Functional Behavior Assessments: A Comparison of Across Three Assessment Methodologies

Background:  

Children with autism may develop problem behaviors such aggression or self-injury that can result in injury to themselves or others (Kanne & Mazurek, 2007; Weiss, 2003). Identifying the consequences responsible for maintaining problem behavior through functional behavioral assessment (FBA) is a critical aspect of successful treatment of problem behavior (Hanley, Iwata, & McCord, 2003). The functional analysis (FA) developed by Iwata, Dorsey, Slifer, Bauman, and Richman (1982/1994) represents one of the most thorough assessments for identifying the function of problem behavior. In an analog FA, antecedent and consequence variables hypothesized to evoke and maintain problem behavior are directly manipulated. Although the FA methodology represents a rigorous demonstration of function, it is a lengthy and resource intensive approach.  Abbreviated functional analyses, such as the Brief Functional Analysis (BFA; Northup et al., 1991), and indirect functional assessments, such as the Questions About Behavioral Function (QABF; Matson & Vollmer, 1995) rating scale, have been developed as less intensive FBA formats. Generally, indirect functional assessments and BFAs are considered to be less valid than analog FAs (Wacker et al., 2004). However, little direct research has been conducted on how well the results of these three FBA formats correspond with one another.

Objectives:  

The purpose of the current investigation was to examine the correspondence between three formats of functional behavior assessments: FA, BFA, and the QABF.

Methods:  

Ten participants who were admitted to an intensive day-treatment program for the assessment and treatment of problem behavior participated in this study. Prior to the admission, caregivers for each participant completed a QABF. This assessment consisted of a series of questions regarding the potential function of their child’s problem behavior. Each question was assigned a score and grouped into a behavioral function (i.e., attention, escape, automatic, or tangible). During the admission to the day-treatment program, a BFA was conducted with each participant. Upon completion of the BFA, a FA was conducted. Conditions in the FA were similar to the BFA with the exception that a staff member served as the therapist.

Participant responses to the QABF were scored using the methods outlined in that measure’s instruction manual.  Results of the BFA and FA were determined by graphing the rates of problem behavior observed during test and control conditions from each assessment into separate line graphs. Graphs were distributed to 7 naïve clinicians who had expertise in interpreting such graphs for clinical purposes.  These clinicians indicated the function of problem behavior identified in each graph using a standardized rating scale.

Results:  

The correspondence between the QABF and BFA, QABF and FA, and BFA and FA was 39%, 41%, and 65% respectively.

Conclusions:  

The results of the current study indicate that the FBAs that utilize direct measures (i.e., BFA and the analogue FA) had greater correspondence as compared to the indirect measure. Additionally, results suggest that the QABF is more likely to identify multiple functions than the direct measures, increasing the opportunity for false positive results.