Skip to main content

Table 7 Impact of SAPs on plot-level productivity (log value of harvest per hectare)

From: Climate variability, agricultural technologies adoption, and productivity in rural Nigeria: a plot-level analysis

Variable

OLS

IVREG2H

Coefficient

Std error

Coefficient

Std error

Improved seeds

0.032

0.055

0.104

0.146

Inorganic fertilizer

0.324***

0.034

0.305*

0.179

Organic fertilizer

0.019

0.040

 − 0.427***

0.140

Legume intercropping

 − 0.092**

0.038

0.034

0.097

Growing season rainfall (mm)

0.000*

0.000

0.000

0.000

Start of wettest dekad in 2018

 − 0.017**

0.008

 − 0.019*

0.010

Household size

0.027

0.029

0.047

0.034

Dependency ratio

0.043***

0.014

0.042***

0.016

Gender

 − 0.001

0.001

 − 0.001

0.001

Age

0.156***

0.051

0.161***

0.056

Education

 − 0.004

0.004

 − 0.007*

0.005

Off farm

 − 0.044

0.033

 − 0.031

0.041

Wage work

 − 0.104**

0.053

 − 0.109*

0.063

Extension

0.093**

0.042

0.105**

0.048

Own plot

0.032

0.039

0.055

0.043

Erosion control

 − 0.132

0.139

 − 0.215

0.139

Steep slope

 − 0.033

0.038

 − 0.039

0.041

Irrigation

0.244***

0.094

0.268***

0.099

Distance to market

0.000

0.000

 − 0.001

0.000

Nutrient constraint

 − 0.148***

0.025

 − 0.134***

0.028

NDVI in 2018

1.805***

0.406

1.635***

0.561

Distance to household

0.002

0.004

0.002

0.004

Wealth Index

0.078***

0.030

0.093***

0.036

Agric capital index

0.019

0.015

0.030

0.019

Tropical livestock unit

 − 0.010***

0.004

 − 0.008

0.005

North Central

0.011

0.139

 − 0.028

0.162

North East

 − 0.398***

0.140

 − 0.409**

0.170

North West

0.040

0.144

0.156

0.195

South East

0.278*

0.156

0.268

0.183

South West

0.613***

0.164

0.545***

0.194

Constant

11.856***

0.283

11.966***

0.374

Sample size

5,616

 

5616

 

R2

0.070

 

0.050

 

Wald

15.8***

 

10.54***

 

Underidentification test

  

136.19**

 

Weakidentification test

  

2.673***

 

Overidentification test (Hansen J statistic)

 

56.60

 
  1. *, ** and *** indicate statistical significance at 10, 5 and 1%, respectively