In estimating insect density and impact, entomologists are understandably interested in accuracy of estimation, but they almost always are dealing with precision because of bias due to an invalid estimator, probability sampling, or nonsampling errors. Definitions related to statistical estimation are reviewed and the concepts of accuracy and precision examined.
Interval estimation and optimum sample size determination related to accuracy and precision, using the concept of allowable error, are examined. Criteria for selecting the best estimator in tenns of accuracy and precision are presented, and the distortion of probability statements due to bias is discussed. Accuracy and precision are compared and contrasted using two examples: (I) estimating insect density and (2) estimating insect impact. Adjusted and more accurate estimators can be obtained if the bias of an estimator can be estimated from a preliminary sample.
Fowler, Gary W. and Witter, John A.
"Accuracy and Precision of Insect Density and Impact Estimates,"
The Great Lakes Entomologist: Vol. 15
, Article 5.
Available at: http://scholar.valpo.edu/tgle/vol15/iss2/5