Monitoring and Evaluation in
Fragile and Conflict-Affected Contexts:
The challenges of measurement
David Fleming, Senior Consultant, Itad
Date: 28th January 2015
Seminar Outline
1. Introducing Itad: Life as an M&E consultant
2. Introducing/recapping M&E: Why monitor
and evaluate and why important in FCAS?
3. Theories of change: what they are, why they
are useful and challenges in FCAS
4. M&E approaches and methods: how to
monitor and evaluate in FCAS; examples from
peacebuilding and humanitarian work
Learning objectives
1. Come away with a better understanding of why we
do M&E and why it’s particularly important in FCAS
2. Learn about and put into practice some of the most
important M&E methods and tools for FCAS
3. Be able to better identify the challenges of doing
M&E in FCAS and how to overcome these
4. Everyone to leave the room with a burning desire to
get involved in M&E at some point in the future!
Life as an M&E consultant…
2. Introducing M&E:
Why Monitor and Evaluate?
“After decades in which development agencies have
disbursed billions of dollars for social programs, and
developing country governments and nongovernmental
organizations (NGOs) have spent hundreds of billions
more, it is deeply disappointing to recognize that we
know relatively little about the net impact of most of
these social programs”
‘When will we ever learn?’ Evaluation Gap Working
Group, Center for Global Development 2006
• Monitoring: “Collection of data with which managers can
assess extent to which objectives are being achieved” (World
Bank)
– Purpose: Collect information on programme outputs and
outcomes to track and improve performance and results
• Evaluation: “Determination of the value of a project,
programme or policy” (World Bank)
– Purpose: evidence-based decisions, accountability,
transparency, lesson learning
– Types: project, programme, policy, organisation, sector, theme,
formative, summative, impact…
Why is M&E important in FCAS?
1. Development trends in FCAS
• By 2015, 50% of world’s poor will live in fragile states
(OECD); by 2030 it might be two thirds (Brookings)
• Support to conflict, violence and fragility becoming a key
priority for most major donors
• ODA to fragile states is falling in quantity but number of
actors multiplying (OECD)
• DFID has been scaling up support to FCAS (commitment
to increase to 30% of ODA by 2015)
• DFID strategies include BSOS, cross-Whitehall CSSF, and
the ‘Beyond Aid’ agenda
Why is M&E important in FCAS?
2. Increasing emphasis on transparency,
accountability and fiduciary risk
• Higher risk to investments in terms of results,
security and fiduciary risk
3. More limited evidence base – need for lesson
learning and evidence of what works
• Support evidence-informed decisions and better
programming by knowing what works and
doesn’t and why and in which contexts
What are the biggest challenges?
Risk of
exacerbating
conflict
Hawthorne
effect
Insecurity
Political
objectives
Longer-term
nature of results
Measurement
challenges
Vulnerability to
biases
Lack of existing
data
Poor data
reliability
Poor data
accessibility
Unpredictable
chains of
causation
Complex and
dynamic
contexts
M&E within the programme cycle
Identification
Problem analysis
Appraisal
Evidence of what
works
Design
Most cost-
effective
intervention/s
Implementation
With M&E built in
from outset
Completion
Measure results
– did it work?
Post
Completion
Feed lessons
into future
decisions
Lesson Learning
and
Feedback
Challenges of programming in FCAS
Identification
Problem analysis
contested
Appraisal
Little robust
data and
research. No
time
Design
Little evidence
to assess cost
effectiveness.
Political
imperatives
Implementation
Great hurry. M&E lags
behind. No baselines/
measurement
strategies
Completion
Not enough data
to say. No
inclination to
admit failure
Post
Completion
Not enough
results
published/
stored/
synthesisd.
Disagreement No knowledge
management/sharing and
lots of uncoordinated
actors
3. Theories of Change (ToC)
Why are ToCs useful for M&E?
A ToC is an iterative and collaborative process for thinking through how a
programme is expect to work within the context of the broader system. It
should create the space for critical reflection and learning and be adjusted
and iterated over time.
