| Burrows | Drone survey | Total |
---|
Potentially active | Inactive |
---|
(a) Grassland |
Ground survey | Potentially active | 46 | 7 | 53 |
Inactive | 20 | 12 | 32 |
 | Total | 66 | 19 | 85 |
(b) Tussock grassland |
Ground survey | Potentially active | 12 | 2 | 14 |
Inactive | 9 | 4 | 13 |
 | Total | 21 | 6 | 27 |
- Most of the individual burrows observed in the study were detected during both ground and drone surveys of their particular transect. Tables a and b show the numbers of these burrows that were classified as potentially active and/or inactive in each of the two grassy communities. To provide a basis for evaluating agreement of this aspect of the two methods, the marginal proportions in each table were tested for homogeneity using McNemar’s test (see text)