Angler Responses to Potential Harvest Regulations in a Norwegian Sport Fishery: A Conjoint-Based Choice Modeling Approach

Abstract

Experimental modeling of angler choice behavior is slowly emerging as an alternative to both the modeling of actual fishing behavior (revealed preferences), and the modeling of behavioral antecedents, such as attitudes, beliefs, and opinions. These conjoint-based choice modeling approaches are particularly useful for evaluating various fisheries management alternatives and programs. In this study we applied a discrete choice experiment (DCE), which relies on the principle of fractional factorial designs, to produce hypothetical profiles of fishing opportunities. The study was conducted in Engerdal, eastern Norway. Engerdal is a popular fishing area for locals and tourists from Norway and Sweden. Proper harvest regulation of a popular sport fishery for brown trout Salmo trutta and European grayling Thymallus thymallus is the major challenge to fisheries management in the area. Data were collected in a mail survey in the fall of 1996. Three regulation variables (minimum size limit, bag limit, and bait type) and two expectation variables (expected average fish size and expected catch numbers) were considered in this DCE. The study uncovered major differences in preferences for fishing opportunities between fly-only anglers and other angler groups. The fundamental difference between the main angling groups suggests that spatial segregation of these segments should be an essential management strategy, if both groups are to be provided with satisfactory experiences. The study shows great promise for using conjoint-based choice modeling in understanding angler responses to different harvest regulation alternatives. Future research is needed to refine its application in the human dimensions research.

Notes

¹ A normal distribution of the error term would result in a probit model, which is computationally much more demanding. Therefore, the Gumbel distribution is assumed for analytic convenience. Further details about the derivation of the MNL model can be found in standard texts such as Ben Akiva and Lerman (1985).