• Links to assumptions box in LF, but goes beyond this in focusing on iterating
through learning shared mental models of how change happens
• Important for developing M&E strategy – test key links and assumptions
(intellectual leaps) in the causal chain over the life of the programme
• Important for evaluability – provides foundation for a theory-based evaluation
• Important to talk of ‘theories’ not ‘theory’ – i.e. to recognise and manage a
range of theories and multiple drivers of change
• Not a tick-box exercise or management tool like the LF but a way of working
and thinking – it’s primarily a process rather than a product
What are the pitfalls in FCAS?
• Time and resource-consuming – so they can often be poorly conceived/
too vague
• Poorly understood/used – as linear tick-box exercise rather than iterative
approach
• Oversimplification of complex contextual (e.g. conflict) factors – reflexivity
and feedback loops in complex conflict systems – black swan idea
• Absence of/poor conflict analysis – must underpin project design
• Difficulties in evidence gathering/data collection – conflict environments
are often data rich but information poor – insecurity, staff turnover
• Difficulties of working with and aiming to influence a range of actors
• Unpacking chains of cause and effect in FCAS can be very difficult
• Death by diagram
• Funnel of attrition
The funnel of attritionOnly these people
may experience
improved outcomes
4. M&E approaches and methods
Recent explosion of new and innovative
approaches to monitoring and evaluation:
1. Use of mobile technology and ICTs for data
collection and analysis – e.g. Ushahidi
2. Influence of complexity science – PDIA, DDD –
enabling environment for experimentation
3. Remote monitoring and verification
4. Rigorous evaluation/impact evaluation designs
Why evaluate?
• White and Waddington (2012):
‘The use of the systematic reviews methodology is
comparatively new among social scientists in the
international development field, but has grown
rapidly in the last 3 years...To date, there has not
been a strong tradition of using rigorous evidence in
international development. The evidence bar has
been rather low, with many policies based on
anecdote and ‘cherry picking’ of favourable cases’.
Why evaluate?
• Accountability and lesson-learning
– Accountability to taxpayers and beneficiaries
– Understanding what works, why, where and for whom
to underpin evidence-based programming
– Priority to evaluate interventions with a weak
evidence base
• Inform scale up of an intervention or transfer to
another context
• Make mid-course corrections
• To support spending decisions
What is impact evaluation?
“Impact evaluation is a with versus without analysis: what happened with the
programme (a factual record) compared to what would have happened in the
absence of the programme (which requires a counterfactual)” (White, 2013)
“Impact evaluation aims to demonstrate that development programmes lead
to development results, that the intervention has a cause and effect” (Stern
et al. 2012)
• Attribution analysis to understand what difference a programme made
• Counterfactual construction through experimental/quasi-experimental
methods for large n (comparison groups); causal chain analysis for small n
• Theory-based impact evaluation – in ideal world, an RCT should be
embedded in a broader theory-based design that addresses questions
across the causal chain (White, 2013)
• Causal chain analysis – rigorous empirical assessment of causal
mechanisms and the assumptions that underlie the causal chain
How do we estimate impact?
How do we estimate impact?
Pros and cons of RCTs
• Pros: RCTs are the “gold standard” for addressing
attribution when an ex ante design is possible with a
large number of units of assignment
• BUT MAJOR DRAWBACKS, ESPECIALLY IN FCAS
– Not suited to complex development pathways with
multiple non-linear causal factors
– Less appropriate where hard to identify comparison
groups – threat to validity
– When extrapolated from their context, RCT findings
lose claims to rigour (Pritchett and Sandefur, 2013)
How best to evaluate in FCAS?
In increasing order of robustness:
• Use of evaluation framework and robust
approach to evidence assessment – e.g.
