contextual Bandits”
 
Statistical inference: Probability and Distribution
Statistical inference: Probability and DistributionStatistical inference: Probability and Distribution
Statistical inference: Probability and Distribution
 
Research methods 2 operationalization & measurement
Research methods 2   operationalization & measurementResearch methods 2   operationalization & measurement
Research methods 2 operationalization & measurement
 

More from MEASURE Evaluation

Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...
Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...
Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...
MEASURE Evaluation
 
Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...
Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...
Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...
MEASURE Evaluation
 

More from MEASURE Evaluation (20)

Managing missing values in routinely reported data: One approach from the Dem...
Managing missing values in routinely reported data: One approach from the Dem...Managing missing values in routinely reported data: One approach from the Dem...
Managing missing values in routinely reported data: One approach from the Dem...
 
Use of Routine Data for Economic Evaluations
Use of Routine Data for Economic EvaluationsUse of Routine Data for Economic Evaluations
Use of Routine Data for Economic Evaluations
 
Routine data use in evaluation: practical guidance
Routine data use in evaluation: practical guidanceRoutine data use in evaluation: practical guidance
Routine data use in evaluation: practical guidance
 
Tuberculosis/HIV Mobility Study: Objectives and Background
Tuberculosis/HIV Mobility Study: Objectives and BackgroundTuberculosis/HIV Mobility Study: Objectives and Background
Tuberculosis/HIV Mobility Study: Objectives and Background
 
How to improve the capabilities of health information systems to address emer...
How to improve the capabilities of health information systems to address emer...How to improve the capabilities of health information systems to address emer...
How to improve the capabilities of health information systems to address emer...
 
LCI Evaluation Uganda Organizational Network Analysis
LCI Evaluation Uganda Organizational Network AnalysisLCI Evaluation Uganda Organizational Network Analysis
LCI Evaluation Uganda Organizational Network Analysis
 
Using Organizational Network Analysis to Plan and Evaluate Global Health Prog...
Using Organizational Network Analysis to Plan and Evaluate Global Health Prog...Using Organizational Network Analysis to Plan and Evaluate Global Health Prog...
Using Organizational Network Analysis to Plan and Evaluate Global Health Prog...
 
Understanding Referral Networks for Adolescent Girls and Young Women
Understanding Referral Networks for Adolescent Girls and Young WomenUnderstanding Referral Networks for Adolescent Girls and Young Women
Understanding Referral Networks for Adolescent Girls and Young Women
 
Data for Impact: Lessons Learned in Using the Ripple Effects Mapping Method
Data for Impact: Lessons Learned in Using the Ripple Effects Mapping MethodData for Impact: Lessons Learned in Using the Ripple Effects Mapping Method
Data for Impact: Lessons Learned in Using the Ripple Effects Mapping Method
 
Local Capacity Initiative (LCI) Evaluation
Local Capacity Initiative (LCI) EvaluationLocal Capacity Initiative (LCI) Evaluation
Local Capacity Initiative (LCI) Evaluation
 
Development and Validation of a Reproductive Empowerment Scale
Development and Validation of a Reproductive Empowerment ScaleDevelopment and Validation of a Reproductive Empowerment Scale
Development and Validation of a Reproductive Empowerment Scale
 
Sustaining the Impact: MEASURE Evaluation Conversation on Maternal and Child ...
Sustaining the Impact: MEASURE Evaluation Conversation on Maternal and Child ...Sustaining the Impact: MEASURE Evaluation Conversation on Maternal and Child ...
Sustaining the Impact: MEASURE Evaluation Conversation on Maternal and Child ...
 
Using Most Significant Change in a Mixed-Methods Evaluation in Uganda
Using Most Significant Change in a Mixed-Methods Evaluation in UgandaUsing Most Significant Change in a Mixed-Methods Evaluation in Uganda
Using Most Significant Change in a Mixed-Methods Evaluation in Uganda
 
Lessons Learned In Using the Most Significant Change Technique in Evaluation
Lessons Learned In Using the Most Significant Change Technique in EvaluationLessons Learned In Using the Most Significant Change Technique in Evaluation
Lessons Learned In Using the Most Significant Change Technique in Evaluation
 
Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...
Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...
Malaria Data Quality and Use in Selected Centers of Excellence in Madagascar:...
 
Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...
Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...
Evaluating National Malaria Programs’ Impact in Moderate- and Low-Transmissio...
 
