Comparisons of Functional Behavior Assessment Procedures to the Functional Analysis of Problem Behavior

Romani, P. W., Luehring, M. C., Hays, T. M., & Boorse, A. L. (2023). Comparisons of functional behavior assessment procedures to the functional analysis of problem behavior. Behavior Analysis: Research and Practice, 23(1), 36–48. https://doi.org/10.1037/bar0000258

Reviewed by
Angela Fuhrmann-Knowles, MA, BCBA, Endicott College

Why research this topic?

Procedures to teach specific skills for individuals with autism Functional behavior assessments (FBA) help determine the function(s) of behavior (i.e., why they occur). Identifying the function of a behavior allows for function-based treatment, meaning the treatment will help a learner meet their needs in a more desirable or safe manner. Research has demonstrated that function-based treatment significantly reduces problem behaviors (Carr & Durand, 1985; Hanley et al., 2003).

FBAs include indirect, descriptive, and experimental assessments. Indirect methods entail collecting information about behaviors through interviews, checklists, or questionnaires without directly observing them. Descriptive assessments involve direct observations of the behavior, in which an observer notes the events that occur before (i.e., antecedents) and after the behavior (i.e., consequences). From this information, patterns that influence or maintain the behavior are identified. Experimental assessment methods, such as the experimental functional analysis (EFA), involve measuring the behavior in the presence or absence of reinforcers, or environmental cues. During an EFA, a learner’s behavior is observed under different arranged conditions, which helps identify what causes it. The EFA is designed using the information gathered from indirect and direct assessment methods.

EFAs are more accurate and reliable for identifying behavior functions than indirect or descriptive methods but are not always used (Colombo et al., 2020; Oliver et al., 2015). Their underutilization means that treatment may not be function-based, which could lead to poorer outcomes. Although FBAs and EFAs can be used in various settings, Romani et al. (2023) highlight the underutilization of EFAs in school settings. Multiple barriers exist to implementing EFAs in school settings, including the amount of training, resources, and expertise required. Research indicates that school professionals may primarily use indirect and descriptive assessment methods (Oliver et al., 2015; Van Acker et al., 2005). Students with challenging behaviors are required to receive an FBA per the Individuals with Disabilities Education Act (IDEA); however, this process varies across school districts. As such, research must compare the accuracy of results from FBA procedures to EFAs.

Thus far, research in this area has produced mixed results, with some studies demonstrating 100% agreement between indirect and descriptive assessments and EFAs (Arndorfer et al., 1994) and others showing poorer agreement (Alter et al., 2008; Tarbox et al., 2009). Comparing the results of different FBA procedures helps identify which are most accurate and reliable and creates the opportunity for learners to receive function-based treatment.

What did the researchers do?

This study had two purposes. The first was to replicate previous studies by showing the exact and partial agreement between the results of indirect and descriptive assessments and the EFA. The second was to extend previous studies by evaluating the agreement between results from combinations of indirect and descriptive assessments to those of the EFA.

Ten male and four female children, aged 6-17, participated. The participants attended a treatment program at a psychiatric unit for children that provided psychiatric and behavioral services. Each participant had an autism spectrum disorder (ASD) diagnosis and another developmental or mental health disorder. The children’s problem behaviors included negative statements, property destruction, aggression, perseverative speech, disrobing, and self-injury. All sessions took place in classrooms in the unit.

A doctoral-level Board Certified Behavior Analyst (BCBA-D) conducted the FBA procedures for each participant. Three indirect assessments were used in this study. These included an Antecedent-Behavior-Consequence Interview (ABC-I), in which the authors used the Open-Ended Functional Assessment Interview (Hanley, 2012), Questions About Behavior Function (QABF; Matson & Vollmer, 1995) and the Functional Analysis Screening Tool (FAST; Iwata et al., 2013). On a participant’s first day of admission, the parents completed an interview in an office in the unit or over the phone. Then, the BCBA-D completed the descriptive assessment using the Antecedent-Behavior-Consequence Narrative (ABC-N). Data were collected in 60-minute intervals: during work, unstructured play, and therapeutic groups on the behaviors identified as most concerning in the interviews. Whenever a problem behavior occurred, the BCBA-D recorded the antecedents and consequences.

After completing the FBA procedures, the unit Board Certified Behavior Analyst (BCBA) began the EFAs. The researchers conducted standard EFA conditions as follows (Iwata et al., 1994):

1. Escape: The BCBA issued an academic or daily living instruction. If problem behavior occurred, the participant received a 30-second break from the instruction.
2. Tangible: Access to a highly preferred item was withheld from the participant. If problem behavior occurred, the participant received 30 seconds of access to that item.
3. Attention: The BCBA instructed the participant to play with a low-preferred item while the BCBA engaged in a different activity. If problem behaviors occurred, the participant received 30 seconds of attention.
4. Free play: The participant received free access to highly preferred items and attention. Problem behavior was ignored.

