Skip to main content
. 2020 Jun 19;8:e9401. doi: 10.7717/peerj.9401

Table 3. Fall and winter bird-window collisions per day predicted by building risk and protective features: Generalized estimating equation results for replication of Brown et al. (2019).

95% Wald C.I. p-value, original
B S.E. p Exp(B) Lower Upper Study
Model 1: All three predictors
Intercept −2.80 0.30 0.01 0.06 0.03 0.11 0.01
Pear trees 0.49 0.42 0.24 1.64 0.72 3.71 0.04
Mirrored windows 0.85 0.39 0.03 2.33 1.08 5.05 0.09
Bird-friendly glass −1.30 0.77 0.09 0.27 0.06 1.23 0.78
Model 1: All three predictors
Intercept −3.14 0.28 0.01 0.04 0.03 0.08 0.01
Pear trees 1.20 0.34 0.01 3.33 1.71 6.49 0.01
Model 3: Mirrored windows only
Intercept −2.82 0.20 0.01 0.06 0.04 0.09 0.01
Mirrored windows 1.37 0.33 0.01 3.92 2.04 7.54 0.01
Model 4: Bird-friendly glass
Intercept −2.28 0.16 0.01 0.10 0.07 0.14 0.01
Bird-friendly glass −1.82 0.73 0.01 0.16 0.04 0.67 0.11
Model 5: Pear + mirrored, best fit
Intercept −3.14 0.28 0.01 0.04 0.03 0.08 0.01
Pear trees .84 0.40 0.04 2.31 1.05 5.06 0.01
Mirrored windows .85 0.39 0.03 2.33 1.08 5.05 0.09

Notes.

B
B coefficient
S.E.
Standard error of B
Exp(B)
odds ratio
CI
confidence interval

Data were collected in fall 2019 and winter 2020 at the University of Utah, Salt Lake City, Utah, USA.