Improved Performance of the Malaria Surveillance, Monitoring, and Evaluation ...
Improved Performance of the Malaria Surveillance, Monitoring, and Evaluation ...Improved Performance of the Malaria Surveillance, Monitoring, and Evaluation ...
Improved Performance of the Malaria Surveillance, Monitoring, and Evaluation ...
 
Lessons learned in using process tracing for evaluation
Lessons learned in using process tracing for evaluationLessons learned in using process tracing for evaluation
Lessons learned in using process tracing for evaluation
 
Use of Qualitative Comparative Analysis in the Assessment of the Actionable D...
Use of Qualitative Comparative Analysis in the Assessment of the Actionable D...Use of Qualitative Comparative Analysis in the Assessment of the Actionable D...
Use of Qualitative Comparative Analysis in the Assessment of the Actionable D...
 
Sustaining the Impact: MEASURE Evaluation Conversation on Health Informatics
Sustaining the Impact: MEASURE Evaluation Conversation on Health InformaticsSustaining the Impact: MEASURE Evaluation Conversation on Health Informatics
Sustaining the Impact: MEASURE Evaluation Conversation on Health Informatics
 

Recently uploaded

1029-Danh muc Sach Giao Khoa khoi 6.pdf
1029-Danh muc Sach Giao Khoa khoi  6.pdf1029-Danh muc Sach Giao Khoa khoi  6.pdf
1029-Danh muc Sach Giao Khoa khoi 6.pdf
QucHHunhnh
 
The basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptxThe basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptx
heathfieldcps1
 
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in DelhiRussian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
kauryashika82
 
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
ZurliaSoop
 
Spellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please PractiseSpellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please Practise
AnaAcapella
 

Recently uploaded (20)

Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
Explore beautiful and ugly buildings. Mathematics helps us create beautiful d...
 
On National Teacher Day, meet the 2024-25 Kenan Fellows
On National Teacher Day, meet the 2024-25 Kenan FellowsOn National Teacher Day, meet the 2024-25 Kenan Fellows
On National Teacher Day, meet the 2024-25 Kenan Fellows
 
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
TỔNG ÔN TẬP THI VÀO LỚP 10 MÔN TIẾNG ANH NĂM HỌC 2023 - 2024 CÓ ĐÁP ÁN (NGỮ Â...
 
1029-Danh muc Sach Giao Khoa khoi 6.pdf
1029-Danh muc Sach Giao Khoa khoi  6.pdf1029-Danh muc Sach Giao Khoa khoi  6.pdf
1029-Danh muc Sach Giao Khoa khoi 6.pdf
 
Food safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfFood safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdf
 
Dyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptxDyslexia AI Workshop for Slideshare.pptx
Dyslexia AI Workshop for Slideshare.pptx
 
Sociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning ExhibitSociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning Exhibit
 
This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.This PowerPoint helps students to consider the concept of infinity.
This PowerPoint helps students to consider the concept of infinity.
 
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdfUGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
UGC NET Paper 1 Mathematical Reasoning & Aptitude.pdf
 
Micro-Scholarship, What it is, How can it help me.pdf
Micro-Scholarship, What it is, How can it help me.pdfMicro-Scholarship, What it is, How can it help me.pdf
Micro-Scholarship, What it is, How can it help me.pdf
 
The basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptxThe basics of sentences session 3pptx.pptx
The basics of sentences session 3pptx.pptx
 
ICT role in 21st century education and it's challenges.
ICT role in 21st century education and it's challenges.ICT role in 21st century education and it's challenges.
ICT role in 21st century education and it's challenges.
 
psychiatric nursing HISTORY COLLECTION .docx
psychiatric  nursing HISTORY  COLLECTION  .docxpsychiatric  nursing HISTORY  COLLECTION  .docx
psychiatric nursing HISTORY COLLECTION .docx
 
Magic bus Group work1and 2 (Team 3).pptx
Magic bus Group work1and 2 (Team 3).pptxMagic bus Group work1and 2 (Team 3).pptx
Magic bus Group work1and 2 (Team 3).pptx
 
ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701ComPTIA Overview | Comptia Security+ Book SY0-701
ComPTIA Overview | Comptia Security+ Book SY0-701
 
