but they depend on certain assumptions that cannot be adequately tested. Such a com- 

 parison would only have validity if "loopholes" were closed, such as the visitors who 

 entered or left on sample days when the interviewers were not on the trail. Otherwise 

 the sampling fraction, based on the proportion of days sampled, exaggerates the propor- 

 tion of use sampled. One way of dealing with this problem would be to define the 

 sampling frame in terms of hours, and choose blocks of hours randomly. The extremely 

 late and early hours, say from 11:00 p.m. to 5:00 a.m., could probably be safely ignored. 



Table 6 . --Stratified vandom- sampling estimates of recreation use on the Mission 

 Mountains Primitive Area, Montana, based on interview of entering and 

 leaving visitors, swrmer-falt 1968 



Stratum 



Visits 



■ Total 



use 



Group 



s 



Estimate 



: Standard 

 : error 



\ Estimate ' 



Standard : 

 error ■'■ : 



Estimate 



-v» Tr% /-\ -v^ L 





Number 



Percent 



Visitor-hours 



Percent 



Number 







BASED 



ON INTERVIEW 



OF VISITORS 



UPON ENTRANCE 







Summer 



2,878 



11 .3 



20,918 



12.9 



748 



10.3 



Fall 



506 



22.9 



12,996 



59.8 



209 



18.9 



Summer + Fall 



3,384 



10.6 



33,914 



24.3 



957 



9.0 



1 



763 



6.8 



6,795 



13.2 



190 



5.8 



2 



692 



11.3 



5,619 



17.6 



209 



8.6 



3 



686 



16.9 



3,774 



23.7 



160 



17.5 



4 



737 



41.5 



4,730 



46.0 



189 



36.0 



5 



228 



37.7 



958 



26.3 



97 



33.0 



6 



200 



31.0 



4,039 



44.6 



83 



25.3 



7 



78 



60.3 



8,154 



92.7 



29 



41.4 





BASED 



ON INTERVIEW OF VISITORS 



UPON DEPARTURE 





Summer 



2,898 



12. 1 



22,275 



8.0 



776 



9.6 



Fall 



668 



20.4 



3,681 



25.0 



273 



19.3 



Summer + Fall 



3,566 



10.6 



25,956 



7.8 



1,049 



8.7 



1 



838 



6.8 



8,038 



9.9 



209 



5.6 



2 



600 



13.8 



4,125 



11.5 



191 



10.5 



3 



837 



16.7 



7,281 



14.8 



206 



16.9 



4 



623 



49.1 



2,832 



38.4 



170 



36.0 



5 



215 



40.0 



652 



37.3 



93 



38.2 



6 



297 



22.2 



2,279 



35.4 



131 



22.1 



7 



156 



52.6 



750 



49.9 



49 



52.5 



Standard error, expressed as percentage of the estimate. 



27 



