Affects of road sign wording on visitor survey - non-response bias
On-site visitor interviewer data collection is a key component of the USDA Forest Service National Visitor Use Monitoring (NVUM) program. In many areas, especially higher speed roads and roads with non-recreation traffic, many vehicles may not stop for an interview. Wording on the sign may condition non-recreation visitors to self-select as to whether or not they decide to stop for an interview. Since the primary purpose of the interview is to calibrate a mechanical traffic counter, such behavior can lead to bias in the resulting visitation estimate. Non-response bias of national forest traffic was examined by using four different wordings for road signs during NVUM interview days. The experiment was performed using a randomized block design with each treatment (sign) being applied to five different road locations (blocks). Statistical analysis was performed to determine if any particular sign wording significantly affected (1) the rate of visitor response and (2) the mix of visitors who stopped for interviews. Data analysis show that the total number of all interviews obtained, the proportion of interviews obtained to overall traffic, and the proportion of non-recreation interviews obtained were different using different sign wording. The total number of recreation interviews obtained and the proportion of recreation interviews obtained were not different statistically.