|
Abstract:
|
Recreational fishing characteristics and catch along the US coasts have traditionally been monitored using combinations of intercept and telephone/mail recall surveys conducted by state and federal agencies. Because of increasing costs and declining response rates, efforts have been initiated to investigate ways to collect data via alternative channels, including logbooks. Even when logbook reporting is mandatory, however, there are concerns about underreporting of trips and/or misreporting of catch, so that it is of interest to combine the logbook data with validation surveys. In this presentation, we discuss a number of approaches to create combined estimators for the charter fishery in South Carolina. These approaches rely on matching of individual logbook and survey trips, followed by different ways to account for discrepancies between the logbook and survey data in estimation and inference. We describe and compare the methods developed for this application, and discuss their usefulness in other contexts that require combination of data from survey and non-survey sources.
|
