. 2021 Jul 7;68(3):353–365. doi: 10.1007/s00267-021-01497-4

Table 4.

Multilevel model estimation of the effects of various factors on the abandoned farmland ratio

Variable	Model 1		Model 2		Model 3		Model 4
Variable	Estimate	SE	Estimate	SE	Estimate	SE	Estimate	SE
Intercept	0.142	0.004^***	0.133	0.020^***	0.209	0.022^***	0.199	0.029^***
Characteristics of communities
Payment			–0.001	3.946E−04^**			–9.82E−04	3.887E−04^*
Community-based activities			–0.073	0.018^***			−0.070	0.018^***
Cooperation between communities			–0.019	0.009^*		−0.017	0.009^*
Cooperation with outside			–0.007	0.013			−0.004	0.013
Slope			0.004	0.001^***			0.004	0.001^***
Distance to the town center			0.728	0.110^***			0.711	0.109^***
Characteristics of farmers
Total farmland					2.157E−04	2.685E−05^***	2.105E−04	2.666E−05^***
Off-farm income dependency					0.009	0.005	0.010	0.005^*
Gender					0.010	0.011	0.009	0.011
Age					–4.093E−04	2.281E−04	–4.232E−04	2.275E−04
Successor					−0.012	0.005^**	–0.012	0.005^**
Labor					0.001	0.001	0.001	0.001
Machine					−0.003	0.002	−0.002	0.002
Crop type					−0.005	0.001^***	−0.005	0.001^***
Sales					−0.016	0.002^***	−0.015	0.002^***
Farmer-level variance	0.030		0.030		0.029		0.029
Community-level variance	0.011		0.008		0.010		0.008
Log-likelihood	1837.5		1881.2		1883.2		1920.4
AICc	–3669.1		–3744.4		–3742.4		–3804.7
delta-AIC	135.7		60.3		62.3		0.0
AIC weight	0.00		0.00		0.00		1.00
df	3		9		12		18
Number of observations	6893		6893		6893		6893
Number of groups	836		836		836		836

A negative number indicates negative effects

AIC akaike information criterion

***P < 0.001; **P < 0.01; *P < 0.05