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in DelhiRussian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
Russian Escort Service in Delhi 11k Hotel Foreigner Russian Call Girls in Delhi
 
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
 
Spellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please PractiseSpellings Wk 3 English CAPS CARES Please Practise
Spellings Wk 3 English CAPS CARES Please Practise
 
Understanding Accommodations and Modifications
Understanding  Accommodations and ModificationsUnderstanding  Accommodations and Modifications
Understanding Accommodations and Modifications
 

Fundamentals of Program Impact Evaluation

  • 1. Peter M. Lance, PhD MEASURE Evaluation University of North Carolina at Chapel Hill MARCH 31, 2016 Fundamentals of Program Impact Evaluation
  • 2. Global, five-year, $180M cooperative agreement Strategic objective: To strengthen health information systems – the capacity to gather, interpret, and use data – so countries can make better decisions and sustain good health outcomes over time. Project overview
  • 3. Improved country capacity to manage health information systems, resources, and staff Strengthened collection, analysis, and use of routine health data Methods, tools, and approaches improved and applied to address health information challenges and gaps Increased capacity for rigorous evaluation Phase IV Results Framework
  • 4. Global footprint (more than 25 countries)
  • 5.
  • 6. How Do We Know IfAProgram MadeADifference? ABrief Helicopter Tour of Methods for Estimating Program Impact
  • 7. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 8. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 9. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 10. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 11. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 12. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 13. • The Program Impact Evaluation Challenge • Randomization • Selection on observables • Within estimators • Instrumental variables
  • 14. Newton’s “Laws” of Motion 𝐹𝑜𝑟𝑐𝑒 = 𝑀𝑎𝑠𝑠 ∙ 𝐴𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛
  • 15.
  • 16.
  • 17. Did the program make a difference?
  • 18. Did the program cause a change in an outcome of interest Y ? (Causality)
  • 19. Our outcome of Interest What happens if an individual does not participate in a program What happens if that individual does participate in a program Potential Outcomes 𝑌 : 𝑌0 : 𝑌1 :
  • 20. Our outcome of interest What happens if an individual does not participate in a program What happens if that individual does participate in a program Potential Outcomes 𝑌𝑖 : 𝑌𝑖 0 : 𝑌𝑖 1 :
  • 21. Our outcome of interest What happens if an individual does not participate in a program What happens if that individual does participate in a program Potential Outcomes 𝑌𝑖 : 𝑌𝑖 0 : 𝑌𝑖 1 :
  • 22. Our outcome of interest What happens if an individual does not participate in a program What happens if that individual does participate in a program Potential Outcomes 𝑌𝑖 : 𝑌𝑖 0 : 𝑌𝑖 1 :
  • 23. What happens if the individual participates {Causal} Program Impact 𝑌𝑖 1 − 𝑌𝑖 0 = Program Impact What happens if the individual does not participate
  • 24. What happens if the individual participates {Causal} Program Impact 𝑌𝑖 1 − 𝑌𝑖 0 = Program Impact What happens if the individual does not participate
  • 25. What happens if the individual participates {Causal} Program Impact 𝑌𝑖 1 − 𝑌𝑖 0 = Program Impact What happens if the individual does not participate
