What is Interobserver Agreement (IOA)?
Interobserver agreement (IOA) refers to the degree to which two or more independent observers report the same observed values after measuring the same events. For behaviour analysts and educators, it's a standard check on the reliability of data collection. High agreement is a good sign that the target behaviour is clearly defined and that observers are applying that definition consistently — though on its own it doesn't prove the data are accurate, since two observers can agree and both be wrong.
Why is IOA Important?
Without reliable data, it's hard to know whether an intervention is actually working. IOA gives you a quantitative measure of how consistently a behaviour is being recorded. It helps surface observer drift, highlights ambiguities in operational definitions, and is often a requirement for publication in peer-reviewed journals.
How to calculate IOA
Paste each observer's record into the calculator above, choose the method that matches how the data were collected, and it returns the agreement percentage with a clinical interpretation. Reliable agreement starts before the observation, though: both observers need the same operational definition of the target behaviour. If two people disagree on what counts, a low score is telling you the definition is ambiguous — not that the behaviour is hard to see.
Methods this calculator uses
Total Count (Frequency)
For event/frequency data: divide the smaller count by the larger and multiply by 100. Quick, but it can hide disagreements — two observers can reach the same total while recording different events.
Interval-by-Interval
For interval recording and momentary time sampling, such as data from the interval recording timer. It compares the two records one interval at a time and divides agreements by the total number of intervals — more precise than a total count.
Partial Interval
Mathematically identical to interval-by-interval, but used when behaviour is scored if it occurs at any point during an interval. The calculator processes it interval by interval so you can see exactly where observers diverged.
Frequently asked questions
What is a good IOA score?
The conventional benchmark is 80% or above. Below that, treat the data as unreliable and revisit the operational definition or observer training before drawing any conclusions.
What does a high IOA score indicate?
A high IOA score — typically 90% or above — indicates that two independent observers recorded the same behaviour consistently: the operational definition is clear and the observers are applying it the same way. It does not by itself prove the data are accurate; both observers could share the same bias, so high agreement is a check on reliability, not validity.
How do I calculate IOA by hand?
For total count, divide the smaller count by the larger and multiply by 100. For interval data, divide the number of agreements by the total number of intervals and multiply by 100. The calculator above does both and shows the interval-by-interval breakdown so you can see where observers diverged.
Which IOA method should I use?
Match the method to your data: total count for simple frequency tallies, interval-by-interval for interval and time-sampling data, and partial interval when you scored whether the behaviour occurred at any point within each interval.