Evaluation of sampling techniques for the Lake Michigan angler survey

Abstract.–The Lake Michigan angler survey currently uses the roving-access method for estimating angling effort and harvest at specific sites in State of Michigan waters. Sites are sampled with equal probability following a stratified random design. Four sampling techniques were evaluated to decrease sampling cost while maintaining or improving accuracy and precision of these estimates. Techniques were: determine site-to-site relationships for predicting angling effort; optimal allocation of clerk effort; proportional allocation of clerk effort; and bus route design. All four techniques were shown to be unsuitable replacements for current sampling techniques. Site-to-site relationships of angling effort were found to be inconsistent, whil