  • 26. What happens if the individual participates {Causal} Program Impact 𝑌𝑖 1 − 𝑌𝑖 0 = Program Impact What happens if the individual does not participate
  • 27. What happens if the individual participates {Causal} Program Impact 𝑌𝑖 1 − 𝑌𝑖 0 = Program Impact What happens if the individual does not participate
  • 28. 𝑃𝑖 = 1ifindividual 𝑖 participates 0if individual 𝑖 does not participate Program Participation
  • 29. 𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖 1 + 1 − 𝑃𝑖 ∙ 𝑌𝑖 0 Observed Outcome
  • 30. 𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖 1 + 1 − 𝑃𝑖 ∙ 𝑌𝑖 0 Observed Outcome 𝑃𝑖 = 1
  • 31. 𝑌𝑖 = 1 ∙ 𝑌𝑖 1 + 1 − 1 ∙ 𝑌𝑖 0 Observed Outcome 𝑃𝑖 = 1
  • 32. 𝑌𝑖 = 𝑌𝑖 1 + 0 ∙ 𝑌𝑖 0 Observed Outcome 𝑃𝑖 = 1
  • 33. 𝑌𝑖 = 𝑌𝑖 1 Observed Outcome 𝑃𝑖 = 1
  • 34. 𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖 1 + 1 − 𝑃𝑖 ∙ 𝑌𝑖 0 Observed Outcome
  • 35. 𝑌𝑖 = 𝑃𝑖 ∙ 𝑌𝑖 1 + 1 − 𝑃𝑖 ∙ 𝑌𝑖 0 Observed Outcome 𝑃𝑖 = 0
  • 36. 𝑌𝑖 = 0 ∙ 𝑌𝑖 1 + 1 − 0 ∙ 𝑌𝑖 0 Observed Outcome 𝑃𝑖 = 0
  • 37. 𝑌𝑖 = 𝑌𝑖 0 Observed Outcome 𝑃𝑖 = 0
  • 42. 𝑌𝑖 1 , 𝑌𝑖 0 Observed Outcome Fundamental Identification Problem of Program Impact Evaluation
  • 43. 𝑌𝑖 1 , 𝑌𝑖 0 Observed Outcome Fundamental Identification Problem of Program Impact Evaluation
  • 45. Individual Population Hi. They call me individual i
  • 47.
  • 50. An expected value for a random variable is the average value from a large number of repetitions of the experiment that random variable represents An expected value is the true average of a random variable across a population Expected Value
  • 51. An expected value for a random variable is the average value from a large number of repetitions of the experiment that random variable represents An expected value is the true average of a random variable across a population Expected Value
  • 52. An expected value is the true average of a random variable across a population 𝐸 𝑋 = sometruevalue Expected Value
  • 53. 𝐸 𝑐 = 𝑐 𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊 𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍 𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍 𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍 Expectations: Properties
  • 54. 𝑬 𝒄 = 𝒄 𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊 𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍 𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍 𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍 Expectations: Properties
  • 55. 𝐸 𝑐 = 𝑐 𝑬 𝒄 ∙ 𝑾 = 𝒄 ∙ 𝑬 𝑾 𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍 𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍 𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍 Expectations: Properties
  • 56. 𝐸 𝑐 = 𝑐 𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊 𝑬 𝑾 + 𝒁 = 𝑬 𝑾 + 𝑬 𝒁 𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍 𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍 Expectations: Properties
  • 57. 𝐸 𝑐 = 𝑐 𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊 𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍 𝑬 𝑾 − 𝒁 = 𝑬 𝑾 − 𝑬 𝒁 𝐸 𝑎 ∙ 𝑊 ± 𝑏 ∙ 𝑍 = 𝑎 ∙ 𝐸 𝑊 ± 𝑏 ∙ 𝐸 𝑍 Expectations: Properties
  • 58. 𝐸 𝑐 = 𝑐 𝐸 𝑐 ∙ 𝑊 = 𝑐 ∙ 𝐸 𝑊 𝐸 𝑊 + 𝑍 = 𝐸 𝑊 + 𝐸 𝑍 𝐸 𝑊 − 𝑍 = 𝐸 𝑊 − 𝐸 𝑍 𝑬 𝒂 ∙ 𝑾 ± 𝒃 ∙ 𝒁 = 𝒂 ∙ 𝑬 𝑾 ± 𝒃 ∙ 𝑬 𝒁 Expectations: Properties
  • 59. 𝐸 𝑊 ∙ 𝑍 ≠ 𝐸 𝑊 ∙ 𝐸 𝑍 𝐸 𝑊 𝑍 ≠ 𝐸 𝑊 𝐸 𝑍 𝐸 𝑓 𝑊 ≠ 𝑓 𝐸 𝑊 Expectations: Properties
  • 60. 𝑬 𝑾 ∙ 𝒁 ≠ 𝑬 𝑾 ∙ 𝑬 𝒁 𝐸 𝑊 𝑍 ≠ 𝐸 𝑊 𝐸 𝑍 𝐸 𝑓 𝑊 ≠ 𝑓 𝐸 𝑊 Expectations: Properties
  • 61. 𝐸 𝑊 ∙ 𝑍 ≠ 𝐸 𝑊 ∙ 𝐸 𝑍 𝑬 𝑾 𝒁 ≠ 𝑬 𝑾 𝑬 𝒁 𝐸 𝑓 𝑊 ≠ 𝑓 𝐸 𝑊 Expectations: Properties
  • 62. 𝐸 𝑊 ∙ 𝑍 ≠ 𝐸 𝑊 ∙ 𝐸 𝑍 𝐸 𝑊 𝑍 ≠ 𝐸 𝑊 𝐸 𝑍 𝑬 𝒇 𝑾 ≠ 𝒇 𝑬 𝑾 Expectations: Properties
