In-home parent training of functional analysis skills.

Abstract

Functional analysis (FA) is recognized as an effective method for determining social and nonsocial influences maintaining problem behavior and subsequently, formulating an effective intervention plan (Hanley, Iwata, & McCord, 2003). Although early FA studies were implemented by experienced research personnel (Iwata, Dorsey, Slifer, Bauman, & Richman, 1994), more recent inquiry has focused on paraprofessional staff and non-behavioral practitioners. For example, Moore, Edwards, Sterling-Turner, Riley, DuBard, & McGeorge (2002) taught classroom teachers to implement FA conditions through modeling, behavior rehearsal, and performance feedback. Similarly, Wallace, Doney, Mintz-Resudek, & Tarbox (2004) showed that educators could learn FA skills using scripts, role playing, and video demonstration. More recently, Phillips and Mudford (2008) combined lecture, written information, modeling, rehearsal, and feedback to train FA methodology with inexperienced residential-care staff. The importance of these studies is that when compared to assessment that is performed under simulated conditions, having natural care-providers conduct FA may capture more precisely the operating contingencies responsible for problem behavior. Parents of children who have problem behavior have participated in FA sessions under professional supervision (Wacker, Berg, & Harding, 2006) but research has not addressed actual skills training with family members. That is, can parents acquire the skills necessary to conduct FA and within their homes? The present study considered this question by evaluating an in-home training program with two families unfamiliar with FA methods and no prior implementation history METHOD Participants and Setting The participants were two sets of parents who had a child with autism. Family A included mother (32 years old), father (38 years old), and their 10-year old son. Family B included mother (36 years old), father (40 years old), and their 12-year old son. The families agreed to participate in the study as a component of consultation they were receiving from a board certified behavior analyst (BCBA). All FA sessions (described below) were conducted in a small area set aside in each family's home. Measurement The participants conducted FA sessions in which a graduate student played the role of a child. Each simulated session lasted 5 minutes and consisted of either the social disapproval, demand, or play condition described by Iwata et al (1994). During sessions, one parent from each family interacted with the student. There were 3 sessions per day (social disapproval, demand, play), implemented in random order, and scheduled over a one-week period. Note that the alone condition of the Iwata et al (1994) methodology was omitted from the study because the child behavior could not occur if the participants were not present. Because the children in each family occasionally hit their parents, the student playing the role of a child displayed aggression as the behavior during FA sessions. The student followed a written script that specified when aggression should occur within each FA condition. The scripts were introduced to the student during two training meetings conducted by the senior author preceding the study. The senior author videotaped each FA session and subsequently, recorded data by viewing the recorded sessions. Using a 30s partial interval recording method, each occurrence of the behavior was scored and whether the participant implemented the appropriate consequence in each FA condition. Implementation accuracy was calculated as a percentage measure by dividing the number of intervals in which the appropriate consequence was demonstrated by the total intervals in which the target behavior occurred and multiplied by 100. Interobserver Agreement Interobserver agreement (IOA) was assessed by having a second observer score data independently with the senior author for 75% of videotaped FA sessions. …