Caregivers’ Agreement and Validity of Indirect Functional Analysis: A Cross Cultural Evaluation Across Multiple Problem Behavior Topographies

Abstract

The Motivation Assessment Scale is an aid for hypothesis-driven functional analysis. This study presents its Spanish cross-cultural validation while examining psychometric attributes not yet explored. The study sample comprised 80 primary caregivers of children with autism. Acceptability, scaling assumptions, internal consistency, factor structure, inter-assessor reliability and agreement, and known-group validity analyses were performed. Scaling assumptions, internal consistency (Cronbach alpha of 0.75) and factor structure were satisfactory other than for the Escape domain which demonstrated low internal consistency (0.65), inadequate scaling assumptions (multitrait analysis, 50% success rate) and did not constitute a separate factor. Caregivers' agreement for the primary function reached 73.9% and known group-validity hypotheses across behavior topographies were partially met. The clinical appropriateness of the scale is discussed.

A review of research on direct-care staff data collection regarding the severity and function of challenging behavior in individuals with intellectual and developmental disabilities

“…The information derived from descriptive and indirect assessments can be useful when forming hypotheses about behavior function and can assist in developing test conditions for an experimental analysis but cannot be used to reliably determine behavior function. (Hall, 2005; Virgués-Ortega et al, 2010). Indirect and descriptive assessments can be used to support experimental analysis outcomes and to learn important information that can lead to more successful and efficiently executed experimental analyses (Herzinger and Campbell, 2006; Virgués-Ortega, et al, 2010; Wightman et al, 2013).…”

“…(Hall, 2005; Virgués-Ortega et al, 2010). Indirect and descriptive assessments can be used to support experimental analysis outcomes and to learn important information that can lead to more successful and efficiently executed experimental analyses (Herzinger and Campbell, 2006; Virgués-Ortega, et al, 2010; Wightman et al, 2013). Furthermore, strengths of such assessments include greater validity, efficient administration, minimal cost, and an organized method for providing parents, teachers, and caregivers information on the challenging behaviors without specialized training (Hall, 2005; Herzinger and Campbell, 2006; Virgués-Ortega et al, 2010; Wightman et al, 2013).…”

“…Despite the utility of a multimodal assessment approach, research has shown a low consistency between multiple measures of assessment (Alter et al, 2008; Fahrenholz, 2004; Hall, 2005; Tarbox et al, 2009), particularly between indirect assessments and experimental functional analyses outcomes. Additionally, inconsistencies have been reported between indirect assessments and direct observations (Alter et al, 2008; Virgués-Ortega et al, 2010). In addition to the inconsistency between assessments, research has shown some staff to be ill-prepared and lack the necessary skills and time to conduct functional behavior assessments (McCahill et al, 2014; Mortenson et al, 2008; Newcomer and Lewis, 2004).…”

“…Furthermore, indirect and descriptive behavior assessments are less desirable as they have low interrater reliability and can be biased with respect to perceived function of behavior (Virgués-Ortega et al, 2010) since outcomes are influenced by the perceptions and opinions of those completing them (Alter et al, 2008; Herzinger and Campbell, 2006). Responses affected by the perceptions of those reporting may present differently depending on the environment a reporter interacts with the individual and may not provide an exclusive look at the specified behaviors targeted for intervention (Alter et al, 2008).…”

“…Experimental analysis methodology, while considered the gold standard for determining behavior function, has also received many criticisms that impact staff’s ability to implement the technique. Experimental analyses have been reported to be time-consuming; require extensive training for design, data collection, and interpreting results; encourage challenging behavior through test conditions meant to evoke the behaviors, procedurally are not cost-effective, and finally, experimental analyses require a great deal of resources and are complex and difficult to implement (Hall, 2005; Herzinger and Campbell, 2006; Virgués-Ortega, et al, 2010). The usability of experimental analyses are further challenged when considering the need for a controlled environment (often not feasible in applied settings) and the danger in challenging behaviors themselves—especially when behaviors are severe and require immediate attention (Herzinger and Campbell, 2006; Wightman et al, 2013).…”