  • 63. 𝑌𝑖 1 − 𝑌𝑖 0 Average Treatment Effect (ATE) 𝐸 𝑌1 − 𝑌0 Average Effect of Treatment on the Treated (ATT) 𝐸 𝑌1 − 𝑌0|𝑃 = 1 Hi there Individual Impact
  • 66. Average Treatment Effect (ATE) 𝐸 𝑌1 − 𝑌0 Average Effect of Treatment on the Treated (ATT) 𝐸 𝑌1 − 𝑌0|𝑃 = 1 Treatment Effects
  • 67.
  • 68.
  • 69. Suppose that we have a sample of 𝑖 = 1,…, 𝑛 individuals…. …but for each individual 𝑖 we observe either 𝑌𝑖 1 or 𝑌𝑖 0 … …but not both So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 70. Suppose that we have a sample of 𝑖 = 1,…, 𝑛 individuals…. …but for each individual 𝑖 we observe either 𝑌𝑖 1 or 𝑌𝑖 0 … …but not both So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 71. Remember, however, a key property of expectations: 𝐸 𝑌1 − 𝑌0 = 𝐸 𝑌1 − 𝐸 𝑌0 …but this means that in principle we could estimate E 𝑌1 and E 𝑌0 separately So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 72. Remember, however, a key property of expectations: 𝐸 𝑌1 − 𝑌0 = 𝐸 𝑌1 − 𝐸 𝑌0 …but this means that in principle we could estimate E 𝑌1 and E 𝑌0 separately So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 73. For instance, suppose that in our sample we have: 𝑛 𝑃 participants(𝑃𝑖 = 1) and 𝑛 𝑁 non-participants(𝑃𝑖 = 0) (hence 𝑛 𝑃 + 𝑛 𝑁 = 𝑛) So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 74. Then an estimator of 𝐸 𝑌1 is 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 calculated with the 𝑛 𝑃 participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 75. Then an estimator of 𝐸 𝑌1 is 𝒀 𝟏 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 calculated with the 𝑛 𝑃 participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 76. Then an estimator of 𝐸 𝑌1 is 𝑌1 = 𝑗=1 𝒏 𝑷 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 calculated with the 𝑛 𝑃 participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 77. Then an estimator of 𝐸 𝑌1 is 𝑌1 = 𝑗=1 𝑛 𝑃 𝒀𝒋 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 calculated with the 𝑛 𝑃 participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 78. Then an estimator of 𝐸 𝑌1 is 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝒏 𝑷 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 calculated with the 𝑛 𝑃 participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 79. Then an estimator of 𝐸 𝑌1 is 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝒀𝒋 𝟏 𝑛 𝑃 calculated with the 𝑛 𝑃 participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 80. Similarly, an estimator of 𝐸 𝑌0 is 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁 calculated with the 𝑛 𝑁 non-participants out of the sample of 𝑛 individuals So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 81. So then an estimate of 𝐸 𝑌1 − 𝑌0 = 𝐸 𝑌1 − 𝐸 𝑌0 is 𝑌1 − 𝑌0 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 − 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁 − 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 So how do we estimate 𝑬 𝒀 𝟏 − 𝒀 𝟎 ??
  • 82. But is it a good estimate??
  • 83.
  • 84.
  • 85.
  • 86. So we have two samples of size 𝒏 By random chance, between the two samples we almost surely have 1. A different precise mix of individuals 2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁) 3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 : 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁
  • 87. So we have two samples of size 𝒏 By random chance, between the two samples we almost surely have 1. A different precise mix of individuals 2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁) 3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 : 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁
