Introduction
Thank you for all those who took the time to send me their results of the online counting training exercise. While I have not been involved with counting for very long, I have heard that people with different experience in counting come up with different results to those who do not have the same experience. For instance, experienced counters tend to under estimate the numbers, and some counting projects rely on averages from several people counting the same flock to get better accuracy.
Methods
Five people responded to the training by sending me their results. These included four experienced counters and one person who had not been exposed to counting flocks of birds before. Percentage error was calculated for each count for each person and then averaged for that person. Average error was then compared between experienced and inexperienced counters.
Results
The average range of error for the observers was a 24% underestimation to a 17% over estimation. When all the counts were averaged per person, the experienced counters had a underestimation of 6% and the inexperienced counter an underesimtation of 3%. The variation between individuals between the various exercises can be seen in the graph below. It is easy to the variation around the 0 line in the middle of the graph for the different count estimates in each exercise. Click on the graph below to view a larger image.
Discussion
The sample so far is too small to compare experienced vs inexperienced estimations. The good news is that when all the counts from various people are averaged, the accuracy was very good. Lets hope that this is reflection of the accuracy of the counts that we are getting from the field for the Migrating Kestrel Project.
Online Practical training
If you stumbled across this page and have not done the practical training on this web site, why not give your estimates and email the results to me?