Effects of Sample Selection on Estimates of Economic Impacts of Outdoor Recreation

Donald B.K English

doi:10.52324/001c.8942

Abstract

Estimates of the economic impacts of recreation often come from spending data provided by a self-selected subset of a random sample of site visitors. The subset is frequently less than half the onsite sample. Biased vectors of per trip spending and impact estimates can result if self-selection is related to spending patterns, and proper corrective procedures are not employed. This paper shows a method for accounting for both sample selection and the censored nature of reported expenditures, via a Tobit model with sample selection. Results from a sample of visitors to Cumberland Island National Seashore indicate a naive (uncorrected) approach overestimates per trip visitor spending by 15 percent and economic impacts to industrial output by 10 percent.

Highlights

Estimates of the economic impacts associated with recreation visitation are useful policy tools for many federal and state agencies in making land management decisions (Jackson, et al, 1992; Johnson and Moore, 1993; USDA Forest Service, 1995)
Approaches used in assessing economic impacts of recreation visitation have been reasonably well defined and documented (Alward and Lofting, 1985; Alward, et al, 1985; Stevens and Rose, 1985; Leitch and Leistritz, 1985; Johnson and Moore, 1993; English and Bergstrom, 1994)
This paper has examined the potential for sample selection bias in visitor expenditure means and in subsequent economic impacts of recreation, when visitor expenditure data are collected in the conventional manner

Summary

INTRODUCTION

Estimates of the economic impacts associated with recreation visitation are useful policy tools for many federal and state agencies in making land management decisions (Jackson, et al, 1992; Johnson and Moore, 1993; USDA Forest Service, 1995). The process begins with an onsite random sample of N visitors or visitor groups All of those contacted provide information on some vector V of data items on themselves and/or their group, such as demographics, use patterns, or trip descriptors. The amount of money spent near the recreation site is assumed to be explained by another vector of variables, X Information on these variables can be obtained through either the onsite contact or mailed survey. This vector may be no more than a series of dummy variables describing primary recreation activities, or it may include variables thought to be related to expenditure levels, such as group size, travel distance, or length of stay. The final sample included 812 onsite observations, of which 370 had provided expenditure data

EMPIRICAL MODELING

ECONOMIC IMPACT ESTIMATES

Findings

DISCUSSION