Assessment & Research

The validation of procedures to assess prevocational task preferences in retarded adults.

Mithaug et al. (1978) · Journal of applied behavior analysis 1978
★ The Verdict

A quick object-pairing game gives a stable, valid ranking of prevocational task preferences for adults with severe ID.

✓ Read this if BCBAs writing prevocational or day-hab programs for adults with severe intellectual disability.
✗ Skip if Clinicians who only treat young children or use questionnaire-based preference tools.

01Research in Context

01

What this study did

Researchers worked with the adults who had severe intellectual disabilities. Staff showed each adult two prevocational objects at a time, like a screwdriver or a paintbrush. The adult pointed to or touched the item they wanted.

Staff repeated every possible pair until each person had chosen 66 times. They counted wins to rank the 12 tasks from most to least preferred.

02

What they found

The same adults picked the same top three tasks on three different days. Rankings stayed stable for two weeks. When staff later let them work on any task, people spent the most time with their top-ranked item and little time with the bottom-ranked one.

The simple paired-choice method gave a quick, reliable list of what each adult liked to do.

03

How this fits with other research

Howlin et al. (2006) later turned the same paired-choice idea into a paper checklist for typical high-school and college students. They dropped the hands-on objects and used self-report questions, but the core logic—rank options head-to-head—remained identical.

Frank-Crawford et al. (2018) asked a tougher follow-up question: do these preference ranks still matter when the job gets hard? They found that high-preference items sometimes lost their power once kids had to work harder. The 1978 tool still gives a valid starting list, but you must test whether the top items actually work when demands rise.

Wouters et al. (2017) reviewed fitness tests for youth with ID. Like the 1978 study, they stressed that any good assessment needs solid repeat scores; only a few fitness tools met that bar, just as only paired-choice met it for task preference.

04

Why it matters

If you support adults with severe ID, you can now build a prevocational program around what each person truly likes. Run a 15-minute paired-choice session, create a ranked list, then let data—not guesswork—guide job training. Re-check the list every month and always test whether the top picks still work when the task becomes work, not play.

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

Pick six work tasks, pair them every which way, let your client choose, and tally wins to create the week's preference order.

02At a glance

Intervention
preference assessment
Design
single case other
Sample size
3
Population
intellectual disability
Finding
positive

03Original abstract

Three severely retarded young adults between the ages of 19 and 21 years participated in a prevocational training program, and worked regularly on six different tasks during the scheduled six-hour day. The study attempted to assess each subject's preferences for the six tasks: collating, stuffing, sorting, pulley assembly, flour-sifter assembly, and circuit-board stuffing. In Phase I, the procedure consisted of randomly pairing each task with all other tasks in a two-choice situation that required the subjects to select one task from each pair combination to work for a seven-minute period. The selection procedure consisted of presenting two representative task objects on a tray and requesting the subject to pick up one object and place it on the work table. The object selected represented the task worked for that period. The 15 possible pair combinations were presented randomly every two days for a period of 34 days to determine the preferences. During the validation phase (Phase II), each subject's least- and most-preferred tasks were paired separately with moderately-preferred tasks. As expected, these manipulations confirmed the baseline data, as choices for the moderately-preferred tasks decreased when consistently paired with the preferred tasks and increased when consistently paired with the least-preferred tasks.

Journal of applied behavior analysis, 1978 · doi:10.1901/jaba.1978.11-153