The researchers compared the agreement between the individual FBA assessments and their combinations to the results of the EFA. First, the researchers evaluated the exact agreement between the functions identified in the FBA procedures and those in the EFA. For example, if the ABC-I, ABC-N, and the EFA showed escape as the function of the problem behavior, the researchers considered this an exact match. Second, the researchers evaluated partial agreement. For example, if the ABC-I suggested that escape and attention were the functions of the problem behavior, and the QABF and the EFA showed escape as the function, the researchers considered this a partial match.

What did the researchers find?

The researchers found that the FAST alone, QABF alone, and ABC-N alone demonstrated the highest exact agreement with the EFA at 21.4%. The next highest exact agreement was demonstrated between the FAST and the ABC-N at 14.3%. The combinations with the next highest exact agreement with the EFA at 7.1% were the QABF and ABC-N; QABF and FAST; QABF, FAST, and ABC-N; and the ABC-I alone (see table below). The remaining combinations never exactly matched the EFA.

Assessment	Percentage of Exact Agreement
FAST alone QABF alone ABC-N alone	21.4%
FAST and ABC-N	14.3%
QABF and ABC-N QABF and FAST QABF, FAST, and ABC-N ABC-I alone	7.1%

Six combinations of FBA procedures demonstrated 85.7% partial agreement with the EFA. The combinations included: (1) ABC-I, QABF, and ABC-N; (2) QABF and ABC-N; (3) ABC-I, FAST, and ABC-N; (4) ABC-I and ABC-N; (5) QABF and FAST; and (6) QABF, FAST, and ABC-N. Of the eight remaining combinations, four produced 78.6% partial agreement and four produced less than or equal to 71% partial agreement with the EFA (see table below). These results are like what other studies have demonstrated previously.

Assessment	Percentage of Partial Agreement
ABC-I, QABF, and ABC-N QABF and ABC-N ABC-I, FAST, and ABC-N ABC-I and ABC-N QABF and FAST QABF, FAST, and ABC-N	21.4%
ABC-I, QABF, and FAST ABC-I and QABF ABC-I and FAST ABC-I alone	78.6%
FAST and ABC-N FAST alone QABF alone ABC-N alone	71% or less

What are the strengths of the study?

Two observers collected data on the problem behaviors to assess interobserver agreement (IOA). IOA evaluates the agreement between two or more observers in how they collect data. The IOA scores ranged from 80 to 100%, with an average score of 84.9%, indicating high agreement. Additionally, a BCBA with over 10 years of experience in conducting assessments reviewed the data collected by the BCBA-D to measure IOA on the hypothesized functions of the participants’ behaviors. Upon reviewing the data, if the BCBA hypothesized the same function of behavior as the BCBA-D, this was marked as an agreement. The BCBA agreed with the BCBA-D about the functions of behavior 100% of the time. Another strength is that the procedures are written clearly, making replication easier for researchers. Lastly, Romani et al. (2023) used indirect and descriptive assessments that were validated and commonly used by professionals in everyday practice, making the purpose and results of this study more clinically relevant.

What are the limitations of the study?

The authors discussed some limitations of this study. First, they did not conduct an alone condition, which is used to assess for an automatic function of behavior, even though some of the participants’ FBA results suggested an automatic function. Second, low-preferred rather than high-preferred items were available during the attention condition. Romani et al. (2023) made this decision due to other research that demonstrated that using low-preferred items competed less with learners’ desire for attention from others. Third, uncommon IOA measures were used.

Additionally, IOA data were not collected during the indirect assessments or the ABC-N. The lack of IOA during the assessments is a limitation because, for example, if the BCBA-D and the BCBA collected data during the ABC-N assessment, they might note different antecedents and consequences for the problem behaviors. If different antecedents and consequence events were noted, the two observers might come to differing conclusions about the function of the problem behavior, which could lead to differences in EFA conditions and treatment plan development. Lastly, treatment integrity data were not collected during this study, meaning the accuracy with which the assessments were conducted is unclear.

What do the results mean?

Romani et al. (2023) evaluated the accuracy of standard assessment practices. They demonstrated that although an exact agreement between FBA procedures and EFA results was low, partial agreement was high. This finding means there is likely to be some agreement between FBA procedures and EFA results. For parents, caregivers, and teachers, this finding means that even if an FBA rather than an EFA is conducted for their child, loved one, or student receiving services, there is likely to be some accuracy in identifying the function of their behavior. As noted previously, behavior analysts do not always conduct EFAs due to constraints, such as the time, expertise, and resources required. As such, if a parent, caregiver, or teacher learns their service provider or behavior analyst does not have the resources to conduct EFAs, there can be assurance that the FBA methods used also have the potential to provide answers regarding behavioral function(s).

There are multiple possibilities for future research and takeaways from this study. First, because conditions were not arranged in the EFA to assess automatic reinforcement functions, researchers could not evaluate the accuracy of direct and indirect assessments in identifying automatic functions. To that point, because automatic reinforcement functions were not assessed, the results of this study may not be directly applicable to individuals who engage in behavior that may be automatically maintained. Second, researchers should continue to evaluate the agreement between FBA procedures and EFA results across different participants, settings, and behaviors. Doing so would help evaluate if and how agreement changes across populations in different settings or when using different indirect and descriptive assessment methods.

