Statistical modelling/Milk offtake

Using CS1Data1 the model to be fitted (LOCATION+AGEC) can be done using Stats → Regression Analysis → Generalized Linear Models.... . The output describes the analysis of variance and shows evidence of a highly statistically significant association accounting for 32.4% of the variation.

A table of parameter estimates follows, the constant giving the intercept on the y-axis for Bilisa (the baseline or reference location (code 1)), LOCATION 2 representing the average difference between Bilisa and Assa, and AGEC representing the regression coefficient or slope for the two parallel regression lines. The standard error (s.e.) gives the precision with which the parameter estimates are determined.

 

 

 

Response variate: TOTALM
Fitted terms: Constant + LOCATION + AGEC

*** Summary of analysis ***

d.f.

s.s.

m.s.

v.r.

Regression

2

27.12

13.5612

40.08

Residual

161

54.47

0.3383

  

Total

163

81.60

0.5006

  

Percentage variance accounted for 32.4
Standard error of observations is estimated to be 0.582

*** Estimates of parameters ***

estimate

s.e.

t(161)

Constant

2.136

0.112

19.14

LOCATION 2

-0.8218

0.0971

-8.46

AGEC

-0.0511

0.0169

-3.02

Parameters for factors are differences compared with the reference level:
Factor Reference level
LOCATION 1