Evidence and impact measurement

MSC’s Evidence and Impact Measurement (EIM) team works across South Asia, Southeast Asia, and Sub-Saharan Africa. Our clients include national and subnational governments, bilateral and multilateral donors, philanthropic foundations, development finance institutions, financial service providers, nonprofits, and private sector partners. Recent assignments span India, Indonesia, Bangladesh, Sri Lanka, Kenya, Uganda, Ghana, and Zambia. 

Our evaluation teams combine international methodological expertise with local knowledge and in-country research capacity. This approach allows us to tailor research designs to local institutions, infrastructure, culture, program delivery, and data availability. It also maintains consistent standards of analytical rigor. 

MSC has worked for more than 27 years in nearly 70 countries across Africa, Asia, and the Pacific, with a team of about 450 staff. This footprint allows EIM teams to draw lessons across countries while keeping recommendations grounded in local conditions. 

MSC’s Evidence and Impact Measurement (EIM) services help funders, governments, development organizations, and mission-driven institutions understand what works, for whom, why, and at what cost across three service areas. These are evaluation and evidence generation; monitoring, evaluation, and learning (MEL) strategy and systems; and institutional capacity building. 

We conduct impact, outcome, process, performance, policy, and complex program evaluations. Depending on the research question, we apply experimental and quasi-experimental methods, mixed-methods research, and theory-based approaches. 

We also develop theories of change, results frameworks, indicators, monitoring systems, dashboards, learning mechanisms, and institutional MEL strategies. For organizations that want to strengthen their internal capabilities, we provide MEL diagnostics, training, system design, technical assistance, and strategic advisory. 

Our focus includes evidence generation and its practical use for program improvement, investment decisions, accountability, policy, and scale.  

Recent examples include the national evaluation of India’s flagship financial inclusion schemes, such as Pradhan Mantri Jan Dhan Yojana (PMJDY), Pradhan Mantri Jeevan Jyoti Bima Yojana (PMJJBY), Pradhan Mantri Suraksha Bima Yojana (PMSBY), and Stand-Up India.  

We conducted this national evaluation for the Development Monitoring and Evaluation Office (DMEO) at NITI Aayog. Another example is the concurrent evaluation of the Soil Health Card program across 54 districts in 18 states. The evaluation covered more than 11,000 farmers.  

Monitoring, evaluation, and learning (MEL) are three interconnected functions in program management.  

Monitoring tracks program implementation, including reach, outputs, expenditure, delivery quality, and progress against targets.  

Evaluation examines what changed, whether and how a program contributed to that change, for whom it worked, and why. Depending on the assignment, it may also assess relevance, efficiency, sustainability, scalability, and value for money. 

Learning turns evidence into decisions. It creates structured opportunities for program teams, funders, and policymakers to interpret findings, test assumptions, adapt implementation, and improve future investments.  

The three functions work best when teams design them together. Monitoring data feeds evaluation, and both feed a learning process that changes what the program does next. 

MSC designs and implements a range of evaluations. These include impact and outcome evaluations; baseline, midline, and endline studies; process  evaluations; program and portfolio evaluations; policy and institutional evaluations; real-time, rapid, and concurrent evaluations; value-for-money and economic assessments; and mixed-method and theory-based evaluations of complex interventions. 

We use approaches such as randomized controlled trials (RCTs), difference-in-differences (DiD), regression discontinuity design (RDD), propensity score matching (PSM), and synthetic control methods when causal attribution is feasible and useful.  

For example, MSC applied a modified RDD to evaluate Indonesia’s Program Keluarga Harapan conditional cash transfer program on behalf of the Ministry of Social Affairs. We also used a DiD design to evaluate the impact of Berendina Micro Investments Company in Sri Lanka. 

We use theory-based approaches such as contribution analysis, process tracing, and realist evaluation when programs operate in complex environments or counterfactual analysis is not feasible. We choose the method based on the evaluation question, program design, data availability, ethical considerations, and operational context. This ensures the evaluation approach fits the context and available evidence. 

Yes. Many development programs involve multiple actors, changing contexts, overlapping interventions, and long pathways between activities and outcomes. 

In such cases, MSC combines quantitative analysis with theory-based and qualitative methods to examine whether change occurred and how the program contributed to it. We select methods based on the assignment. These may include contribution analysis, process tracing, realist approaches, comparative case analysis, mixed methods, longitudinal evidence, administrative data, and stakeholder research. 

