3.4.2. Structural Equation Modeling

A structural equation model was applied to the determinants of smallholder livestock farming households' resilience to food insecurity (Table 2). The model applies a factor analysis-type model to measure the latent variables via observed variables, simultaneously using a regression-type model for the relationship among the latent variables [36,42]. The structural equation model for a household *i* is illustrated as (Equation (6)):

$$\text{ADRI}\_{\text{i}} = \text{f}(\text{ASS}\_{\text{i}} \text{ ADC}\_{\text{i}} \text{ SSF}\_{\text{i}} \text{ CH}) \tag{6}$$


**Table 2.** Description of variables used in structural equation modeling.

Source: Authors' observation (2020).

Equation (6) is disaggregated in detail as (Equation (7)):

ADRIi = f(ASSi(herd/flock size (HFS); agricultural assets (AA); non-agricultural assets (NAA)),

ADCi (perception; source of income (Incsource); migration; credit),

SSFi (cash; training; food support; water rights; garden equipment; sanitary latrines, farm input), herd/flock size (HFS);

CH (drought occurrence and intensity))

(7)