The findings of this study are valuable for those tasked with identifying behavioral functions and those with a child, a loved one, or a student receiving behavioral services. Although some functional assessment and analysis methods lead to greater accuracy than others, no functional assessment or analysis method is accurate 100% of the time. If an intervention is unsuccessful with one’s child, loved one, or student, one possible reason is that an incorrect function was identified, leading to the development of an intervention that does not address the reasons a behavior is occurring. Importantly, functions of behavior should be viewed as dynamic, meaning that although a behavior can occur for the same reason every time, the function can also change (Ala’i-Rosales et al., 2019; Hanley, 2012; LaBelle & Charlop-Christy, 2002). Romani et al. (2023) demonstrated some of the FBA methods that may produce accuracy like EFAs in identifying behavioral functions, which parents, caregivers, or teachers may find to be a useful resource when working with behavior analysts or service providers. Understanding which FBA procedures lead to the greatest accuracy with EFAs may save time and resources and allow learners to receive effective and function-based treatment.

References

Ala’i-Rosales, S., Cihon, J. H., Currier, T. D., Ferguson, J. L., Leaf, J. B., Leaf, R., McEachin, J., & Weinkauf, S. M. (2019). The big four: Functional assessment research informs preventative behavior analysis. Behavior Analysis in Practice, 12(1), 222-234. https://doi.org/10.1007/s40617-018-00291-9

Alter, P. J., Conroy, M. A., Mancil, G. R., & Haydon, T. (2008). A comparison of functional behavior assessment methodologies with young children: Descriptive methods and functional analysis. Journal of Behavioral Education, 17(2), 200-219. https://doi.org/10.1007/s10864-008-9064-3

Arndorfer, R. E., Miltenberger, R. G., Woster, S. H., Rortvedt, A. K., & Gaffaney, T. (1994). Home-based descriptive and experimental analysis of problem behaviors in children. Topics in Early Childhood Special Education, 14(1), 64-87. https://doi.org/10.1177/027112149401400108

Carr, E. G., & Durand, V. M. (1985). Reducing behavior problems through functional communication training. Journal of Applied Behavior Analysis, 18(2), 111-126. https://doi.org/10.1901/jaba.1985.18-111

LaBelle, C. A., & Charlop-Christy, M. H. (2002). Individualizing functional analysis to assess multiple and changing functions of severe behavior problems in children with autism. Journal of Positive Behavior Interventions, 4(4), 231-241. https://doi.org/10.1177/10983007020040040601

Colombo, R. A., Taylor, R. S., & Hammond, J. L. (2020). State of current training for severe problem behavior: A survey. Behavior Analysis in Practice, 14(1), 11-19. https://doi.org/10.1007/s40617-020-00424-z

Hanley, G. P. (2012). Functional assessment of problem behavior: Dispelling myths, overcoming implementation obstacles, and developing new lore. Behavior Analysis in Practice, 5(1), 54-72. https://doi.org/10.1007/bf03391818

Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem behavior: A review. Journal of Applied Behavior Analysis, 36(2), 147-185. https://doi.org/10.1901/jaba.2003.36-147

Iwata, B. A., DeLeon, I. G., & Roscoe, E. M. (2013). Reliability and validity of the Functional Analysis Screening Tool. Journal of Applied Behavior Analysis, 46(1), 271-284. https://doi.org/10.1002/jaba.31

Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S. (1994). Toward a functional analysis of self-injury. Journal of Applied Behavior Analysis, 27(2), 197-209. https://doi.org/10.1901/jaba.1994.27-197 (Reprinted from Analysis and Intervention in Developmental Disabilities, 2, 3-20, 1982)

Matson, J. L., & Vollmer, T. R. (1995). User’s guide: Questions About behavioral function (QABF). Scientific Publishers, Inc.

Oliver, A. C., Pratt, L. A., & Normand, M. P. (2015). A survey of functional behavior assessment methods used by behavior analysts in practice. Journal of Applied Behavior Analysis, 48(4), 817-829. https://doi.org/10.1002/jaba.256

Tarbox, J., Wilke, A. E., Najdowski, A. C., Findel-Pyles, R. S., Balasanyan, S., Caveney, A. C., Chilingaryan, V., King, D. M., Niehoff, S. M., Slease, K., & Tia, B. (2009). Comparing indirect, descriptive, and experimental functional assessments of challenging behavior in children with autism. Journal of Developmental and Physical Disabilities, 21(6), 493-514. https://doi.org/10.1007/s10882-009-9154-8

Reference for this article:

Fuhrmann-Knowles, A. (2025). Research synopsis: Comparisons of functional behavior assessment procedures to the functional analysis of problem behavior. Science in Autism Treatment, 23(4).

Other Related Research Synopses:

Related ASAT Articles and Reviews:

#BehaviorAnalysts #Centerbased #ChallengingBehavior #EarlyCareerAndStudents