 Total duration inner observer agreement this one's different than total count because you're measuring a duration But you still take the total duration during the during the measurement period Of one observer and divided by the total duration during the interval period during the measurement period by the second observer It's horrifically inaccurate as well again That's why we have intervals and so you can break them down and make it more accurate to find out how well people are Doing so this isn't another one that you should probably not use but we have to talk about it anyway Because sometimes that's all you got right and I get that and we all get that you just do the best You can with what you have but this particular method Grossly overestimates accuracy of inner observer agreement just like the total count does so again total duration You're just taking the the shortest duration divided by the longest duration You know the total durations right between the two observers the shortest one divided by the longest one gets your ratio Multiply 100 you got yourself a percentage of agreement not that it's very accurate