If the target behavior is not observable and it does not make sense to modify the behavior's definition so that it is observable.
If you're faced with an unobservable behavior like "feeling sad" and want to stay in a behavioral framework by only focusing on addressing observable behaviors, then change your behavior's definition to something observable: "feeling sad" becomes "cries and lays head on desk."
However, if the goal is to address the cognitive components of "being sad" (i.e., thoughts and feelings), then you should not do an FBA. FBA is rooted in the principles of applied behavior analysis, which exclusively targets observable behaviors. FBA is one tool in our large educational and social/emotional toolbox. It serves a very specific purpose, and sometimes it will not be the most appropriate choice.
If you have not been trained to conduct one.
Fundamentally, conducting an FBA is a complex skill that requires a theoretical and conceptual understanding of applied behavior analysis, an understanding of data-based decision making, and knowledge of and practical expertise in the assessment practices involved.
If you do not have a solid understanding of ABA principles, have not conducted multiple FBAs with supervision and guidance, and otherwise lack the skills required to conduct an FBA, then you should not do one.
If the behavior in question is not sufficiently important for that student's quality of life or the safety of others that it deserves a considerable amount of resources allocated to addressing it.
In other words, if it makes sense to try something less resource-intensive before doing an FBA, then try the less-intensive thing first.
If your team has a reasonable idea that this student might be engaging in mild disruptive behavior because that behavior generally results in the student obtaining adult attention, and your school has access to a small-group intervention like Check-In-Check-Out which is oriented towards that function of behavior, then it might make sense to first place the student in that small-group intervention and determine whether that was sufficient for changing the mild disruptive behavior.
This recommendation is closely linked to tiered and prevention-oriented decision-making, wherein we typically use less-intrusive means of intervention and determine whether those worked before advancing on to more intrusive means.
That isn't to say that a student must always go through Tier 2 support before having an FBA conducted. Sometimes, a student's behavior may be serious enough (e.g., self-injurious, dangerous to self or others) that an FBA is immediately warranted.