humanitarian evaluations
• Use of theories of change and contribution
analysis to test causation and assumptions
• Realist evaluation design looking at how
different mechanisms operate in contexts
Using an evaluation framework
Questions
Theory/
Approach
Methods
Tools
Establishing a framework for the evaluation provides a
consistent and systematic means to designing the
evaluation, collating and analysing the existing evidence
and the new data created, and generating and
interpreting the results. (Magenta Book para 6.1)
Theory or approach
•Results-oriented
•Theory-based
•Participatory/
empowerment
•Utilization-focused
Methods
•Qualitative
•Case study
•Experimental
•Value for Money
•Contribution analysis
Tools
•Document review
•Key informant interview
•FGD
•Direct observation
•Questionnaire survey
•Participatory data collection
•SWOT
•Forcefield
•Stakeholder analysis
•Ranking and scoring
•[Types of analysis; CBA;
VfM; QCA; etc]]
Methods define
which tools and
how to use them
Evaluating peacebuilding
• Most useful definition of impact – understand
effects of intervention on conflict drivers
• Conflict analysis is critical – understand/test
relevance of intervention to conflict drivers
• Use of ToC to understand/test assumptions
about how intervention contributes to change
• Experimental approaches usually not useful –
better to look at contribution
M&E Group Exercise
• Split into 4 groups
• 2 groups will be responsible for designing an
outline M&E system for a peacebuilding
programme
• 2 groups will be responsible for designing an
outline proposal to do an external evaluation
of the same programme
Further Reading
Literature on M&E approaches and methods
• L. Morra Imas, Rist, R., The Road to Results (World Bank, 2009)
• S. Funnell, Rogers, P., Purposeful Program Theory (Wiley, 2011)
• E. Stern et al., ‘Broadening the range of designs and methods for impact evaluation’, DFID
working paper 38, April 2012
• H. White, Phillips, D., ‘Addressing Attribution of cause and effect in small n impact
evaluations’, 3ie Working Paper 15, June 2012
• G.Westhorp, ‘Realist impact evaluation: an introduction’, September 2014
Literature on M&E with specific reference to FCAS
• DFID, ‘Results in Fragile and Conflict-affected States and Situations’, 2012
• DFID, ‘Back to Basics, A compilation of best practices in design, monitoring and evaluation in
fragile and conflict-affected environments,’ March 2013
• L. Schreter, Harmer, A., Delivering Aid in Highly Insecure Environments, 2013
• S. Herbert, ‘Perceptions surveys in fragile and conflict-affected states’, GSDRC Helpdesk
Research Report, March 2013
• DFID, ‘Evaluating impacts of peacebuilding interventions’, May 2014
• J. Puri et al. ‘What methods may be used in impact evaluations of humanitarian assistance’,
3ie working paper 22, December 2014
Thank you for listening - any questions?
david.fleming@itad.com

FCAS M&E Seminar

  • 1.
    Monitoring and Evaluationin Fragile and Conflict-Affected Contexts: The challenges of measurement David Fleming, Senior Consultant, Itad Date: 28th January 2015
  • 2.
    Seminar Outline 1. IntroducingItad: Life as an M&E consultant 2. Introducing/recapping M&E: Why monitor and evaluate and why important in FCAS? 3. Theories of change: what they are, why they are useful and challenges in FCAS 4. M&E approaches and methods: how to monitor and evaluate in FCAS; examples from peacebuilding and humanitarian work
  • 3.
    Learning objectives 1. Comeaway with a better understanding of why we do M&E and why it’s particularly important in FCAS 2. Learn about and put into practice some of the most important M&E methods and tools for FCAS 3. Be able to better identify the challenges of doing M&E in FCAS and how to overcome these 4. Everyone to leave the room with a burning desire to get involved in M&E at some point in the future!
  • 5.
    Life as anM&E consultant…
  • 6.
    2. Introducing M&E: WhyMonitor and Evaluate? “After decades in which development agencies have disbursed billions of dollars for social programs, and developing country governments and nongovernmental organizations (NGOs) have spent hundreds of billions more, it is deeply disappointing to recognize that we know relatively little about the net impact of most of these social programs” ‘When will we ever learn?’ Evaluation Gap Working Group, Center for Global Development 2006
  • 7.
    • Monitoring: “Collectionof data with which managers can assess extent to which objectives are being achieved” (World Bank) – Purpose: Collect information on programme outputs and outcomes to track and improve performance and results • Evaluation: “Determination of the value of a project, programme or policy” (World Bank) – Purpose: evidence-based decisions, accountability, transparency, lesson learning – Types: project, programme, policy, organisation, sector, theme, formative, summative, impact…
  • 8.
