A.4 Entering data and updating graphs

What the outline says

Task A.4 in content area A, Data Collection and Graphing (13 of the 75 scored questions): "Enter data and update graphs."

Why graphs matter

A data sheet tells you what happened in one session. A graph shows what is happening across many sessions. Your supervisor looks at the graph to see whether a skill is improving, whether problem behavior is decreasing, and whether the plan needs to change. That decision is the supervisor's, but it can only be as good as the data you enter.

Accurate documentation is part of the RBT role. The RBT Ethics Code asks RBTs to accurately complete all required documentation, including client data, and to correct inaccurate information they have submitted to a supervisor or employer.

Parts of a line graph

Most session data in ABA are shown on a line graph. Learn its parts by name, because exam items and supervisors use these terms:

• Horizontal (x) axis: time, usually sessions or dates.
• Vertical (y) axis: the measure of behavior, such as count, rate, minutes or percentage correct. Its label tells you what each point means.
• Data point: one session's value, placed above that session at the height of its value.
• Data path: the line joining data points within one condition.
• Phase change line: a vertical line between the last session of one condition and the first session of the next.
• Condition labels: names such as "Baseline" and "Intervention" written above each phase.
• Legend or labels: used when more than one data path appears, each with its own marker.

Data paths are never connected across a phase change line. Each condition gets its own data path, so the reader can compare conditions without the line blending them together.

Entering data step by step

• Check the data sheet: right client, right target, right date, and every trial or interval filled in.
• Read the vertical-axis label and enter the measure the graph calls for. If the graph shows percentage correct and your sheet shows 8 correct out of 10 trials, enter 80%. If it shows rate and the session ran 20 minutes with 10 responses, enter 0.5 per minute.
• Place the point at the next session and at its value, and connect it to the previous point in the same condition.
• Compare what you entered against the data sheet before you close the file.
• Enter data as soon as you can after the session, following your organization's timelines.
• Use only approved devices and systems, and keep others from seeing client information while you work.

If the supervisor tells you a new condition starts, the phase change line goes in the space between the last session of the old condition and the first session of the new one, with a label for the new phase. You add it when directed; you do not start a new phase on your own.

Missing, unusual and wrong data

A zero is a real data point. It says the behavior was measured and did not occur. If a session was cancelled, or you did not record data, do not enter a zero, an estimate, an average, or the previous session's value. Leave that session without a data point and tell your supervisor what happened.

If a value looks very different from the rest, first check it against the data sheet. If it matches, it is real data and stays on the graph, and you report it. Never delete, average or redraw a point because it looks out of place.

If you find an error you made, correct it through your organization's procedure and tell your supervisor. Never change the original data sheet to match a graph. If a value does not fit the graph, for example 25 responses on an axis that stops at 20, tell the supervisor rather than plotting it at the top of the axis.

Other common displays

A bar graph compares summary values across categories, such as the mean number of tantrums in baseline and in intervention. It does not show session-by-session change the way a line graph does.

A cumulative record adds each session's count to a running total. Its data path can rise or stay flat but never goes down; a session with no responses leaves the line level. A steeper rise means more responses in less time.

How this connects to other tasks

Entering data depends on collecting it correctly (A.1 to A.3) and calculating summaries such as rate and percentage (A.6). The graphs you update are the ones you and your supervisor read for trends (A.7). Inaccurate entries are one of the risks of unreliable data (A.8), and reporting what you notice to your supervisor ties into documentation and reporting (Domain E).

Common mistakes

• Entering a zero for a session that did not happen. A zero means the behavior was measured and did not occur; leave the session blank and report why.
• Entering a count when the graph shows a percentage or rate. Read the vertical-axis label before every entry.
• Filling in missed data from memory. If data were not recorded, say so; an estimate is not data.
• Deleting or averaging an unusual point. Check it against the data sheet; if it matches, it stays.
• Connecting data points across a phase change line.
• Saving up a week of data to enter at once. Late entries leave the supervisor working from an out-of-date graph.

How it's tested

Expect short scenarios: a cancelled session, a typing error, a value that needs converting before it is plotted, or a question about where a phase change line goes. Some items show a graph and ask what is wrong with it. The right answer keeps the data accurate and honest, matches the measure the graph calls for, and leaves decisions about phases and plan changes to the supervisor.