BMC Public Health

Table 3 Sensitivity analysis results

From: Bayesian belief network modelling of household food security in rural South Africa

	a: Expert elicitation			b: Data-learned			c: Data + expert prior
Variable	I	I/H	_S2	I	I/H	_S2	I	I/H	S²
CGS	0.0099	0.0115	0.0030	0.0000	0.0000	0.0000	0.0002	0.0004	0.0000
EdL	0.0078	0.0091	0.0023	0.0000	0.0000	0.0000	0.0001	0.0001	0.0000
EmL	0.0177	0.0205	0.0052	0.0001	0.0002	0.0000	0.0082	0.0165	0.0013
HHG	0.0000	0.0000	0.0000	0.0002	0.0004	0.0000	0.0000	0.0001	0.0000
LLV	0.0264	0.0307	0.0079	0.0000	0.0000	0.0000	0.0028	0.0056	0.0005
ND	0.0021	0.0025	0.0006	0.0001	0.0001	0.0000	0.0008	0.0016	0.0001
NWAA	0.0007	0.0008	0.0002	0.0007	0.0014	0.0001	0.0005	0.0011	0.0001
RCA	0.0083	0.0097	0.0025	0.0000	0.0001	0.0000	0.0066	0.0133	0.0010
RS	0.0000	0.0000	0.0000	0.0007	0.0016	0.0001	0.0003	0.0005	0.0000
SCL	0.0108	0.0126	0.0034	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
SES	0.0067	0.0078	0.0020	0.0112	0.0244	0.0014	0.0114	0.0228	0.0017
UCL	0.0039	0.0046	0.0011	0.0011	0.0025	0.0001	0.0008	0.0017	0.0001
UWF	0.0334	0.0388	0.0101	0.0043	0.0093	0.0007	0.0066	0.0133	0.0012
WA	0.0049	0.0056	0.0014	0.0000	0.0000	0.0000	0.0001	0.0003	0.0000

Mutual information reduction I (also divided by food security information H) and expected change of belief S² for food security given each of the other variables for the three different networks
CGS child grant status, EdL education level, EmL employment level, FS food security, HHG household head gender, LLV level of local vegetation, ND number of dependents, NWAA number of working age adults, RCA receipt of communal aid, RS refugee status, SCL selling of crops and livestock, SES socio-economic status, UCL use of crops and livestock, UWF use of wild foods, WA water access

Back to article page

ISSN: 1471-2458

Contact us

Submission enquiries: bmcpublichealth@biomedcentral.com
General enquiries: ORSupport@springernature.com