10 minutes
Measures that show access and outcomes Build indicators that describe whether people can get in, get through and get results, instead of counting attendance or checking a compliance box.
What you’ll be able to do Select access, experience and outcome measures for a program area, identify the data source and owner for each, and state what each measure does and does not show. Compare the responses to this question and explain your choice: What is the strongest response? Draft a one-page measurement plan for your work, including measure and layer, source and owner, protection rule. Three layers of measurement Most programs already count something: people trained, meetings held, policies signed. Those are activity measures. They are easy and they are not wrong, but they answer “what did we do,” not “did it work.” Practitioners add two layers. Access measures ask whether a person could get in and take part: could the document be read, was the request resolved in time, did the meeting have captions. Outcome measures ask whether the result was fair: did disabled participants finish the process at the same rate, stay in the job as long, report the same satisfaction.
The best measures share three features. They describe a system, so they can be improved without judging a person. They can be repeated, so change is visible. And they come with a known limit written next to them, so nobody reads more into the number than it holds.
One rule is absolute. Measures describe documents, requests, meetings, services and processes. They never score, compare or rank individual employees, and they never combine disability information with individual performance information.
Suggested measures Percentage of public documents meeting accessibility standards, with severity of remaining issues.
Average and longest time to resolve accommodation requests, by request category.
Percentage of meetings and events planned with accessible practices: materials in advance, captions, an access-needs line and a named contact.
Satisfaction with access, broken down by service type where the group size allows.
Complaint themes about access and the pattern of how they were resolved.
Hiring, retention, promotion and engagement indicators, using voluntary self-identification data only in aggregate and only above the minimum group size.
Number and severity of barriers found through user testing with disabled people, and the share fixed within an agreed window.
A one-page measurement plan The fields that turn a good intention into a measure someone will actually collect, read and act on.
Measure and layer What is counted, and whether it describes activity, access or outcome. Example: share of program notices passing an accessibility check (access).
Source and owner Where the data comes from and the named role responsible for producing it on schedule.
Protection rule The minimum group size, suppression approach and which of the five data types it draws on.
Limit and review point One sentence on what the measure cannot show, and when the group that owns it will look at it and decide what to change.
Complete the plan for one program area and review it with the people the measures are meant to serve before adopting it.
Browse and download only. Staff notes are not typed or saved on this page.
Download this published draft Measure the system you can change What you can change You control whether a measure is written about a process or about a person.
What to watch for Do not accept “accommodations per supervisor” or “absences among staff with accommodations” as inclusion measures. They point at individuals and invite exactly the judgment the data must never support.
Your next step Rewrite any person-pointing measure in your area so the subject of the sentence is a document, request, meeting or process.
Measure or mirage? “Percent of staff who completed the course”
Activity. Useful as context, useless alone. Pair it with a measure of what changed in documents, meetings or requests.
“Median days from accommodation request to decision”
Access. Describes the process, repeatable, actionable by the office that owns the process. Report the longest wait too; medians hide the person who waited four months.
“Retention after two years, disabled and non-disabled staff”
Outcome. Powerful at agency or division scale using voluntary self-identification in aggregate. Never produced for a team small enough to identify anyone.
“Barriers found in user testing, by severity”
Access, and the most direct barrier measure available. Requires paying disabled testers and fixing what they find.
“Number of people with disabilities on each team”
Not a measure of inclusion. It identifies people in small teams and says nothing about whether the team is accessible.
Carry this forward Attendance and training completion measure exposure; access measures whether a person could use the thing; outcome measures whether it worked for them.
Every good measure names the system it describes: documents, requests, meetings, services, complaints, products. Not one names a person.
A small set of measures reviewed on a schedule beats a large set nobody reads.
A one-page measurement plan with three to five measures, each with a source, an owner, a review point and a line on its limits.
Draft the measurement plan for one program area using the artifact in this lesson. Ask a disabled colleague or advisory member whether the measures would show the barriers they actually meet, and compensate their time where appropriate.
Browse and download only. Course notes are not typed or saved on this page.
Reset this lesson
Previous lesson Next lesson Participation and course completion in this program do not count toward DHS-required training credits unless management, a director, or DHS leadership expressly approves an exception.
