How to calibrate protection vulnerability scoring?
1
Introduction
2
Executive summary: what you need to know as a manager
2.1
There’s no simple solution to a complex problem
2.2
Scoring with a protection lens
2.3
The reliability of your scores will reflect the quality of your data
2.4
Be ready for auditing
3
Why and when using vulnerability scoring?
3.1
from the identification of vulnerables to the scoring of household vulnerability
3.2
Different models of vulnerability: from food insecurity & poverty to a “protection vulnerability framework”
4
Potential approaches to calibrate scores
4.1
Vulnerability as a latent variable
4.2
Avoid compensability effects
4.3
What data can be used?
4.3.1
Calibration using expert opinions
4.3.2
Calibration using statistically representative dataset
5
Risks when calibrating vulnerability scores
5.1
There’s no magic formula
5.2
Building consensus
5.3
Including and communication with communities
6
Budget allocation
6.1
Selecting experts
6.2
Speed-up consultation with quadratic voting
6.3
Restitution of results
7
Conjoint analysis
7.1
Principle of opinion reverse engineering
7.2
Cutomise the indicators
7.3
Setting up the consultation
8
Analytic Hierarchy Process
8.1
Defining weights through expert judgment
8.2
The Analytic Hierarchy Process (AHP)
8.3
Limitation of AHP
8.4
How to? A step by step approach
8.5
Define vulnerability criteria
8.5.1
Criteria without hierarchy
8.5.2
Criteria with hierarchy
8.6
Build the expert consultation form
8.7
Form
8.8
3. Generate the AHP file from the collected data
8.9
4. Run AHP algorithm
8.10
5. Review results
8.11
6. Apply the formula
9
Vulnerability proxy Regression
9.1
10
Data Envelopment and Deprivation
10.1
What is deprivation analysis?
11
Item Response Theory
11.1
Vulnerability as a “
latent
” variable
11.2
Building composite indicators with IRT
11.2.1
Step 0: Data Preparation
11.2.2
Step 1: Clustering algorithm / Dimensionality assessment
11.2.3
Step 2: Model selection
11.2.4
Step 3: Items aggregation using the weights
12
Robustness and Sensitivity analysis
12.1
Assessing exclusion & Inclusion errors
13
Eligibility and Prioritization
13.1
Eligibility
14
Conclusion
14.1
Source file
How to calibrate protection vulnerability scoring?
9
Vulnerability proxy Regression
9.1