Rice Genomics and Genetics 2015, Vol.6, No.7, 1-10
1
Table 5 Description of the variables specified in the model
Variable
Meaning
Description
Expected signs
Dependent variable (first stage)
Grown improved
variety
Whether a farmer grows at least
one improved rice variety or not.
Dummy (grows = 1, 0 otherwise)
Dependent variable (second stage)
Area grown to
improved variety
Proportion of farm area grown
to improved rice variety
Continuous (hectares)
Independent variables
Gender
Gender of the farmer
Dummy (male =1, Female =0)
+
Age
Age of the household head
Continuous (years)
_
Education
Education level of the household head
Continuous (years of formal education
+
Household size
number of people in household
Continuous (Number)
+
Membership in
association
Whether a farmer belong to any association Dummy (member = 1 , 0 otherwise)
+
Access to media
Whether a farmer had access to
media or not.
Dummy (access to media = 1, 0 otherwise) +
Ownership of mobile
phone
Whether a farmer owned a mobile
phone or not
Dummy (owns a mobile phone= 1, 0
otherwise)
+
Farm size
Size of farmland
Continuous hectares
+
Main occupation
Main occupation of the rice farmer
Dummy (1= farming, 0 otherwise)
±
Access to credit
Whether a farmer has access to credit or not Dummy (access to credit = 1, 0 otherwise) +
Agricultural training
Whether a farmer has attended
agricultural training or not
Dummy (attends agricultural training = 1,
0 otherwise)
+
Off-farm employment
Whether a farmer has any
off farm employment or not
Dummy (engage in non-agricultural
activities = 1, 0 otherwise)
_
Agricultural income
Income from all agricultural activities
Continuous (naira)
±
β
i
=A vector of parameter estimates
X
i
=A vector of explanatory variables which include;
X
1
=Age of the household head (years)
X
2
=Education level of the household head (years)
X
3
=Household size (number of people in
household)
X
4
=Membership in association (Yes=1, 0 otherwise)
X
5
=Access to media (Yes=1, 0 otherwise)
X
6
=Ownership of mobile Phone (Yes=1, 0 otherwise)
X
7
=Main Occupation (farming=1, 0 otherwise)
3.3.2 Heckman two-stage sample selection model
The Heckman two-stage sample selection model was
used to identify the determinants of adoption and its
intensity among the respondents (table 5). The first
stage employed the probit regression model with the
dependent variable specified as dichotomous assuming
value 1 if the farmer adopts one or more of the
improved rice varieties and 0 otherwise (Usman et
al., 2011, Awotide et al., 2011)
(2)
Z
i
*=whether a farmer adopts improved rice variety
or not (1=yes and 0 otherwise)
X
i
=vector of factors influencing the adoption of
improved rice varieties.
X
1
=Gender of the farmer
X
2
=Age of the household head
X
3
=Education level of the household head
X
4
=Household size (number of people in household)
X
5
=Membership in association
X
6
=Access to media
X
7
=Ownership of mobile Phone
X
8
=Farm Size
X
9
=Main Occupation
X
10
=Access to credit
