Equity evaluation: measuring if a project truly leaves no one behind
In the development sector, projects are generally evaluated through their ability to produce measurable results: raising school enrolment rates, improving access to drinking water, strengthening household incomes, or increasing vaccination coverage. Monitoring and evaluation systems have been designed to answer one essential question: were the targets achieved?
This performance-centred approach has done much to professionalise development project management. It draws on logical frameworks, performance indicators and recognised evaluation criteria, notably those of the OECD Development Assistance Committee (DAC). Yet as development policies have placed the reduction of inequalities and inclusion at the heart of their priorities, a limitation has become apparent: a project can perform well without being equitable.
Take a simple example. A water access programme raises coverage from 55% to 85% in a region. On the face of it, the result looks remarkable. But a more detailed analysis reveals that this progress mainly concerns urban centres, while several remote villages remain excluded. The overall indicators point to success. The reality on the ground shows that the most vulnerable populations continue to be left behind.
It is no longer enough to demonstrate that a project works. We must understand for whom it works.
This is precisely what equity evaluation is about.
Equality, equity and effectiveness: three notions to distinguish
Before addressing methods for evaluating equity, three often-confused notions need to be distinguished, as they carry very different methodological implications.
This distinction is essential in development projects. Delivering the same service to all populations does not guarantee that everyone can benefit from it under the same conditions. Geographic, economic, social, cultural or disability-related constraints can limit some groups’ access to interventions, even when those interventions are open to all.
Equity does not stand in opposition to effectiveness; it is an essential complement to it. In monitoring and evaluation, effectiveness mainly answers the question “have the expected results been achieved?”. Equity evaluation, in turn, raises other fundamental questions:
- Who benefited from the results?
- Which groups were left out?
- Did gaps between populations narrow or widen?
- Did the delivery mechanisms enable or restrict access for certain categories of beneficiaries?
A project can be highly effective while unintentionally reinforcing existing inequalities.
Why traditional evaluations are no longer enough
Monitoring and evaluation systems have historically been designed to measure the overall performance of interventions. Aggregated indicators such as the number of beneficiaries, coverage rates, infrastructure delivered or execution rates provide a synthetic view that is essential for project steering.
The problem with this view: averages mask disparities.
- A 90% school enrolment rate can hide significant gaps between girls and boys.
- High vaccination coverage can conceal poor accessibility in certain rural areas.
- An average rise in income can mainly benefit households that were already best integrated into markets.
Put differently, two projects showing similar performance can have very different effects on inequalities.
This limitation has become particularly visible with the adoption of the Sustainable Development Goals (SDGs) and the Leave No One Behind principle. Organisations are now expected to measure not only the results achieved, but also how they are distributed across different population groups. This challenge connects with a broader, well-documented structural issue: the disconnect between data produced and decisions made.
Education programme in sub-Saharan Africa:
when an aggregated indicator masks massive exclusion
A five-year regional primary schooling programme, funded by a consortium of multilateral donors, targeted 240,000 children across eight districts. At the midterm review, the annual report showed a regional enrolment rate of 90%. The programme was considered an exemplary success.
An external evaluation, commissioned by one of the donors and focused on equity, produced a radically different reading by disaggregating the same data by gender and place of residence. The actual enrolment rate for girls in rural areas was capped at 62%, a gap of 34 points compared with boys in urban areas.
The report also revealed that school infrastructure had been built primarily in district capitals, and that children with disabilities were not counted in the monitoring system. The programme had delivered outstanding aggregated results while reproducing, and even reinforcing, existing inequalities.
The cost of this statistical blind spot is twofold: loss of intervention effectiveness for the most vulnerable populations, and loss of credibility with donors aligned on the SDGs.
How to integrate equity into an M&E system
Embedding equity into a monitoring and evaluation system is not a cosmetic adjustment. This integration rests on four complementary steps, which must be thought through in an articulated way.
1. Disaggregate the data
The first step is to move beyond overall figures through disaggregated data. This approach reveals gaps that are often invisible in aggregated statistics.