    Why is M&Eimportant in FCAS? 1. Development trends in FCAS • By 2015, 50% of world’s poor will live in fragile states (OECD); by 2030 it might be two thirds (Brookings) • Support to conflict, violence and fragility becoming a key priority for most major donors • ODA to fragile states is falling in quantity but number of actors multiplying (OECD) • DFID has been scaling up support to FCAS (commitment to increase to 30% of ODA by 2015) • DFID strategies include BSOS, cross-Whitehall CSSF, and the ‘Beyond Aid’ agenda
  • 9.
    Why is M&Eimportant in FCAS? 2. Increasing emphasis on transparency, accountability and fiduciary risk • Higher risk to investments in terms of results, security and fiduciary risk 3. More limited evidence base – need for lesson learning and evidence of what works • Support evidence-informed decisions and better programming by knowing what works and doesn’t and why and in which contexts
  • 10.
    What are thebiggest challenges? Risk of exacerbating conflict Hawthorne effect Insecurity Political objectives Longer-term nature of results Measurement challenges Vulnerability to biases Lack of existing data Poor data reliability Poor data accessibility Unpredictable chains of causation Complex and dynamic contexts
  • 11.
    M&E within theprogramme cycle Identification Problem analysis Appraisal Evidence of what works Design Most cost- effective intervention/s Implementation With M&E built in from outset Completion Measure results – did it work? Post Completion Feed lessons into future decisions Lesson Learning and Feedback
  • 12.
    Challenges of programmingin FCAS Identification Problem analysis contested Appraisal Little robust data and research. No time Design Little evidence to assess cost effectiveness. Political imperatives Implementation Great hurry. M&E lags behind. No baselines/ measurement strategies Completion Not enough data to say. No inclination to admit failure Post Completion Not enough results published/ stored/ synthesisd. Disagreement No knowledge management/sharing and lots of uncoordinated actors
  • 13.
    3. Theories ofChange (ToC)
  • 14.
    Why are ToCsuseful for M&E? A ToC is an iterative and collaborative process for thinking through how a programme is expect to work within the context of the broader system. It should create the space for critical reflection and learning and be adjusted and iterated over time. • Links to assumptions box in LF, but goes beyond this in focusing on iterating through learning shared mental models of how change happens • Important for developing M&E strategy – test key links and assumptions (intellectual leaps) in the causal chain over the life of the programme • Important for evaluability – provides foundation for a theory-based evaluation • Important to talk of ‘theories’ not ‘theory’ – i.e. to recognise and manage a range of theories and multiple drivers of change • Not a tick-box exercise or management tool like the LF but a way of working and thinking – it’s primarily a process rather than a product
  • 15.
    What are thepitfalls in FCAS? • Time and resource-consuming – so they can often be poorly conceived/ too vague • Poorly understood/used – as linear tick-box exercise rather than iterative approach • Oversimplification of complex contextual (e.g. conflict) factors – reflexivity and feedback loops in complex conflict systems – black swan idea • Absence of/poor conflict analysis – must underpin project design • Difficulties in evidence gathering/data collection – conflict environments are often data rich but information poor – insecurity, staff turnover • Difficulties of working with and aiming to influence a range of actors • Unpacking chains of cause and effect in FCAS can be very difficult • Death by diagram • Funnel of attrition
  • 16.
    The funnel ofattritionOnly these people may experience improved outcomes
  • 17.
    4. M&E approachesand methods Recent explosion of new and innovative approaches to monitoring and evaluation: 1. Use of mobile technology and ICTs for data collection and analysis – e.g. Ushahidi 2. Influence of complexity science – PDIA, DDD – enabling environment for experimentation 3. Remote monitoring and verification 4. Rigorous evaluation/impact evaluation designs
  • 19.
    Why evaluate? • Whiteand Waddington (2012): ‘The use of the systematic reviews methodology is comparatively new among social scientists in the international development field, but has grown rapidly in the last 3 years...To date, there has not been a strong tradition of using rigorous evidence in international development. The evidence bar has been rather low, with many policies based on anecdote and ‘cherry picking’ of favourable cases’.
  • 20.
    Why evaluate? • Accountabilityand lesson-learning – Accountability to taxpayers and beneficiaries – Understanding what works, why, where and for whom to underpin evidence-based programming – Priority to evaluate interventions with a weak evidence base • Inform scale up of an intervention or transfer to another context • Make mid-course corrections • To support spending decisions
  • 21.
