Beyond Filling in the Blanks: Thinking Like a Data Analyst as a BCBA

What is a data analyst?

A data analyst is a person who organizes, examines, and interprets data to identify patterns and help make decisions. A data analyst turns data into meaningful information that can guide those decisions. Behavior analysts (supervisors) are not data collectors. We are data analysts and clinical decision-makers!

A completed form is not an analysis. Every BCBA has seen it: the FBA template is filled in, the ABC boxes are checked, the graph is updated, and the session note says, "continue current program." Every field has an answer. But was anything actually analyzed?

The forms we use are tools, not thinking. A template can prompt you to write down an antecedent, but it cannot tell you whether that antecedent matters. A graph can show a line going up or down, but it cannot tell you why. That part is our job as behavior analysts!

At BH Field, we believe the most important skill a BCBA brings to a case is not the ability to complete documentation. It is the ability to think like a data analyst: to question, to look for patterns, to test explanations, and to change course when the evidence says so.

The second "A" in ABA stands for Analysis

Our field is called Applied Behavior Analysis for a reason. Collecting data is not the same as interpreting it, and graphing is not the same as deciding. The clinical process should run all the way through:

Assessment → Hypothesis → Intervention → Data → Analysis → Clinical Decision

Too often, the process stops at the graph. The data are collected faithfully, plotted neatly, and then filed. When that happens, a learner can spend months in a program that stopped working weeks ago, or in one that never addressed the real function of the behavior.

A data analyst does not just record numbers. A data analyst asks what the numbers mean, what else could explain them, and what should happen next. That is the standard our learners and families deserve.

20 skills that separate analysis from paperwork

Thinking like a behavior analyst is a set of skills you can practice and develop. We group them into five habits of mind.

1. See clearly before you explain

• Objective observation. Separate what you saw from what you concluded. "He refused because he was angry" is an interpretation. "He pushed the worksheet off the table and said 'no' within 5 seconds of the instruction" is an observation.
• Critical thinking. Before accepting the first explanation, ask: What do I actually know? What am I assuming? What evidence supports my conclusion?
• Detecting bias. Watch for confirmation bias, anchoring, prior diagnoses, caregiver narratives, and your own expectations. They quietly shape what you notice and what you ignore.
• Knowing when you don't know. One of the strongest analytic skills is recognizing that the information is insufficient for a defensible conclusion, and collecting more before acting.

2. Think functionally, not in forms

• Functional thinking. Look past what the behavior looks like and ask what maintains it. Two identical behaviors can serve completely different functions.
• Identifying relevant variables. Consider antecedents, consequences, motivating operations, reinforcement history, setting events, environmental changes, skill deficits, caregiver responses, and treatment integrity.
• Skill deficit vs. performance deficit. Can the learner not do it, or can they do it but the environment is not supporting it? The answer changes the entire intervention.
• Conceptually systematic thinking. You should be able to explain why an intervention should work in behavioral terms. "This is the program we normally use" is not a rationale.

3. Read the data like an analyst

• Data interpretation, not just collection. Look at a graph and ask: What is changing? How much? How consistently? Is the change clinically meaningful? What happened when the intervention changed?
• Pattern recognition. Look for meaningful patterns across time, people, settings, conditions, and behaviors, rather than reacting to one bad session.
• Correlation vs. causation. Two things happening together does not mean one caused the other. A drop in behavior the same week a new program started is a lead, not proof.
• Integrating multiple sources. Combine direct observation, assessment results, caregiver and staff report, treatment data, and history. No single score or source tells the whole story.

4. Treat every explanation as a hypothesis

• Hypothesis development. Build a reasonable explanation from the evidence, and remember a hypothesis is not a fact.
• Hypothesis testing. Ask what evidence would support or contradict it, then use systematic observation, assessment, or a treatment change to find out.
• Parsimonious reasoning. Start with the simplest explanation that accounts for the evidence before reaching for complicated ones.
• Considering alternative explanations. Always ask, "What else could explain what I am seeing?" An apparent treatment failure could reflect poor treatment integrity, weak reinforcers, shifting motivation, too few opportunities to respond, a poorly defined target, environmental changes, or an incorrect functional hypothesis.

5. Decide, and keep deciding

• Clinical decision-making. Turn assessment and data into action. The analysis is only finished when it produces a decision.
• Cognitive flexibility. Good clinical judgment is not defending the original plan. It is responding to new information.
• Generalization thinking. Don't stop at "Can the learner do it in therapy?" Ask whether it happens with different people, materials, settings, and natural contingencies.
• Social validity and clinical relevance. Not everything measurable needs treatment. Ask whether the target meaningfully improves communication, independence, safety, relationships, access, autonomy, or quality of life.

What this looks like on a real case

Imagine a learner whose aggression has been flat for six weeks on an escape-maintained protocol. The fill-in-the-blank response: "Behavior remains stable. Continue current program. Will monitor."

The analytic response takes more effort, but it makes the supervisor ask better questions, and it gets learners unstuck:

• Is the plan actually being implemented? Check treatment integrity data and watch a session before blaming the protocol.
• Is the reinforcer still working? Motivation shifts. A break that was valuable in March may not be in May.
• Is the function right? Look at the ABC data again with fresh eyes. Is there evidence of attention or tangible access we explained away?
• Did something else change? New staff, a new classroom, medication, sleep, a family event.
• Is the target well defined? If two staff would score the same episode differently, the data cannot be trusted.
• What would prove me wrong? Decide what result would change the plan, then go find out.
• Is the RBT collecting the data correctly? You might need to retrain the RBT, change the definition of the behavior, or run IOA.

Questions to bring to every data review and documentation

• What do I actually know, and what am I assuming?
• What is changing, by how much, and how consistently?
• What else could explain this?
• What evidence would contradict my hypothesis?
• Is this change clinically meaningful for this learner's life?
• Do I have enough information to decide, or do I need more?

The BH Field analytic mindset

Think about the consequences of your own behavior: how every clinical decision will affect a client who has medical necessity and multiple impairments. Think smart. Think as a behavior analyst. Ask, "What if…?" If you remember nothing else, remember these six lines:

• Observe before interpreting.
• Ask before assuming.
• Analyze before intervening.
• Measure before concluding.
• Question your own hypothesis.
• Change your clinical decision when the data tell you to.

The next time you open a template, treat every blank as a question to investigate, not a box to fill. Your learners are not paperwork. Their progress and overall outcomes depend on how well you think.