Page 39 - Contributed Paper Session (CPS) - Volume 2

CPS1414 Bashiru. I.I S. at el.
be important to allow decision makers to delineate areas where the situation
is rampant and also for effective allocation of limited resources (Li et al., 2013).
This study addresses this gap by examining the spatial pattern of child
labour practices and its determinants among various districts in Ghana using
geographically weighted linear model (GWLM). The use of GWLM approach
provides a deterministic tool that is capable of assessing spatial variation in
the contributing factors of child labour practices.

2. Materials and Methods

Figure 1: Maps of Ghana showing the centroid of the various districts

2.2 Geographically Weighted Linear Model Specification

i 
y =  ( u , v +  ( u , v ) x +  i (1)
)
i
0
i
i
ik
k
i
k

where (u i , v i )denoting the two dimensional geographical coordinates
th
of the i district (centroid of the district),  0 (u i , v i ) represent the
th
intercept at location (u i , v i ), x denote the k independent variable,
ik
 k (u i , v i ) represent the model coefficient associated with explanatory
variable at location (u i , v i ) and  represent error term N , 0 (  i 2 ) .This
i
means the parameter  = ( 0 , 1 ,..., k ) estimated in (4) are allowed to
be different between districts. Thus, the spatial heterogeneity is
addressed in the GWLM modelling framework. The parameter can be
expressed in the matrix notation as

28 | I S I W S C 2 0 1 9