Lovell and Carter: Sampling weights in regression models of recreational fishing-site choice 
247 
The number of angler fishing trips, by state, that targeted groupers (Epi- 
nephelus spp., Hyporthodus spp., or Mycteroperca spp.) from private boats 
in the Gulf of Mexico in 2009. Data are from the National Marine Fisher- 
ies Service’s Access Point Angler Intercept Survey. Trips are distributed 
across coastal counties in Florida, Alabama, Mississippi, and Louisiana 
according to locations of intercept sites on shore where anglers were in- 
terviewed for the survey. 
by the sample proportions (Hj) to construct 3 versions 
of the weights for the WESML estimator (Table 1). On 
the basis of the lower limit of the population propor- 
tion, 12 of the 28 zones had no predicted trips in the 
scenario; these zones were left out of the model in the 
lower-limit scenario. In other words, whereas the other 
models were estimated with 28 zones, the lower-limit 
WESML model was estimated with 16 zones. 
We used APAIS catch data to calculate the un- 
weighted and weighted average harvest rates for grou- 
pers and red snapper for each of the 28 zones over the 
5 years (2004-08) that preceded our sample period of 
2009 (Table 1). Average harvest rates were calculated 
with the Surveymeans procedure in SAS. These target- 
ed harvest rates represent the number of fish caught 
and kept per angler on a targeted trip and are used as 
a proxy for fishing quality in the vector, q, of attributes 
in the demand model. We also considered including the 
catch rates for other snappers (as a group) into the 
model as did Haab et al. (2012). However, there were 
very few target trips for offshore varieties of snappers 
in the APAIS data set (gray snapper [ Lutjanus griseus ] 
was excluded because it is primarily an inshore spe- 
cies). Zones with historically high targeted harvest 
rates for groupers and red snap- 
per should be preferred by anglers 
targeting these species. Note that 
the unweighted conditional logit 
demand model was based on the 
unweighted harvest rates, whereas 
the WESML demand models were 
based on the weighted harvest 
rates. Our goal was to compare 
estimated parameters and angler 
WTP between a typical demand 
model that does not use any APA- 
IS weight information and typical 
demand models that do use this 
information. 
Travel cost, c, to each zone was 
calculated as the round-trip dis- 
tance in miles, from the centroid of 
the angler’s zip code of residence 
to the centroid of the first zip code 
in the zone, multiplied by the cost 
per mile. The distances were cal- 
culated with PC*Miler, vers. 23 
(ALK Technologies, Princeton, NJ). 
We used $0.59 as the cost per mile 
on the basis of the standard busi- 
ness mileage rate for 2009 from 
the Internal Revenue Service 
(IRS 4 ), adjusted to 2012 dollars 
using the Consumer Price Index 
(CPI Inflation Calculator, Bureau 
of Labor Statistics, http://www.bls. 
gov/data/inflation_calculator.htm). 
We did not include the opportunity 
cost of time in the model because information on an- 
gler income or whether or not the angler took time off 
from work to fish was not collected in the 2009 APAIS. 
Centner (2007) notes that relatively few (e.g., <4% in 
2000) anglers reported having foregone income when 
asked to be part of the APAIS. 
Results 
In 2009, anglers who targeted primarily groupers took 
trips from central and southwestern Florida (Fig. 1). 
Anglers in Louisiana and Mississippi targeted red 
snapper more often than groupers (Fig. 2). In the Flor- 
ida panhandle area and in Alabama, anglers tended to 
target both species, but the proportion of trips during 
which groupers were targeted there was less than the 
proportion of trips in other areas of Florida. In the 
models, 990 observations were available for use; 725 
trips during which anglers targeted groupers and 265 
4 IRS (Internal Revenue Service). 2008. IRS announces 
2009 standard mileage rates. IRS News Release IR-2008- 
131, 24 November. [Available from http://www.irs.gov/uac/ 
IRS-Announces-2009-Standard-Mileage-Rates.] 
