methods

We rewatched the videos and choose a 5 minute period and 10 behaviors that occurred in that time frame for a reliability analysis. One group member watched the 5 minutes worth of video and used Jwatcher to indicate the times at which each behavior took place. This member rewatched the same five minutes of video and used Jwatcher to indicate the times at which each behavior occurred a second time. The reliability was determined between the two data sets and we redid the indication of when each behavior occurred until the reliability was above 65%. A second group member watched the same 5 minutes of video and indicated when behaviors occurred using Jwatcher. This was redone until there was above a 65% reliability between the first and second group member.