In recent years, rather than being used to assess the potential function of a response, descriptive assessment methods have been applied to evaluate potential consequences or contingencies for problem behavior (Borrero, Woods, Borrero, Masler, & Lesser in Journal of Applied Behavior Analysis, 43, 71-88. doi: 10.1901/jaba.2010.43-71, 2010) or to assist with designing baseline conditions to approximate caregiver behavior (Casey et al. in Behavior Modification, 33, 537-558. doi: 10.1177/0145445509341457, 2009). It has been shown that descriptive assessments of some forms of problem behavior (e.g., self-injury, aggression) are not good indicators of behavioral function and should not be used exclusively when conducting functional behavior assessments (Thompson & Iwata in Journal of Applied Behavior Analysis, 40, 333-338. doi: 10.1901/jaba.2007.56.06/epdf, 2007). However, the extent to which descriptive assessments of inappropriate mealtime behavior can predict behavioral function is not yet clear. We conducted descriptive assessments of inappropriate mealtime behavior and compared the results to functional analyses for ten children with severe food refusal. Results showed that, for 71 % of participants, the descriptive and functional analyses matched. These results suggest that the correspondence between descriptive and functional analyses, at least for inappropriate mealtime behavior, may be higher than that for other forms of problem behavior.
Keywords: Conditional probabilities; Descriptive analysis; Descriptive assessment; Feeding; Inappropriate mealtime behavior; Pediatric food refusal.