Kamm Matthew matthew.kamm@tufts.edu Reed J Michael 2018 Results from multiple independent observers classifying one aerial image of an American kestrel nest box site in Massachusetts, USA using ENVI dataset Tufts University https://doi.org/10.1594/PANGAEA.884662 2018-01-08T11:01:42 https://doi.org/10.1594/PANGAEA.884669 Five naive observers (graduate students at Tufts University, MA, USA) each classified an aerial photo around an American Kestrel nest box site in Massachusetts, USA. Five different cover categories were used (Grass, Herbaceous, Woody, Human-modified, and Bare Ground) but each observer drew their own training samples. The results of a confusion matrix evaluating how many pixels of each type were correctly identified by the Supervised Classification scheme are included here. Each observer (five naive observers and the authors) was assigned a number, 1 through 6, and summary statistics of kappa coefficients are included. -71.53 -71.53 42.23 42.23 -71.53 42.23 not specified US-MA Massachusetts 42.23 -71.53 United States Multiple investigations MULT 11.7 CC-BY-3.0 Creative Commons Attribution 3.0 Unported https://creativecommons.org/licenses/by/3.0/ citable author63878 author63879 event2848477 geocode1600 geocode1601 license101 method10667 sourceinst20586 term31243 topotype1