    What is impactevaluation? “Impact evaluation is a with versus without analysis: what happened with the programme (a factual record) compared to what would have happened in the absence of the programme (which requires a counterfactual)” (White, 2013) “Impact evaluation aims to demonstrate that development programmes lead to development results, that the intervention has a cause and effect” (Stern et al. 2012) • Attribution analysis to understand what difference a programme made • Counterfactual construction through experimental/quasi-experimental methods for large n (comparison groups); causal chain analysis for small n • Theory-based impact evaluation – in ideal world, an RCT should be embedded in a broader theory-based design that addresses questions across the causal chain (White, 2013) • Causal chain analysis – rigorous empirical assessment of causal mechanisms and the assumptions that underlie the causal chain
  • 22.
    How do weestimate impact?
  • 23.
    How do weestimate impact?
  • 24.
    Pros and consof RCTs • Pros: RCTs are the “gold standard” for addressing attribution when an ex ante design is possible with a large number of units of assignment • BUT MAJOR DRAWBACKS, ESPECIALLY IN FCAS – Not suited to complex development pathways with multiple non-linear causal factors – Less appropriate where hard to identify comparison groups – threat to validity – When extrapolated from their context, RCT findings lose claims to rigour (Pritchett and Sandefur, 2013)
  • 25.
    How best toevaluate in FCAS? In increasing order of robustness: • Use of evaluation framework and robust approach to evidence assessment – e.g. humanitarian evaluations • Use of theories of change and contribution analysis to test causation and assumptions • Realist evaluation design looking at how different mechanisms operate in contexts
  • 26.
    Using an evaluationframework Questions Theory/ Approach Methods Tools Establishing a framework for the evaluation provides a consistent and systematic means to designing the evaluation, collating and analysing the existing evidence and the new data created, and generating and interpreting the results. (Magenta Book para 6.1)
  • 27.
    Theory or approach •Results-oriented •Theory-based •Participatory/ empowerment •Utilization-focused Methods •Qualitative •Casestudy •Experimental •Value for Money •Contribution analysis Tools •Document review •Key informant interview •FGD •Direct observation •Questionnaire survey •Participatory data collection •SWOT •Forcefield •Stakeholder analysis •Ranking and scoring •[Types of analysis; CBA; VfM; QCA; etc]] Methods define which tools and how to use them
  • 28.
    Evaluating peacebuilding • Mostuseful definition of impact – understand effects of intervention on conflict drivers • Conflict analysis is critical – understand/test relevance of intervention to conflict drivers • Use of ToC to understand/test assumptions about how intervention contributes to change • Experimental approaches usually not useful – better to look at contribution
  • 29.
    M&E Group Exercise •Split into 4 groups • 2 groups will be responsible for designing an outline M&E system for a peacebuilding programme • 2 groups will be responsible for designing an outline proposal to do an external evaluation of the same programme
  • 30.
    Further Reading Literature onM&E approaches and methods • L. Morra Imas, Rist, R., The Road to Results (World Bank, 2009) • S. Funnell, Rogers, P., Purposeful Program Theory (Wiley, 2011) • E. Stern et al., ‘Broadening the range of designs and methods for impact evaluation’, DFID working paper 38, April 2012 • H. White, Phillips, D., ‘Addressing Attribution of cause and effect in small n impact evaluations’, 3ie Working Paper 15, June 2012 • G.Westhorp, ‘Realist impact evaluation: an introduction’, September 2014 Literature on M&E with specific reference to FCAS • DFID, ‘Results in Fragile and Conflict-affected States and Situations’, 2012 • DFID, ‘Back to Basics, A compilation of best practices in design, monitoring and evaluation in fragile and conflict-affected environments,’ March 2013 • L. Schreter, Harmer, A., Delivering Aid in Highly Insecure Environments, 2013 • S. Herbert, ‘Perceptions surveys in fragile and conflict-affected states’, GSDRC Helpdesk Research Report, March 2013 • DFID, ‘Evaluating impacts of peacebuilding interventions’, May 2014 • J. Puri et al. ‘What methods may be used in impact evaluations of humanitarian assistance’, 3ie working paper 22, December 2014
  • 31.
    Thank you forlistening - any questions? david.fleming@itad.com