Table 2.

Model results for the effects of study year (Y), plant species origin (O; native, non-native), earthworm density (E; low, high) and fencing (F; open, fenced) on vegetation cover (%) according to life form at 12 sites at West Point, NY from 2009 to 2012. We ran independent linear mixed models for each life form. All models included site, plot within site and quadrat within plot as random factors. Non-significant study factors were dropped from selected models and not included in the table.

Factor		Intercept	Year	Origin (non-native)	Earthworm (low)	Fencing (open)	O × Y	O × E	O × F	Y × F
(Bi)Annuals
	Est*	0.006		−0.002
	SE	0.002		0.001
	χ2**			15.16
	P			<0.001
Graminoids
	Est	3.85	−0.44	9.47	−2.12	−2.34	−1.36	−10.91	5.61	1.29
	SE	3.87	0.51	1.36	6.36	2.15	0.59	1.39	1.31	0.59
	χ2						5.37	82.5	121.47	4.85
	P						0.02	<0.001	<0.001	0.03
Perennials
	Est	0.022		−0.021	−0.01	−0.004		0.009	0.004
	SE	0.002		0.001	0.002	0.001		0.001	0.001
	χ2							109.51	24.82
	P							<0.001	<0.001
Woody
	Est	0.027	0.001	−0.015	0.011	−0.006		−0.013	0.009
	SE	0.005	0.0005	0.002	0.008	0.002		0.002	0.002
	χ2		5.22					34.29	18.61
	P		0.02					<0.001	<0.001

Est Estimate; SE estimate standard error.

^*Estimates and standard errors are reported from the model fitted with restricted maximum likelihood.

^**Chi-squared statistics and P-values are from likelihood ratio tests with each parameter removed from the maximum likelihood-based model, with all other parameters retained. It was not possible to test the significance of all terms because of higher order interactions. All chi-square tests have 1 degree of freedom.