Table 2.

Household well-being and livelihood strategies for a panel of 769 households bordering protected areas (PAs), within PAs, and in matched control areas outside PAs in northern Cambodia, 2008–2011

							Test of differencea
	Border PA		Within PA		Control		(within-PA vs. controls)
	2008	2011	2008	2011	2008	2011	2008	Change	2011
Households	141	141	443	443	185	185
Well-being variables
Poverty	10.5	12.5	9.6	11.8	8.0	11.4	*b	ns	ns
Rice harvest (kg)	2181	3015	1851	2506	1293	2329	ns	ns	ns
Food security (kg)	219	1942	−230	1337	−633	1109	ns	ns	ns
Livelihood strategiesc
Resin tapper (%)	32	30	55	59	28	37	***	ns	***
Rice farmer (%)	94	96	91	96	94	95	ns	ns	ns
>1 ha of paddy fields (%)	90	90	73	85	63	79	*	ns	ns
Mini tractor (%)	36	54	30	60	26	37	ns	**	***
Rice shifting cultivation (%)	38	27	37	26	45	39	*	*	**
Cash crops	n/a	5	n/a	2	n/a	10			**
Employed (%)	11	10	6	9	3	4	ns	ns	*
Service or shop (%)	23	24	14	26	14	29	ns	ns	ns

^aTests of difference are mixed effects regression models for continuous variables (poverty, rice harvest, food security, cattle) and generalized mixed effects models with a binomial link function for categorical variables.

^bTests of difference significance values: ns, not significant;

^*P < 0.05;

^**P < 0.01;

^***P < 0.001.

^cHouseholds could have more than one livelihood strategy.