ANALYSIS 
Limitations of the Survey Method 
Since survey routes are not selected at random 
they may not be typical of all woodcock habitat. The 
extent to which counts reflect population changes is not 
known. At present, we can only assume that drastic changes 
in the size of the breeding population will be detected. 
It should be remembered that the Singing Ground 
Survey attempts to measure changes in the relative size of 
the breeding population, not annual production, which must 
be measured by other methods. 
Summarization of Field Data 
This analysis 1s based on data from States and 
Provinces in which one or more comparable routes were run 
in 1960 and 1961. Several routes covered in 1961 were not 
included in the analysis, either because they were run 
during the unreliable period from 2 days before to 1 da 
after the full moon, or because stops were less than O. 
mile apart. If a route was covered more than once, only 
the last trip was included in the computations. These 
procedures were followed in order to make the survey results 
comparable with surveys in previous years. The reliability 
of the annual survey depends upon the comparability of the 
counts from each route, State by State, and region by region, 
Therefore, it is important that cooperators follow the survey 
instructions as closely as possible. A route should be run 
only once each year unless factors such as wind, temperature, 
traffic or other noise result in a non-representative count. 
If a route is covered more than once, the cooperator should 
clearly indicate the reason for the second trip. 
Adjustment for Length of Route 
Survey data of previous years have been analyzed 
in two ways: the number of woodcock heard per stop, and 
the number of woodcock heard per route. Robbins has shown 
