Rate
Rate is the number of times a behavior occurs divided by the length of the observation, such as responses per minute or per hour.
Rate turns a count into a measure that can be compared across sessions of different lengths. The formula is: rate = count ÷ observation time. If a client asks for help 6 times in a 30-minute session, the rate is 6 ÷ 30 = 0.2 responses per minute, or 12 per hour. The unit of time (per minute, per hour, per session day) must always be stated with the number.
RBTs collect the count and record the start and end time of the observation; the rate can then be calculated by hand or by the data system. Rate is the preferred way to report a count whenever observation length varies, for example when one session is cut short because a client goes home sick. Comparing raw counts in that situation would make behavior look lower on the short day just because there was less time for it to happen.
When the number of opportunities is controlled by the teacher, as in discrete-trial teaching where the RBT presents a set number of trials, rate is less useful and percentage correct is usually reported instead. Rate also inherits the limits of counting: it does not fit behaviors that last a long or variable time. Because some sources use "frequency" and "rate" interchangeably, the safest habit is to report both the count and the time.
Example and non-example
| Example | A client hits 9 times in a 45-minute session on Monday and 6 times in a 15-minute session on Tuesday. The rates are 0.2 per minute and 0.4 per minute, so Tuesday's rate was higher even though the count was lower. |
|---|---|
| Non-example | Recording that a student answered 8 of 10 questions correctly is a percentage, not a rate, because it divides by opportunities rather than by time. |
Don't confuse it with
| Frequency | Frequency is the raw count; rate is that count divided by the time observed. |
|---|---|
| Percentage | Percentage divides by the number of opportunities or the total; rate divides by time. |
Test yourself
-
Question 1A client's sessions vary from 30 to 90 minutes. The supervisor wants to compare hitting across sessions. Besides counting each hit, what must the RBT record?
Show answer
A: The length of time observed in each session. A count from a 90-minute session cannot be fairly compared with a count from a 30-minute session. Recording the observation time lets the count be turned into a rate (count divided by time), which can be compared.
- B: Knowing who was nearby can be useful context, but it does not make counts from sessions of different lengths comparable.
- C: Percentage of intervals comes from a discontinuous procedure; it is not something added to a count to make sessions comparable.
- D: Where a hit lands describes its form, not how often hitting occurs over time.
Task A.1 · 40-hour curriculum section C
-
Question 2An RBT counts 18 mands during a 30-minute session. What is the rate of manding?
Show answer
B: 0.6 mands per minute. Rate is the count divided by the observation time: 18 mands ÷ 30 minutes = 0.6 mands per minute (the same as 36 per hour).
- A: 0.3 would result from dividing 18 by 60, but the session lasted 30 minutes.
- C: 1.67 is 30 divided by 18, which is minutes per mand, not mands per minute.
- D: 18 is the total count for the whole session, not the count per minute.
Task A.1 · 40-hour curriculum section C
-
Question 3An RBT is counting a client's out-of-seat behavior during a 60-minute session but must leave the room for 10 minutes to help in an emergency. Another staff member stays with the client but takes no data. What should the RBT do?
Show answer
B: Record the count and note that only 50 minutes were observed. Data must reflect what was actually observed. Recording the true observation time keeps the count accurate and lets it be turned into a correct rate; the gap is documented, not filled in.
- A: Estimating the missed period means recording numbers that were never observed.
- C: Treating 50 minutes as 60 makes the rate look lower than it was and misrepresents the record.
- D: Leaving the sheet blank with no note loses the data that were collected and leaves the gap unexplained.
Task A.1 · 40-hour curriculum section C · RBT Ethics Code 2.02
Answer every question to see your score.
On the RBT exam
This term belongs to task A.1 Continuous measurement in content area A, Data Collection and Graphing, which is 13 of the 75 scored questions.