Assessment & Research

Validity evidence for self-report with assistance to measure physical activity behavior in adults with intellectual disabilities.

Johnson et al. (2014) · Intellectual and developmental disabilities 2014
★ The Verdict

A staff member can reliably ask adults with ID about their exercise and get data that match step counters.

✓ Read this if BCBAs running health programs in day-hab or group-home settings.
✗ Skip if Clinicians who only work with verbal, mild-ID clients who already fill out written surveys.

01Research in Context

01

What this study did

Marquell and colleagues asked adults with intellectual disability to talk about their weekly exercise.

A helper read each question aloud and wrote down the answer.

The team then compared these answers to step counts from pedometers and accelerometers the adults wore.

02

What they found

The helper-assisted interviews matched the device data well.

The study showed high reliability and moderate to strong validity for this simple interview method.

03

How this fits with other research

Lecavalier et al. (2006) warned that plain Likert scales fail for most adults with ID unless you add pictures and keep the disability level mild.

Marquell’s team side-stepped this problem: they used spoken questions and a helper, so even adults with moderate ID could take part.

Johnson et al. (2009) had already proven that three days of pedometer data gives a solid picture of weekly steps in the same population; Marquell now shows you can get equally solid data just by asking, as long as someone gives a gentle assist.

Guerra-Balic et al. (2015) validated the six-minute walk test for mobility in adults with ID; together these papers give you three cheap tools—interview, pedometer, and walk test—to measure activity without lab gear.

04

Why it matters

You can trade expensive devices for a ten-minute assisted interview.

Use the script from the paper, let the client pick easy words like “a little” or “a lot,” and record answers on the spot.

Pair the interview with a quick pedometer check on day one and day three to keep everyone honest; if numbers line up, you can fade the devices and just talk.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Try the assisted interview: read each activity question aloud, let the client answer in their own words, and record it yourself.

02At a glance

Intervention
not applicable
Design
other
Sample size
37
Population
intellectual disability
Finding
positive
Magnitude
medium

03Original abstract

Self-reported physical activity (PA) behavior with assistance from a secondary source has previously been used with adults with an intellectual disability (ID). Limited evidence of reliability and validity have been provided for this approach. This study examined evidence of convergent (CV) and discriminant (DV) validity for self-report with assistance from a secondary source as a measure of PA in adults with ID. PA of 37 participants with ID were assessed using (a) self-report, (b) accelerometers, and (c) pedometers. The multitrait-multimethod (MTMM) analysis was used to evaluate validity. MTMM analysis revealed high reliability among variables, low to strong CV, and moderate DV. The study outcomes provide initial convergent and discriminant validity evidence for this measure of PA in adults with ID.

Intellectual and developmental disabilities, 2014 · doi:10.1352/1934-9556-52.4.273