  • 88. So we have two samples of size 𝒏 By random chance, between the two samples we almost surely have 1. A different precise mix of individuals 2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁) 3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 : 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁
  • 89. So we have two samples of size 𝒏 By random chance, between the two samples we almost surely have 1. A different precise mix of individuals 2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁) 3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 : 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁
  • 90. So we have two samples of size 𝒏 By random chance, between the two samples we almost surely have 1. A different precise mix of individuals 2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁) 3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 : 𝑌1 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 = 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁
  • 91. So we have two samples of size 𝒏 By random chance, between the two samples we almost surely have 1. A different precise mix of individuals 2. A different number of participants (𝑛 𝑃) and non-participants (𝑛 𝑁) 3. Different estimates 𝑌1 and 𝑌0 of 𝐸 𝑌1 and 𝐸 𝑌0 : 𝒀 𝟏 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝒏 𝑷 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑌0 = 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁 𝒀 𝟏 𝑬 𝒀 𝟏
  • 92.
  • 93.
  • 94.
  • 95.
  • 96.
  • 97.
  • 98.
  • 99.
  • 100.
  • 101.
  • 102. 𝒀 𝟏 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝒏 𝑷 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷
  • 103. 𝒀 𝟏 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝒏 𝑷 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷
  • 104.
  • 105.
  • 106.
  • 107.
  • 108.
  • 109. 𝒀 𝟏 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝒏 𝑷 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
  • 110. 𝒀 𝟏 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝒏 𝑷 = 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏
  • 111.
  • 112.
  • 113. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀 𝟏 = 𝒏 𝑷 ∙ 𝑬 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿 2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁 𝑬 𝒊=𝟏 𝒏 𝑷 𝑿𝒊 = 𝒊=𝟏 𝒏 𝑷 𝑬 𝑿𝒊
  • 114. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀 𝟏 = 𝒏 𝑷 ∙ 𝑬 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿 2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁 𝑬 𝒊=𝟏 𝒏 𝑷 𝑿𝒊 = 𝒊=𝟏 𝒏 𝑷 𝑬 𝑿𝒊
  • 115. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿 2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁 𝑬 𝒊=𝟏 𝒏 𝑷 𝑿𝒊 = 𝒊=𝟏 𝒏 𝑷 𝑬 𝑿𝒊
  • 116. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿 2nd Rule: 𝑬 𝑿 + 𝒁 = 𝑬 𝑿 + 𝑬 𝒁 𝑬 𝒊=𝟏 𝒏 𝑷 𝑿𝒊 = 𝒊=𝟏 𝒏 𝑷 𝑬 𝑿𝒊
  • 117. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿
  • 118. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋 𝟏 1ST Rule: 𝑬 𝒄 ∙ 𝑿 = 𝒄 ∙ 𝑬 𝑿 𝑬 𝟏 𝒏 𝑷 ∙ 𝑿 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝑿
  • 119. 𝑬 𝒀 𝟏 = 𝑬 𝒋=𝟏 𝒏 𝑷 𝒀𝒋 𝟏 𝒏 𝑷 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝑬 𝒋=𝟏 𝒏 𝑷 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝟏 𝒏 𝑷 ∙ 𝒏 𝑷 ∙ 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏
  • 120. 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀𝒋 𝟏 = 𝑬 𝒀 𝟏
  • 121. 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀𝒋 𝟏 = 𝑬 𝒀 𝟏
  • 122. 𝑬 𝒀 𝟏 = 𝑬 𝒀𝒋 𝟏 𝑬 𝒀 𝟏 = 𝑬 𝒀 𝟏 𝑬 𝒀𝒋 𝟏 = 𝑬 𝒀 𝟏
