1
$\begingroup$

I´d like to analyse the effect of a treatment (treatment : Factor w/ 2 levels "ambient","elevated") in tree diameter increment. Tree diameter is influenced by tree size. To do so, I performed the following lm:

Call:
lm(formula = BAI2013 ~ diameterJul12 * treatment, data = bandNA)

Residuals:
     Min       1Q   Median       3Q      Max 
-16.6493  -3.1740  -0.3767   3.3631  22.7267 

Coefficients:
                                 Estimate Std. Error t value Pr(>|t|)    
(Intercept)                     -20.49357    2.12883  -9.627  < 2e-16 ***
diameterJul12                     1.24194    0.08876  13.992  < 2e-16 ***
treatmentelevated                10.72336    3.45783   3.101 0.002295 ** 
diameterJul12:treatmentelevated  -0.54953    0.14795  -3.714 0.000285 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 6.461 on 153 degrees of freedom
Multiple R-squared:  0.6035,    Adjusted R-squared:  0.5958 
F-statistic: 77.63 on 3 and 153 DF,  p-value: < 2.2e-16

How can I interpret the results? I need to figure out whether the slope of BAI~diameter is steeper for elevated trees than for ambient. Thanks

$\endgroup$
5
$\begingroup$

The maximum likelihood fitted model is

BAI2013 = 1.24194 * diameterJul12 + 10.72336 * treatmentelevated - 0.54953 * diameterJul12 * treatmentelevated - 20.49357

Here treatmentelevated is a binary variable which is 1 for "elevated". All terms have significant p-values (these come from Wald statistics) suggesting they should be kept in the model. The interaction term is negative suggesting the slope of BAI~diameter is less steep (by 0.54953) in the elevated treatment group.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.