Assessment & Research

Agreement of function across methods used in school-based functional assessment with preadolescent and adolescent students.

Kwak et al. (2004) · Behavior modification 2004
★ The Verdict

Teacher, student, and observer functional-assessment sources rarely agree—triangulate with direct observation before treatment.

✓ Read this if BCBAs writing plans for preadolescent or adolescent students in general-ed classrooms.
✗ Skip if Clinicians who only run clinic-based functional analyses.

01Research in Context

01

What this study did

The researchers asked teachers, students, and trained observers to name the reason for problem behavior in middle-school kids.

They compared the answers to see if everyone picked the same function and ranked the causes in the same order.

02

What they found

Agreement was low. The adults and the kids rarely chose the same main function or the same rank order.

Because the stories did not match, the team warned that one source alone can mislead treatment.

03

How this fits with other research

Contreras et al. (2023) found the same split decision: descriptive assessments match functional analyses only half the time.

Farrant et al. (1998) seems to disagree. They got big behavior drops when teachers ran classroom-based FBAs. The key is that A et al. added brief tests and direct observation, not just interviews.

Saini et al. (2020) adds that open-ended caregiver interviews boost rater agreement but still miss half of the functions later proven by analysis.

04

Why it matters

Low agreement is the norm, not a fluke. Collect at least two sources—interview plus brief direct observation—before you write a behavior plan. If the stories clash, run a short functional analysis in the classroom. This extra step keeps you from chasing the wrong function and saves weeks of off-target treatment.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Add a 10-minute ABC observation to every teacher interview; treat only when the stories match.

02At a glance

Intervention
functional behavior assessment
Design
single case other
Sample size
19
Population
not specified
Finding
negative

03Original abstract

As we begin to apply functional assessment procedures in mainstream educational settings, there is a need to explore options for identifying behavior function that are not only effective but efficient and practical for school personnel to employ. Attempts to simplify the functional assessment process are evidenced by the development of informant assessment measures (e.g., interviews, rating scales). In this study, the agreement (i.e., on relative rankings and primary function) across sources of information regarding behavior function was examined for 19 students in a middle school setting. These measures included teacher ratings, student ratings, student interviews, observer ratings, and conditional probabilities. In addition, for 1 student, whether information obtained through these sources was consistent with that obtained through a brief analog analysis of function was examined. Results indicated low agreement regarding rank order of behavior function and on primary function across all sources of information.

Behavior modification, 2004 · doi:10.1177/0145445503258990