Skip to main content

Table 3 Australia in 2016. Multiple linear regression model outputs of factors influencing daily milk yields in does on a large dairy goat enterprise in south western Victoria

From: The prevalence of Coxiella burnetii shedding in dairy goats at the time of parturition in an endemically infected enterprise and associated milk yield losses

Variable

Coefficient (SE)

t

p value

95% CI

Intercept

2.38 (0.11)

21.26

< 0.001

2.15 to 2.59

qPCR status:

 Negative

Reference

 

-

-

 Positive low

-0.08 (0.12)

-0.68

0.495

-0.33 to 0.16

 Positive high

-0.53 (0.23)

-2.33

0.02

-0.98 to -0.08 a

Parity:

1

Reference

 

-

-

 2

0.34 (0.10)

7.29

< 0.001

0.54 to 0.94

 3

1.22 (0.11)

10.7

< 0.001

0.99 to 1.44

 4+

0.71 (0.13)

5.5

< 0.001

0.46 to 0.96

Kidding season:

 March to April

Reference

 

-

-

 June to July

0.06 (0.11)

0.49

0.62

-0.17 to 0.28

 September to October

0.53 (0.12)

4.52

< 0.001

0.3 to 0.77

 November to December

0.54 (0.12)

4.57

< 0.001

0.3 to 0.77

Farm:

 A

Reference

 

-

-

 B

0.24 (0.10)

2.34

0.02

0.04 to 0.44

 C

-0.71 (0.11)

-6.72

< 0.001

-0.92 to -0.5

Random effect

SD

   

 Goat id

0.82

   

Within goat temporal correlation with first order autoregressive structure (AR1): 0.23

  1. AIC: 343081.1
  2. aInterpretation: After adjusting for the effect of parity, kidding season, days in milk and farm, daily milk yields from does that were qPCR-positive high were 530 (95% CI 80 to 980) mL less than does that were qPCR negative. The random effect at the goat level accounts for repeated measures and individual animal variability.
  3. SE: standard error.
  4. CI: confidence interval