For instance, knowing that 10,000 people have benefited from a training programme is useful. Knowing that only 18% of these beneficiaries are women living in rural areas provides far more strategic information for decision-making.
2. Analyse inequalities and how they evolve
Disaggregating data is only a first step. The analysis must also measure gaps between groups and track how they change over time.
The objective is no longer solely to improve a performance indicator, but to check whether differences between populations are genuinely narrowing. This approach helps project managers better allocate their resources and improve the impact of interventions on priority populations.
3. Understand the mechanisms of exclusion
Quantitative data reveal gaps, but they rarely explain their causes. Qualitative methods are therefore essential in order to understand:
- Social norms that limit women’s participation;
- Indirect costs that hinder access to services;
- Mobility difficulties for persons with disabilities;
- Language barriers;
- Lack of information on available programmes;
- Low trust in certain institutions.
Combining quantitative and qualitative data produces evaluations that are more relevant and more useful for decision-making. It aligns with the logic of an M&E system that is accountable to beneficiaries, in which the populations concerned take part in interpreting the results.
4. Evaluate the relevance of targeting
Many projects put in place targeting mechanisms aimed at the most vulnerable populations. The evaluation must verify:
- Whether the selection criteria were relevant given the real needs;
- Whether the targeted beneficiaries were effectively reached;
- Whether certain vulnerable groups were unintentionally excluded;
- Whether targeting produced unintended effects such as social tensions or stigmatisation.
A targeting strategy is genuinely effective when it improves access for priority populations without creating new forms of exclusion.
Equity must be embedded from project design
One of the most common mistakes is to treat equity as an issue that only concerns the final evaluation. In reality, this dimension must be embedded from the planning phase.
The choice of M&E indicators, the definition of targets, data collection modalities and the way the information system is organised all directly shape the ability to measure inequalities several years later.
For example, an indicator formulated as “number of people trained” provides useful but limited information. Conversely, an indicator that distinguishes beneficiaries by sex, age, location or vulnerability level offers a much more precise view of the project’s impact, and above all of its distribution.
Structure inclusive M&E from design onwards
Delta Monitoring builds data disaggregation into its methodological foundations as a core principle. The platform lets teams define differentiated targets by group, compare performance across territories and automatically identify populations that are insufficiently covered throughout the project cycle.
The practical challenges of equity evaluation
Despite the methodological progress, embedding equity into M&E remains a demanding exercise.
The first challenge concerns data quality. Information systems do not always collect the variables required for reliable disaggregated analysis. When these variables are not planned from the design phase, correcting them mid-project is often impossible.
The second challenge is methodological. Depending on the context, equity may refer to equal opportunities, equal outcomes or the reduction of gaps between territories. It is therefore essential to define the chosen analytical framework clearly, in agreement with stakeholders and donors.
The third challenge is ethical. Collecting sensitive data requires strict guarantees around confidentiality, informed consent and data protection. These requirements are particularly important for vulnerable or marginalised groups.
Finally, intersectionality represents a major challenge. Inequalities do not simply add up; they combine. A woman living in a rural area and with a disability may face very different obstacles from those encountered by other groups. This reality calls for cross-tabulating disaggregation variables rather than treating them in isolation.
A new way of appreciating project performance
Evaluating equity does not mean abandoning traditional M&E approaches. Effectiveness, efficiency, impact and sustainability remain essential dimensions of performance.
The real question is no longer only:
“Did the project deliver results?”
but also:
“Did these results reach the populations that needed them most?“
This shift marks a profound change in results-based management. A project’s performance is now measured as much by its ability to deliver results as by its contribution to reducing inequalities.
In this context, digital M&E platforms play a strategic role. They make it possible not only to produce dashboards, but also to analyse results by beneficiary group, track differentiated targets, compare performance across territories and quickly identify populations that are insufficiently covered.
This is precisely the approach behind DELTA Monitoring. The platform makes it possible to structure results frameworks, manage M&E indicators, produce disaggregated data, analyse gaps between groups and strengthen data-driven decision-making throughout the full life cycle of development projects.