Table 3.
Effect of treatment on visitor variables measured. The means shown are the mean frequency of visitor behaviours (per seven 30 min observations blocks per 2-day period) considered as events and ambient noise level per 2-day period. All measurements were square root transformed, except for ambient noise level and the number of visitors (*). Back-transformed means are presented in parentheses.
| Visitor Variables | Covariate | Open Exhibit | p-Value | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Increased Viewing Distance | Normal Viewing Distance | Proximity a | Regulation Effects a | Proximity × Regulation Interaction | |||||
| Regulated | Unregulated | Regulated | Unregulated | s.e.d. c | |||||
| Ambient noise (dB) * | Maximum temperature | 62 | 61 | 61 | 62 | 0.62 | 0.49 | 1.00 | 0.11 |
| Number of visitors * | Visitor gate numbers | 360 | 300 | 500 | 460 | 46 | 0.0047 | 0.080 | 0.77 |
| Behaviour | |||||||||
| Tactile contact with enclosure features | 0.91 (0.83) | 1.7 (2.8) | 5.1 (26) | 6.2 (39) | 1.18 | 0.00076 | 0.30 | 0.84 | |
| Loud vocalisations | Proportion of males | 8.5 (73) | 9.8 (97) | 7.0 (48) | 11 (123) | 1.87 | 0.98 | 0.075 | 0.33 |
| Looming with penguin > 0.5 m (from looming position) | 0.57 (0.32) | 1.6 (2.6) | 22 (492) | 21 (461) | 1.06 | 3.2 × 10−9 | 0.82 | 0.28 | |
| Looming with penguin < 0.5 m (from looming position) | 0 | 0 | 0.86 (0.74) | 1.3 (1.6) | 0.85 d | 0.061 b | 0.64 b | 0.64 b | |
| Tactile contact with water | 0 | 0 | 3.6 (13) | 5.1 (26) | 0.85 d | 0.00054 b | 0.09 b | 0.09 b | |
| Tactile contact with penguin | 0 | 0 | 0 | 0.64 (0.41) | 0.36 d | 0.18 b | 0.18 b | 0.18 b | |
| Sudden movement | 2.0 (4) | 1.1 (1) | 5.0 (25) | 5.0 (25) | 1.63 | 0.019 | 0.72 | 0.73 | |
ap values were calculated using F tests based on 1, 7 or 1, 8 degrees of freedom depending on whether a covariate was included in the analysis. p values less than 0.05 are in bold. b p values calculated using permutation test. c s.e.d. denotes standard error of difference. d sed calculated using residual standard deviation obtained from analysis only using observations with normal viewing distance.