This approach lets us assess complex programs without making attribution claims the available evidence cannot support. We clearly state the strength of the evidence behind each finding so readers understand how much weight it carries. 

MSC treats theory of change and results frameworks as practical tools for program management and learning, rather than a static compliance requirement. We work with technical program teams, funders, implementers, and program participants so that the logic reflects how change actually happens. The process includes facilitated workshops, stakeholder consultations, and iterative review to produce frameworks that are clear, measurable, and useful for adaptive management. For example, MSC has used the theory of change and logframe methods to build Stanbic Bank Kenya’s integrated monitoring, evaluation, research, and learning (MERL) framework. 

We translate this logic into measurable results frameworks, indicator definitions, data sources, reporting responsibilities, review cycles, and learning questions. As evidence emerges, we help teams revisit assumptions and refine the theory of change so that it remains useful throughout implementation.  

MSC designs evaluations with use in mind from the outset. We work with decision-makers to identify the questions that matter, when evidence will be needed, and which program, investment, or policy decisions the evidence should inform. 

During implementation, we use dashboards, learning reviews, sensemaking or validation workshops, rapid analyses, briefs, and other formats to make evidence accessible and timely. At the end of an evaluation, we go beyond presenting findings. We help stakeholders interpret what the evidence means for program design, implementation, targeting, resource allocation, scale, and future research.

MSC builds inclusion into evaluation design rather than treating it as a reporting requirement. Depending on the context, this may include purposeful representation of women, persons with disabilities, youth, low-income households, remote and underrepresented communities, or underserved groups; gender- and disability-disaggregated analysis aligned with gender equality, disability, and social inclusion (GEDSI) standards; accessible research tools; and qualitative research to surface experiences that aggregated data may conceal.  

We also examine whether programs work differently for different groups, what barriers affect participation and outcomes, and which contextual factors influence results. 

Our teams work closely with local researchers, field teams, program partners, and communities so that methods, tools, language, and interpretation reflect local realities. Research ethics, informed consent, privacy, safeguarding, and do-no-harm principles are built into every stage of the research process, as set out in the next question. 

Every Evidence and Impact Measurement (EIM) assignment is designed around do-no-harm principles. Before fieldwork begins, we assess ethical risks and, when required by the assignment or the client, obtain approval from an accredited institutional review board or national ethics committee. 

Our standard protocols cover informed, documented consent in the respondent’s own language; the right to decline or withdraw at any point; additional safeguards for children and other vulnerable respondents; and safe referral pathways when research surfaces protection concerns.  

Field teams are trained and certified on ethics and safeguarding before deployment.  

For data protection, we apply role-based access controls, encryption in transit and at rest, pseudonymization or anonymization of personally identifiable information, agreed retention and deletion schedules, and compliance with applicable data protection law. These include the European Union’s General Data Protection Regulation (EU GDPR), India’s Digital Personal Data Protection Act (DPDP Act), and country-specific requirements. We agree on data ownership and data-sharing terms with the client during contracting.

MSC’s Evidence and Impact Measurement (EIM) practice works across development sectors where programs, policies, and investments need credible evidence of performance and impact. 

Our experience includes financial inclusion and digital financial services, social protection, and digital public infrastructure (DPI).  

We also work on women’s economic empowerment and gender equality, disability, and social inclusion (GEDSI) approach; micro, small, and medium enterprises (MSMEs) development; and agriculture and food systems. Our work also covers livelihoods and employment, climate resilience, skills and technical and vocational education and training (TVET), government programs, and institutional development. 

EIM also works closely with MSC’s sector specialists. This combines evaluation expertise with an understanding of program operations, markets, institutions, user behavior, policy, and service delivery systems. 

The Evidence and Impact Measurement (EIM) practice brings together evaluation specialists, economists, statisticians, qualitative researchers, and sector experts. MSC’s in-house data science team supports this work. Most team members hold advanced degrees in economics, statistics, public policy, or public health. They also bring a decade or more of field experience in low- and middle-income countries. 

