Disability-Inclusive M&E: The Washington Group Short Set

When to Use

Use the Washington Group Short Set (WG-SS) when you need to know whether program results differ for people with and without disability, and you want a tested way to ask without a long module. It is a set of six questions, built for censuses and surveys, that asks about difficulty in basic activities rather than about diagnoses or labels.

Disability-inclusive M&E means more than one question. Include people with disabilities in design, make data collection accessible, and report disaggregated results. The WG-SS supplies the disaggregation variable, which sits beside sex and age as in disaggregation.

How It Works

The WG-SS asks about difficulty in six basic, universal activities. The wording is the Washington Group's own, and they ask that it be used as written; see the source page for the official questions and translations. In summary, the six areas are:

1. Seeing, even when wearing glasses
2. Hearing, even when using a hearing aid
3. Walking or climbing steps
4. Remembering or concentrating
5. Self-care, such as washing all over or dressing
6. Communicating, such as understanding others or being understood

Each question has four response categories: no difficulty, some difficulty, a lot of difficulty, and cannot do at all.

For international comparison the Washington Group recommends a standard cut-off. A person is counted as having a disability if they answer "a lot of difficulty" or "cannot do at all" on at least one of the six. State the cut-off in your report, along with the age range covered and how the questions were used. A person who answers "some difficulty" on everything is not counted at this cut-off, so keep the full response data to allow other cut-offs.

Key Components

• Wording and order. The Washington Group advises against changing question wording, order, response categories or the cut-off, because changes tend to reduce accuracy and comparability. Where pretesting shows part of a question is irrelevant or confusing in your setting, contact the Secretariat before adapting it.
• Placement. Put any extra functioning or health questions after the six, not before.
• Introduction. The short lead-in sentence may be left out, but should not be replaced by one that uses the word "disability."
• Who answers. Where possible, each adult answers for themselves. Proxy answers lose accuracy; record when a proxy was used.
• Translation. Use the translations on the Washington Group site and test any new ones with the group's translation guidelines.
• Reporting. Report disability prevalence by the standard cut-off, and cross it with your main indicators, for example school attendance or income by disability status.

Best Practices

• Train enumerators on the six questions. Practice reading them exactly as written and not suggesting answers. The enumerator training checklist has a role-play format that fits.
• Make data collection accessible. Check that venues, formats, interpreters and consent processes work for people with sensory, mobility or communication difficulties. Cross-check with do no harm and ethics in M&E.
• Plan the sample for it. A subgroup that is a small share of the population needs a large enough sample to compare. Check numbers before fixing the sample size.
• Include disability in other analyses. Cross it with sex and age; an older woman with a disability may face barriers that neither variable shows alone. See gender-responsive M&E.
• Add program-specific questions after the six. Questions about barriers, access and participation tell you more than prevalence does.
• Ask for the right permissions. Check the Washington Group's terms before copying their question text into your form or report, and cite them.

Common Mistakes

Rewording the questions. A "friendlier" paraphrase breaks comparability with national data and can change who is counted.

Replacing the intro with the word "disability." The questions work by asking about difficulty. A label can cause people to under-report.

Using it for children. The WG-SS is built for adults. Use the child module for children.

Treating it as a full picture. It identifies people more likely to face participation restrictions; it does not measure all impairments or diagnose. Anxiety and depression are better captured by the extended set or a dedicated tool.

Collecting it and never using it. If the variable is not crossed with indicators in the analysis plan, it is only a count.

Examples

A fictional adult literacy program, Lakeshore Learning, runs a baseline of 1,200 adults. It adds the six questions to the household roster, after age and sex, with the official wording read aloud. Enumerators practice them in training. In analysis, the team counts anyone with "a lot of difficulty" or "cannot do at all" on at least one question and compares class attendance and test gains between those with and without disability.

The result is that attendance among adults with difficulty seeing is far lower, because classes are held in dim rooms. The program changes the venue lighting and large-print materials. Without the disaggregation, the average attendance would have hidden the problem.

Compared To