Skip to main content

Table 6 Determinants of organic cotton adoption: results of the probit model

From: Socioeconomic determinants of organic cotton adoption in Benin, West Africa

Variable

Label

Coefficient (a)

Marginal effect (a)(b)

AGE

Age of the producer (year)

0.025* (0.014)

0.010* (0.005)

SEX

Sex of the producer (1 = male, 0 = female)

−1.542*** (0.441)

−0.490*** (0.107)

EDUC

Formal education of the producer (year)

−0.118** (0.053)

−0.046** (0.021)

EXPCOT

Experience in cotton production in general (year)

−0.034* (0.021)

−0.013* (0.008)

CLASSP

Class of poverty as perceived by producers (1 = very poor, 2 = poor, 3 = medium, 4 = rich)

−0.102 (0.206)

−0.039 (0.080)

CENTRE

Regional dummy (1 = Farmers located in the Centre of Benin, 0 = otherwise)

1.346** (0.641)

0.451*** (0.173)

NORTHW

Regional dummy (1 = Farmers located in the northwest-Benin, 0 = otherwise)

−0.318 (0.473)

−0.124 (0.185)

LNDISTEX

Logarithm of the distance between the producer’s house and his farm (km)

−1.112*** (0.326)

−0.431*** (0.126)

HHSIZE

Household size (number of members)

0.052* (0.031)

0.020* (0.012)

LNSUPTOT

Logarithm of the amount of land available (ha)

−0.235 (0.237)

−0.091 (0.092)

OXEN

Ownership of oxen (1 = yes)

0.452 (0.572)

0.168 (0.202)

LNDISTM

Logarithm of the distance between the village and the nearest market place where producer sell their products (km)

0.075 (0.117)

0.029 (0.045)

ACRED

Obtained credit (1 = yes, 0 = no)

−0.340 (0.341)

−0.133 (0.134)

COOP

Membership of producers’ organization (1 = yes)

0.554 (0.371)

0.217 (0.144)

NVISIT

Contact with extension agents (number of visits between January and October 2012)

0.355*** (0.073)

0.138*** (0.026)

CONS

Constant

0.192 (0.738)

 
  1. Number of observation = 191; LR chi2(14) = 144.43***, Log likelihood = −60.110, Pseudo R2 = 0.5457.
  2. (a)Figures in parenthesis are standard errors; (b)The partial derivative is given by \( \frac{\partial E\left[y\Big|x\right]}{\partial x}=\phi \left({x}^{\prime}\beta \right)\beta \) continue variables and \( \mathrm{Prob}\left[y=1\Big|{\overline{x}}_{(d)},d=1\right]-\mathrm{Prob}\left[y=1\Big|{\overline{x}}_{(d)},d=0\right] \), for dummy variables (Greene 2008: 775), with \( {\overline{x}}_{(d)} \) the means of all the other variables in the model.
  3. ***Significant at 1%, **Significant at 5%, *Significant at 10%.