For quantitative analysis, we use Stata, R, and Python, which include econometric modeling and, where useful, machine learning. For qualitative analysis, we use NVivo and comparable coding platforms. We collect primary data through computer-assisted personal interviewing platforms such as SurveyCTO and KoboToolbox. These platforms support GPS stamps, audio audits, and built-in logic and consistency checks. We deliver findings through Power BI or Tableau dashboards when clients need live tracking. 

Quality assurance is built into the workflow. This includes instrument piloting and translation back-checks, enumerator training and certification, independent back-checks and spot checks on a sample of interviews. We also conduct high-frequency data quality checks during fieldwork, use version-controlled and reproducible analysis code, and complete an internal peer review of every deliverable before it reaches the client.

Yes. MSC can support an assignment from initial program design through evaluation and institutional learning.  

We can tailor our support to the client’s needs. This support can include theory-of-change development, evaluation questions, indicator frameworks, sampling strategies, research instruments, and primary and secondary data collection. It can also include data quality assurance, quantitative and qualitative analysis, dashboards, evaluation reports, learning products, dissemination, and recommendations. 

We can also strengthen the systems behind the evidence. These systems include data architecture, reporting processes, review mechanisms, staff capabilities, and institutional monitoring, evaluation, and learning (MEL) arrangements. Clients can engage MSC for an independent evaluation, a specific technical component, or longer-term MEL and learning support. 

MSC helps clients assess whether a program works and whether its results justify the resources invested. 

Depending on the program and available data, we may examine economy, efficiency, effectiveness, and equity. Donors such as the Foreign, Commonwealth and Development Office (FCDO) use these four dimensions to assess value for money, along with sustainability. We apply cost-effectiveness analysis, cost-benefit analysis, benefit-cost ratios, net present value, and unit-cost benchmarking. 

We combine financial and program data with evidence on implementation and outcomes. This approach helps us explain both the numbers and the operational factors behind them. 

This analysis helps funders and program managers compare approaches, identify opportunities for greater efficiency, and make informed decisions about continuation, redesign, or scale.

MSC supports programs that link funding, management, or accountability to measurable results. 

Our work includes the definition of outcomes and indicators, the establishment of baselines and targets, and the design of verification and reporting systems.We also assess data quality, develop results frameworks, and create monitoring, evaluation, and learning (MEL) arrangements to track progress toward agreed outcomes. 

We can support clients with results-based financing mechanisms, including outcome funds and, where relevant, development and social impact bonds. 

Our support can cover feasibility assessments, outcome metric design, verification approaches, and evaluation design.  

For example, MSC provided monitoring, results measurement, and reporting support for the Micro and Small Enterprise Recovery Fund of FSD Uganda and the Mastercard Foundation. The revolving facility tracked performance through concurrent monitoring and a nationwide survey of micro and small enterprises.  

Our focus is to link incentives to meaningful, measurable outcomes that we can feasibly verify. Our goal is to avoid excessive reporting burdens and perverse incentives.

Primary surveys are only one source of evaluation evidence. MSC assesses and triangulates evidence from existing program management information systems (MIS), administrative records, transaction data, monitoring databases, government datasets, and other digital data. We first assess what existing data can tell us and then identify what additional information we need to collect.

Where data quality and coverage permit, these sources can support longitudinal analysis, quasi-experimental evaluation, segmentation, outcome tracking, implementation diagnostics, and triangulation with primary research. 

We also assess data completeness, consistency, bias, interoperability, and fitness for the intended analysis. When existing data cannot answer the evaluation questions on its own, we design targeted primary research to fill the gaps. This approach reduces respondent burden and research costs. It also helps organizations make better use of the data they already collect. We handle all administrative and personal data under the protections described in the question on research ethics and data protection.

Clients engage MSC through competitive tenders, framework and panel arrangements, direct commissioning, and as long-term partners for monitoring, evaluation, and learning (MEL) or learning partner to a program or portfolio. 

Assignments usually begin with a short scoping conversation to clarify the decisions the evidence needs to inform, the questions that follow from them, the data already available, and the timeline. From there, we prepare a technical and financial proposal that sets out the design, sampling approach, instruments, deliverables, and team. 

Timelines vary with the design. A rapid assessment or diagnostic can be delivered in six to twelve weeks; a full mixed-methods outcome evaluation usually runs four to nine months; and experimental designs that require baseline and endline rounds run over multiple years.  

To start a conversation, write to us or use the contact form on this page.