Factors influencing hunting license sales in urban and rural areas of Illinois

Sales of hunting licenses have fallen in the past decades. To seek the means to maintain or increase current hunter numbers, state agencies need to understand their existing market base before developing strategies to boost sales. This can be achieved by exploring the characteristics of the hunting population and factors influencing hunting license sales. Recent studies have examined these factors influencing hunting license sales at an aggregated scale. These studies helped to understand not only the influence of these factors, but also the environment in which individual indicators are embedded. This study used a similar approach, but with more factors and different models. The study area was the state of Illinois. The study consisted of three parts. The first created a regression model for the entire state. Different socioeconomic and biophysical factors were included in the model. Model was transformed to reduce heteroscedasticity and non-normality. Stepwise regression was applied to the transformed model to select variables. The second part addressed the differences between rural and urban areas in Illinois, using the same methods as in the first part. The third part considered spatial dependency of the model residuals, using the Moran test and Lagrange Multiplier test for diagnosis. Global models (spatial lag model, spatial error model, and hierarchical linear model [HLM]) and a local model (geographically weighted regression [GWR]) were applied to deal with spatial autocorrelation of the residuals. The first part found that accessibility to hunting resources, economic status, age structure, education, race and ethnicity, and competition with general recreation influenced Illinois hunting license sales. The second part found that the significant factors for the entire state, rural areas, and urban areas were different. The influence of the eight variables was robust over different models. The third part found spatial dependency in the residuals of the model used in the first part. Spatial regression (spatial lag and spatial error models), HLM, and GWR were applied to reduce spatial dependency. GWR had the best fit of all the models considered. Spatial lag regression had the best fit of all the global models. The spatial lag regression model excluded spatial dependency in the residuals, but there was some spatial dependency in the residuals of the other models.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Natural Res & Env Sciences

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Zhang, Xiaohan

Contributors dc:contributor

• Miller, Craig
• Brazee, Richard
• McLafferty, Sara
• Vaske, Jerry

• Hunting license
• regression
• rural and urban
• spatial dependency
• spatial regression

Statement dc:rights

• Copyright 2019 Xiaohan Zhang

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/106241

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/106241