Download - Quasi Experimental Methods I
AADAPT Workshop South AsiaGoa, December 17-21, 2009
Non-Experimental Methods
Quasi Experimental Methods I
Florence Kondylis
What we know so far
Aim: We want to isolate the causal effect of our interventions on our outcomes of interest Use rigorous evaluation methods to answer our
operational questions Randomizing the assignment to treatment is
the “gold standard” methodology (simple, precise, cheap)
What if we really, really (really??) cannot use it?!
>> Where it makes sense, resort to non-experimental methods
4
When does it make sense? Can we find a plausible counterfactual?
Natural experiment? Every non-experimental method is
associated with a set of assumptions The stronger the assumptions, the more
doubtful our measure of the causal effect Question our assumptions
▪ Reality check, resort to common sense!
5
Principal Objective▪ Increase maize production
Intervention▪ Fertilizer vouchers distribution▪ Non-random assignment
Target group▪ Maize producers, land over 1 Ha &
under 5 Ha Main result indicator
▪ Maize yield
Example: Fertilizer Voucher Program
Before After0
2
4
6
8
10
12
14Control GroupTreatment Group
6
(+) Impact of the program
(+) Impact of external factors
Illustration: Fertilizer Voucher Program (1)
Before After0
2
4
6
8
10
12
14Control GroupTreatment Group
7
(+) BIASED Measure of the program impact
Illustration: Fertilizer Voucher Program (2)
“Before-After” doesn’t deliver results we can believe in!
Before After0
2
4
6
8
10
12
14Comparison GroupTreatment Group
8
« After » difference btwnparticipants andnon-participants
Illustration: Fertilizer Voucher Program (3)
« Before» difference btwnparticipants and nonparticipants
>> What’s the impact of our intervention?
9
Difference-in-Differences Identification Strategy (1)Counterfactual: 2 Formulations that say the same thing1. Non-participants’ maize yield after the
intervention, accounting for the “before” difference between participants/nonparticipants (the initial gap between groups)
2. Participants’ maize yield before the intervention, accounting for the “before/after” difference for nonparticipants (the influence of external factors)
1 and 2 are equivalent
Difference-in-DifferencesIdentification Strategy (2)Underlying assumption:Without the intervention, maize yield for participants and non participants’ would have followed the same trend
>> Graphic intuition coming…
11
Data -- Example 1
Average maize yield
(T / Ha)2007 2008 Difference
(2007-2008)
Participants (P) 1.3 1.9 0.6Non-participants (NP)
0.6 1.4 0.8
Difference (P-NP) 0.7 0.5 -0.2
12
Data -- Example 1
Average maize yield
(T / Ha)2007 2008 Difference
(2007-2008)
Participants (P) 1.3 1.9 0.6Non-participants (NP)
0.6 1.4 0.8
Difference (P-NP) 0.7 0.5 -0.2
13
NP2008-NP2007=0.8
Impact = (P2008-P2007) -(NP2008-NP2007)
= 0.6 – 0.8 = -0.2
2007 200800.20.40.60.8
11.21.41.61.8
2
Participants Non-Participants
P2008-P2007=0.6
14
2007 200800.20.40.60.8
11.21.41.61.8
2
Participants Non-Participants
P-NP2008=0.5
Impact = (P-NP)2008-(P-NP)2007= 0.5 -
0.7 = -0.2
P-NP2007=0.7
Assumption of same trend: Graphic Implication
2007 200800.20.40.60.8
11.21.41.61.8
2
Participants Non-Participants
Impact=-0.2
Summary Negative Impact:
Very counter-intuitive: Increased input use should not decrease yield once external factors are accounted for!
Assumption of same trend very strong 2 groups were, in 2007, producing at very different
levels➤ Question the underlying assumption of same
trend!➤When possible, test assumption of same trend with
data from previous years
2006 2007 20080
0.5
1
1.5
2
2.5
participantsnon-participants
Questioning the Assumption of same trend: Use pre-pr0gram data
>> Reject counterfactual assumption of same trends !
18
Data – Example 2
Average maize yield
(T / Ha)2007 2008 Difference
(2007-2008)
Participants (P) 1.5 2.1 0.6Non-participants (NP)
0.5 0.7 0.2
Difference (P-NP) 1.0 1.4 0.4
192007 20080
0.5
1
1.5
2
2.5
participantsnon-participants
P08-P07=0.6
NP08-NP07=0.2
Impact = (P2008-P2007) -(NP2008-NP2007)
= 0.6 – 0.2 = + 0.4
Assumption of same trend: Graphic Implication
2007 20080
0.5
1
1.5
2
2.5
participantsnon-participants
Impact = +0.4
Conclusion
Positive Impact: More intuitive
Is the assumption of same trend reasonable?
➤ Still need to question the counterfactual assumption of same trends !➤Use data from previous years
Questioning the Assumption of same trend: Use pre-pr0gram data
>>Seems reasonable to accept counterfactual assumption of same trend ?!
2006 2007 20080
0.5
1
1.5
2
2.5
participantsnon-participants
Caveats (1)
Assuming same trend is often problematic No data to test the assumption Even if trends are similar the previous
year…▪ Where they always similar (or are we lucky)?▪ More importantly, will they always be similar?
▪ Example: Other project intervenes in our nonparticipant villages…
Caveats (2) What to do?
>> Be descriptive! Check similarity in observable characteristics
▪ If not similar along observables, chances are trends will differ in unpredictable ways
>> Still, we cannot check what we cannot see… And unobservable characteristics might matter more than observable (ability, motivation, patience, etc)
Matching Method + Difference-in-Differences (1)Match participants with non-participants on the basis of
observable characteristicsCounterfactual: Matched comparison group
Each program participant is paired with one or more similar non-participant(s) based on observable characteristics
>> On average, participants and nonparticipants share the same observable characteristics (by construction)
Estimate the effect of our intervention by using difference-in-differences
25
Matching Method (2)
Underlying counterfactual assumptions
After matching, there are no differences between participants and nonparticipants in terms of unobservable characteristics
AND/OR Unobservable characteristics do not affect
the assignment to the treatment, nor the outcomes of interest
How do we do it?
Design a control group by establishing close matches in terms of observable characteristics Carefully select variables along which to
match participants to their control group So that we only retain
▪ Treatment Group: Participants that could find a match
▪ Comparison Group: Non-participants similar enough to the participants
>> We trim out a portion of our treatment group!
Implications
In most cases, we cannot match everyone Need to understand who is left out
Example
Score
NonparticipantsParticipants
MatchedIndividuals
Wealth
Portion of treatmentgroup trimmed out
29
Conclusion (1)
Advantage of the matching method Does not require randomization
30
Conclusion (2)
Disadvantages: Underlying counterfactual assumption is
not plausible in all contexts, hard to test▪ Use common sense, be descriptive
Requires very high quality data: ▪ Need to control for all factors that influence
program placement/outcome of choice Requires significantly large sample size to
generate comparison group Cannot always match everyone…
31
Summary Randomized-Controlled-Trials require
minimal assumptions and procure intuitive estimates (sample means!)
Non-experimental methods require assumptions that must be carefully tested
More data-intensive Not always testable
Get creative: Mix-and-match types of methods! Adress relevant questions with relevant
techniques
Thank you
Financial support from
Is gratefully acknowledged