Comparison of the predictive validity and consistency among preference assessment procedures: a review of the literature.

Kang et al. (2013) read 14 studies that compared different ways to find reinforcers. They looked only at people with developmental delays. They asked which methods give the same top items over time and which items really work as reinforcers.

The team coded how each study ran the test, how many items it used, and how well it predicted later reinforcer strength.

Single-stimulus and paired-stimulus tests usually picked the same high-preference items. Free-operat tests often picked different favorites. When studies ran a quick reinforcer check, items from MSW or paired formats worked best.

Longer sessions and pictures instead of real items lowered the match between the test and later reinforcer strength.

Kodak et al. (2009) found the same split: MSW and free-operat lists did not line up for half their kids. Soyeon’s review pools that data and shows the mismatch is common.

Conine et al. (2019) showed kids now pick leisure items more than older papers reported. Soyeon’s review did not include this shift, so you may want to test both edibles and leisure today.

Villafaña et al. (2023) tested picture cards for food-avoiding kids and got clean matches to real edible tests. Soyeon warned pictures can lower validity, but the new study shows that adding a quick taste sample fixes the problem.

Pick one format and stick with it. Use paired-stimulus or MSW for the most reliable reinforcer picks. If you must use pictures, give a brief taste or touch sample right after the choice. Re-test every few weeks and always run a short reinforcer probe before starting teaching programs.