Adoption of rice cultivation technologies and its effect on technical efficiency in Sagnarigu District of Ghana

Figures & data

Table 1. Descriptive statistics of variables

Variable description	Mean	Min	Max
Age (in years)	33.35	18	64
Fertilizer use (in kilogrammes)	411.4	50	1,235
Rice output (in kilogrammes)	3,440.5	100	16,302
Education (No. of years in school)	1.5	0.0	12
Farm size (in hectares)	1.0	0.5	4
Household labour (No. of persons)	2.9	1.0	16
Contract farming (dummy, 1 = yes, 0, otherwise)	0.43	0.0	1
Seed used (in kilogrammes)	8.2	0.5	40
Weedicides used (in litres)	3.1	0.0	15
Agricultural extension (dummy, 1 = access, 0, otherwise)	0.7	0.0	1
Training in rice cultivation (dummy, 1 =access, 0, otherwise)	0.7	0.0	1
Farmer group association (dummy, 1 = yes, 0, otherwise)	0.6	0.0	1
Fertilizer subsidy (dummy, 1=access, 0, otherwise)	0.5	0.0	1
Credit dummy, 1= access, 0, otherwise)	0.1	0.0	1
Gender (dummy, 1 =male, 0, female)	0.7	0.0	1

Source: Authors’ Computation, 2017.

Table 2. Maximum likelihood estimates of the poisson model

Variable	Coefficient.	Standard error	p value
Constant	0.244	0.1681	0.147
Contract farming	0.636***	0.119	0.000
Farmers group	0.421***	0.145	0.004
Farm size	−0.004	0.079	0.955
Access to agric. extension	0.674***	0.178	0.000

Source: Authors’ Computation, 2017.

*** Means statistically significant at 1%.

Table 3. Test for choice of functional form and presence of inefficiency

Test	Null hypothesis	Log likelihood function (H0)	Test statistic λ	Critical value	Decision
Choice of functional form	H0:β1 = … = β20 = 0	−143.703	32.119	12.592 (15)	Reject H0: Translog appropriate
Presence of inefficiency term	$H_0:{\delta }_1=\cdots ={\delta }_6=0$	−110.3105	30.369	24.996(6)	Reject H0

Notes: Critical values are at 5% significance level and obtained from χ² distribution table. Figures in brackets are the number of restrictions.

Table 4. Maximum likelihood estimates of the translog and inefficiency models

Variable	Coefficient	Standard error	p-value
Constant	8.330	0.00188	0.000
Household labour	0.269***	0.0217	0.000
Seed	−0.058***	0.00736	0.000
Weedicides	0.517***	0.00713	0.000
Fertilizer	0.235**	0.0846	0.005
Farm size	0.228***	0.0535	0.000
Household labour squared	−3.005***	0.755	0.000
Seed squared	−0.267***	0.00318	0.000
Weedicides squared	0.227	0.209	0.278
Fertilizer squared	−0.067	0.050	0.183
Farm size squared	−0.847**	0.518	0.102
Household labour*Seed	0.102***	0.0182	0.000
Household labour*Weedicides	−0.325**	0.131	0.013
Household labour*Fertilizer	−0.372***	0.089	0.000
Household labour*Farm size	0.797***	0.204	0.000
Seed*Weedicides	0.199**	0.064	0.002
Seed*Fertilizer	0.081***	0.0162	0.000
Seed*Farm size	−0.172**	0.069	0.013
Weedicides*Fertilizer	0.617**	0.369	0.094
Weedicide*Farm size	−0.645***	0.099	0.000
Fertilizer*Farm size	0.109	0.389	0.779
Inefficiency term
Constant	−1.403	0.395	0.000
Gender	0.392	0.324	0.227
Access to credit	1.689***	0.451	0.000
Access to agric. extension	1.622***	0.448	0.000
Predicted value adoption	−0.558***	0.130	0.000
Sigma square	−33.612*	122.426	0.061
Gamma	0.274***	0.022	0.000
Returns to scale	1.191
Log likelihood	110.311

Source: Authors’ Computation, 2017.

* Means statistically significant at 10%.

** Means statistically significant at 5%.

*** Means statistically significant at 1%.

Figure 1. Average efficiency and access to agricultural extension.

Figure 2. Average efficiency and adoption of rice production technologies.

Table 5. Technical efficiency scores

Efficiency range	Adopters		Non-adopters		Pooled Data
	Frequency	Percent	Frequency	Percent	Frequency	Percent
≤ 0.50	23	38.33	37	61.67	60	50
0.51–0.60	6	10	6	10	12	10
0.61–0.70	7	11.67	4	6.67	11	9.17
0.71–0.80	5	8.33	3	5	8	6.67
0.81–0.90	3	5	3	5	6	5
0.91–1.00	16	26.67	7	11.67	23	19.17
Total	60	100	60	100	120	100
Mean TE	0.62		0.46		0.54
Min.	0.0007		0.008		0.00074
Max.	1.00		1.00		0.99
Standard deviation	0.298		0.293		0.31

Source: Authors’ Computation, 2017.

Table 6. Results of effect of adoption of cultivation technologies on rice output

	Propensity score matching (PSM)		Nearest neighbor matching (NNM)		Regression adjustment^a
	Coefficient	Std. Error	Coefficient	Std. Error	Coefficient	Std. Error
ATE	0.101*	0.061	0.101**	0.060
ATT	0.121*	0.720	0.121**	0.731
Mean efficiency adopters					0.584***	0.039
Mean efficiency non-adopters					0.479***	0.053

Source: Authors’ Computation, 2017.

* Indicate statistical significance at 10%.

** Indicate statistical significance at 5%.

*** Indicate statistical significance at 1%.

^a Similar results obtained using inverse probability weights.