Methods of estimating mean school size from IWC sightings survey data

Research output: Contribution to journalArticlepeer-review


Available methods are reviewed in the context of the IWC sighting surveys. Using regression methods, mean school sizes are estimated from the 1991/1992 minke whale Balaenoptera acutorostra survey data and from eight earlier datasets for which school size estimation had been problematic. Estimation by regression of ln (school size) on estimated g(y) performs better than the currently standard IWC estimation method on the 1991/92 dataset, and it performs better than the regression method initially used on the eight problem datasets. In the case of these earlier datasets, direct comparison of the currently accepted estimates obtained from the currently standard IWC method and this new regression method are difficult because of differences in the datasets which have come about as a result of changes in what are considered the best data for school size estimation. However, comparison suggests that the regression method is more robust than the currently standard IWC method. The new method also has a number of theoretical advantages over the currently standard IWC method, including the fact that valid variance estimates are easily obtainable and that decisions on stratification can be made objectively and independently of the stratification decided on for estimation of effective strip half-width. -from Author

Original languageEnglish
Pages (from-to)429-437
Number of pages9
JournalJournal of Cetacean Research and Management
Publication statusPublished - 1 Jan 1994


Dive into the research topics of 'Methods of estimating mean school size from IWC sightings survey data'. Together they form a unique fingerprint.

Cite this