Assessment & Research

Preference testing: a comparison of two presentation methods.

Windsor et al. (1994) · Research in developmental disabilities 1994
★ The Verdict

Paired preference tests give steadier choices than group arrays, even if they take longer.

✓ Read this if BCBAs running preference assessments in schools or clinics for kids with severe delays.
✗ Skip if Teams only working with verbal teens who can self-report likes.

01Research in Context

01

What this study did

The team tested two ways to find out what kids like. They worked with the children who had severe to profound delays. Each child tried both the paired method (two items at a time) and the group method (six items at once). The sessions alternated so every child used both formats.

They measured how often the same top choice came up across trials. This told them which method gave steadier results.

02

What they found

Paired testing won. When kids saw only two items at a time, their favorite stayed the same far more often. The group method made picks jump around. Even though paired took longer, it gave clearer answers.

03

How this fits with other research

MacNaul et al. (2021) looked at 20 studies and said the same thing: paired-stimulus keeps working best when you re-check likes every 8-30 days. This 1994 paper is one of the bricks in that wall.

Bigwood et al. (2026) took the paired idea and tweaked it for adults with dementia. They slowed the pace and added mood checks. The item choices stayed the same, but people felt happier. This shows the paired format can travel across ages and needs.

Cymbal et al. (2020) tested a different tool (the PDC-HS), yet the spirit is the same: pick the format that gives steady data.

04

Why it matters

If you need rock-solid reinforcer lists for kids with severe delays, use paired preference tests. The extra minutes pay off in steady data. You can still borrow tricks from Bigwood et al. (2026) like slowing down or checking smiles if the child seems stressed.

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

Run one paired preference session instead of a six-item array and track if the top pick stays the same across three trials.

02At a glance

Intervention
preference assessment
Design
alternating treatments
Sample size
8
Population
developmental delay
Finding
positive
Magnitude
small

03Original abstract

Paired and group presentation methods of preference testing were compared with eight learners with severe-profound developmental disabilities. Each presentation method was also compared with staff rankings of learners' preferences. Similar preferences were identified with both presentation methods. Although the paired presentation took more time to administer, it elicited more consistent preference information than the group presentation. Staff preference rankings were not highly correlated with either the group or paired presentation. However, items identified as most preferred by staff and by both presentation methods were similar.

Research in developmental disabilities, 1994 · doi:10.1016/0891-4222(94)90028-0