If two observers record the same data and their observations match most of the time, what statistic is demonstrated?

• A. Latency
• B. IOA
• C. Duration
• D. IRT

Answer: (C)

When data come from more than one observer and they match most of the time, what you’re seeing is interobserver agreement, abbreviated IOA. This statistic shows how consistently observers are recording the target behavior or event, which is essential for trusting the data.

IOA is calculated by comparing the observers’ data point by point (or interval by interval) and computing the proportion of agreements. A simple way to think about it is: agreements divided by the total number of observations, then multiplied by 100 to get a percentage. For example, if you have 20 opportunities and the observers agree on 18 of them, IOA is 90%.

Why this matters: high IOA means the data are reliable and not just a reflection of one person’s recording style. If IOA is low, you’d review definitions, train observers more, or adjust the data collection method to improve consistency.

The other timing-related measures—latency (time to start a response), duration (how long a response lasts), and IRT (interresponse time, the time between responses)—are about the characteristics or timing of the behavior itself, not about how well two observers agree in recording it.