  • 124.
  • 125.
  • 127. 𝑃 = 0 𝑃 = 0 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝑃 = 0 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝒀 𝟏
  • 128. 𝑃 = 0 𝑃 = 0 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 0 𝑃 = 1 𝑃 = 1 𝑃 = 0 𝑃 = 0 𝑃 = 1 𝑃 = 0 𝒀 𝟏
  • 129. 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝒀 𝟏
  • 130. Z W “Z Causes W” 𝑬 𝑾|𝒁 ≠ 𝑬 𝑾
  • 131. Z W “Z causes W” 𝑬 𝑾|𝒁 ≠ 𝑬 𝑾
  • 132. Z W “Z causes W” 𝑬 𝑾|𝒁 ≠ 𝑬 𝑾
  • 133. X Y1
  • 134. X Y P
  • 136. X Y P
  • 137. X Y P
  • 138. 𝐸 𝑋|𝑃 = 1 ≠ 𝐸 𝑋 𝐸 𝑌1 |𝑃 = 1 ≠ 𝐸 𝑌1 X Y1
  • 139. 𝐸 𝑋|𝑃 = 1 ≠ 𝐸 𝑋 𝐸 𝑌1 |𝑃 = 1 ≠ 𝐸 𝑌1 X Y1
  • 140. 𝐸 𝑋|𝑃 = 1 ≠ 𝐸 𝑋 𝐸 𝑌1 |𝑃 = 1 ≠ 𝐸 𝑌1 X Y1
  • 141. X Y P
  • 142. 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1 𝑃 = 1
  • 143.
  • 144. 𝐸 𝑌0 |𝑃 = 0 ≠ 𝐸 𝑌0 𝐸 𝑌0 |𝑃 = 1 ≠ 𝐸 𝑌0 𝐸 𝑌0 |𝑃 = 0 ≠ 𝐸 𝑌0 |𝑃 = 0 𝐸 𝑌1 |𝑃 = 0 ≠ 𝐸 𝑌1 𝐸 𝑌1 |𝑃 = 1 ≠ 𝐸 𝑌1 𝐸 𝑌1 |𝑃 = 0 ≠ 𝐸 𝑌1 |𝑃 = 0
  • 145. 𝐸 𝑌0 |𝑃 = 0 ≠ 𝐸 𝑌0 𝐸 𝑌0 |𝑃 = 1 ≠ 𝐸 𝑌0 𝐸 𝑌0 |𝑃 = 0 ≠ 𝐸 𝑌0 |𝑃 = 0 𝐸 𝑌1 |𝑃 = 0 ≠ 𝐸 𝑌1 𝐸 𝑌1 |𝑃 = 1 ≠ 𝐸 𝑌1 𝐸 𝑌1 |𝑃 = 0 ≠ 𝐸 𝑌1 |𝑃 = 0
  • 146. The estimator 𝑌1 − 𝑌0 = 𝑗=1 𝑛 𝑃 𝑌𝑗 𝑛 𝑃 − 𝑘=1 𝑛 𝑁 𝑌𝑘 𝑛 𝑁 = 𝑘=1 𝑛 𝑁 𝑌𝑘 0 𝑛 𝑁 − 𝑗=1 𝑛 𝑃 𝑌𝑗 1 𝑛 𝑃 of 𝐸 𝑌1 − 𝑌0 would be biased if some individuals occurred only among participants or non-participants Or more often among one of the two groups
  • 147. X Y P
  • 148. X Y P
  • 150. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 151. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 152. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 153. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 154. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 155. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 156. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 157. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 158. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 159. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 160. Strength: How strong is the relationship? Consistency: How consistently is link found? Specificity: How specific is the setting or disease? Temporality: Does the cause precede the effect? Gradient: Does more cause lead to more effect? Analogy: Do similar “causes” have similar effect? Coherence: Are field and laboratory findings similar? Experiment: Was variation in the cause random? Plausibility: Does theory agree? Bradford Hill Criteria
  • 161. We are presented with data in the form of a sample: Causality: Our Approach 𝒀𝒊, 𝑷𝒊, 𝑿𝒊 , 𝒊 = 𝟏, . . , 𝒏
  • 162. We are presented with data in the form of a sample: Causality: Our Approach 𝒀𝒊, 𝑷𝒊, 𝑿𝒊 , 𝒊 = 𝟏, . . , 𝒏 Assumptions Model E(Y1-Y0), E(Y1-Y0|P=1), Etc.
  • 163. We are presented with data in the form of a sample: Causality: Our Approach 𝒀𝒊, 𝑷𝒊, 𝑿𝒊 , 𝒊 = 𝟏, . . , 𝒏 Assumptions Model E(Y1-Y0), E(Y1-Y0|P=1), Etc.
  • 165. Links: The manual: http://www.measureevaluation.org/resources/publications/ms- 14-87-en The webinar introducing the manual: http://www.measureevaluation.org/resources/webinars/metho ds-for-program-impact-evaluation My email: pmlance@email.unc.edu
  • 166. MEASURE Evaluation is funded by the U.S. Agency for International Development (USAID) under terms of Cooperative Agreement AID-OAA-L-14-00004 and implemented by the Carolina Population Center, University of North Carolina at Chapel Hill in partnership with ICF International, John Snow, Inc., Management Sciences for Health, Palladium Group, and Tulane University. The views expressed in this presentation do not necessarily reflect the views of USAID or the United States government. www.measureevaluation.org