Designing disaggregated indicators for equity and avoiding double counting

Development projects often measure progress through indicators, such as school enrollment rates, access to clean water, or vaccination coverage. But indicator tracking that looks only at overall averages can hide inequities. A national literacy rate might look impressive, while girls in rural areas remain far behind. Equity-focused indicator tracking ensures that every group’s progress is visible. By disaggregating indicators, practitioners can uncover hidden disparities and take targeted action. This leads to fairer, more effective development programs and better accountability to the communities they serve.

Uncover hidden disparities and avoid double counting
Uncover hidden disparities and avoid double counting

To understand the topic better, we first need to understand the difference between equality and equity:

  • Equality gives everyone the same resources or opportunities.
  • Equity recognizes different circumstances and allocates resources to achieve fair outcomes.

For example, providing the same number of textbooks to every school (equality) may still disadvantage schools in remote areas that lack trained teachers. Equity means measuring and responding to differences.

In indicator tracking, disaggregation means breaking down indicators by relevant categories (e.g., gender, age, location) to reveal differences in outcomes. Instead of reporting a single figure, data is reported for subgroups to show who benefits and who might be left behind.

For instance, “80% access to clean water” may mask gaps between urban and rural regions or inequities by socio-economic status. Disaggregation transforms indicator tracking from a single average into a tool for equity analysis.

Disaggregation categories

The most common disaggregation categories used in development projects include:

  • Gender: to highlight differences between men, women, and gender-diverse people
  • Age: useful for identifying gaps between children, youth, adults, and older people
  • Geography: urban vs. rural, regional breakdowns, or community-level data
  • Socio-economic status: income quintiles, employment status, or poverty level

Although these are essential, they aren’t always sufficient to expose the full complexity of inequities. Other less common but powerful categories include:

  • Disability status: often overlooked but critical for inclusion
  • Ethnicity, language or caste: can reveal structural exclusion
  • Migration or displacement status: especially relevant in humanitarian or urban contexts
  • Sexual orientation and gender identity (SOGI): when ethically appropriate
  • Occupation or livelihood group: for example informal vs. formal workers
  • Religion or faith group: in contexts where religious identity affects service access
  • Digital access: increasingly important for programs using online tools
  • Environmental vulnerability: for example flood-prone vs. non flood-prone communities
  • Intersectional combinations: sometimes there might be a combination of categories, for example gender & disability or age & migration status

It is important to always do ethical data collection, consider community consent, and privacy, especially for sensitive categories. Because disaggregated indicator tracking can involve sensitive data we need to keep these principles in mind:

  • Informed consent and voluntary participation.
  • The Do no harm principle to avoid reinforcing stigma or discrimination.
  • Data protection best practices to ensure we use secure systems and anonymization where needed.
  • Ensure community engagement to involve affected groups in indicator design and interpretation.
  • Be politically aware and prepared for sensitive conversations as disaggregated data can challenge dominant narratives.

Designing disaggregated indicators

Before starting out it is useful to discuss with your project team the following questions:

  • Are your project objectives explicitly equity-focused?
  • Have you identified priority groups likely to face inequities?
  • Have you selected relevant disaggregation categories (common + context-specific)?
  • Have you planned data collection methods and ensured sample sizes are adequate?
  • Do your data tools include appropriate indicator tracking fields?
  • Have you considered ethical safeguards and community engagement?
  • Do you have a plan to analyze and use disaggregated data for decision-making?

As for the sources and the data collection methods you should keep in mind that each method has its own benefits and limitations. For example, household surveys can be highly detailed but they can be costly, administrative data can have wide coverage but their quality may vary, GIS data provides important insights on geographic disaggregation but is limited for social variables.

Once you have the answers to the questions above you can follow the steps below which apply to indicator tracking in general:

  • Clarify the purpose: Align indicators with your equity objectives.
  • Select relevant categories: Combine common and context-specific disaggregation.
  • Define indicators clearly: Use consistent definitions across groups.
  • Plan data collection: Adjust tools and sampling strategies.
  • Build capacity & systems: Train staff and strengthen infrastructure.
  • Analyze & use data: Feed findings back into program design and decision-making.

Take a look at common mistakes to avoid in indicator tracking.

Analyzing and visualizing disaggregated data while avoiding double-counting

Following the collection of the disaggregated data, to make the disparities visible and easy to act upon you need to analyze and visualize the data. Common data analysis techniques include cross-tabulation (e.g. vaccination rate by gender and district), gap analysis (e.g. rural vs urban outcomes), etc. As for visualization, bar charts, pivot tables, and dashboards can highlight key differences and help spur action.

A common risk we can encounter when working with disaggregated indicators is double-counting individuals, households, or activities. This can inflate results, distort subgroup comparisons, and undermine the credibility of indicator tracking. The more finely we disaggregate the higher the chance of overlaps between categories. Each additional category introduces intersections that need to be managed carefully.

Double-counting happens when there are:

  • Overlapping categories: If individuals belong to multiple subgroups (e.g., a woman with a disability who is also a migrant), and the data is summed across categories without proper adjustment, total numbers may exceed the real population.
  • Multiple data collection points: If data is gathered from different sources or over time, and unique identifiers are not used, the same participant can be counted more than once.
  • Poorly defined indicators: Vague definitions (e.g., “number of people reached”) can make it unclear how to treat people who participate in multiple activities.
  • Aggregation errors: When disaggregated data are recombined without clear rules (e.g., summing gender and disability breakdowns), it can lead to inflated totals.

For example if a project reports results disaggregated by gender, age, and disability, one woman with a disability may appear in multiple subgroup tallies (e.g., “female,” “18–24,” and “people with disabilities”). Without clear rules, she could be counted three times.

To avoid double-counting, you need to plan your approach right from the start of the data collection process:

  • Use unique identifiers where possible to track individuals across categories and time.
  • Design indicators clearly to specify whether results represent unique individuals or instances of participation.
  • Structure your data to allow for intersectional analysis (e.g., using multi-variable breakdowns or pivot tables) rather than simply summing across categories.
  • Train data collectors and analysts to understand the difference between subgroup totals and overall totals.
  • Document your counting rules explicitly in the M&E plan or metadata.

Avoiding double-counting with ActivityInfo

In ActivityInfo, we offer the tools to design data systems that help prevent double counting. It is possible to uniquely identify records, link records to each other so as to track unique individuals across multiple interventions, rather than counting them separately in each activity.

In addition, you can apply validation rules to your data collection forms or require unique field combinations to catch double entries at the point of data input. With the duplicate scanner you can review potential duplicates and merge them and with the [audit log](/support/docs/maintenance/viewing-a-database-audit-log.html you can trace when changes were applied so as to resolve issues on time.

As for data analysis, ActivityInfo offers powerful pivot tables that allow you to get totals based on structured relationships rather than ad hoc Excel formulas, reducing aggregation errors.

All in all, disaggregated indicator tracking is a powerful way to turn data into a tool for equity, making hidden disparities visible and guiding fairer decisions. Yet as categories multiply, so do the risks of errors like double counting, where individuals may be counted more than once across overlapping groups. By setting clear counting rules, using careful analysis, and treating data both technically and ethically, practitioners can ensure their indicators reflect reality and truly support inclusive development.

Would you like to learn how ActivityInfo could be used to support your organization in indicator tracking? You can always contact us for a discussion and a tailored demonstration.