METHODS 



An 18-mile round trip, much 

 of it on snowshoes, was 

 required to service this 

 self-registration station at 

 Elk Pass in late October. 



The variables in each stratum were measured using a questionnaire for interviews 

 with groups entering or leaving (fig. 5) . Supplementary (covariate) information was 

 obtained by establishing registration boxes on each trail, where entering groups could 

 register by filling out a wilderness registration card (Appendix I) . This was a con- 

 densed version of the questionnaire used for trail interviews. Mechanical counters 

 placed on some of the most heavily used trails and access roads provided additional 

 covariate data. 



Interviewing Procedures 



Four interviewers worked as two-man teams checking sample locations from 8 a.m. to 

 8 p.m. from June 15 until October 20. From October 20 through November 11 the checking 

 period was shortened to 12 noon to 6 p.m. because the reduction in daylight hours 

 coupled with changes in use lessened the chances for missing parties. This made it 

 possible to check twice as many locations each day during the last 3 weeks--four loca- 

 tions on weekends, two on weekdays. 



Interviewers stationed themselves up the trail, out of sight of the registration 

 station. They drove personal cars rather than official Forest Service vehicles to 

 avoid influencing visitor registration. Interviewers were dressed in official uniform. 

 They conducted a brief interview with every party that entered or left. 



11 



