Sources of wheat production technical inefficiency among smallholder farmers in Northwestern Ethiopia: Beta regression approach

Figures & data

Figure 1. Map of the study area.

Table 1. Description, measurement, and expected signs of independent variables

Dependent variables	Independent variables	Unit of measurement
Wheat production	Labor	Man equivalent
	Oxen	Number
	Wheat farm size	Hectare
	DAP	Kilogram
	Urea	Kilogram
	Seed	Kilogram
Technical inefficiency	Sex of respondents	1 if male, 0 otherwise
	Educational status	1 if literate, 0 otherwise
	Farm experience	Year
	Dependency ratio	Number
	Livestock ownership	TLU
	Distance to all weather road	Minute
	Distance to the local wheat market	Minute
	Household income	USD^1
	Distance to the extension office	Minute
	Credit access	1 if user, 0 otherwise
	Access to wheat demand information	1 if yes, 0 otherwise
	Access to wheat price information	1 if yes, 0 otherwise
	Access to wheat supply information	1 if yes, 0 otherwise

Table 2. Descriptive statistics of categorical and continuous variables

Variables	Count	Percent	Mean	SD^2	Minimum	Maximum
Sex of respondents	268	66.9
Educational status	127	31.8
Credit access	132	33.1
Wheat demand information	223	55.8
Wheat supply information	125	31.2
Wheat price information	234	58.4
Dependency ratio			0.53	0.34	0	1.5
Household income			286.93	323.10	23.37	1883.12
Farming experience			23.79	12.99	3	50
Livestock ownership			4.35	2.05	1.15	10
Distance to the extension office			35.75	18.03	5	60
Distance to all weather road			24.35	13.83	15	65
Distance to the local wheat market			42.08	19.58	10	120

Source: Survey data, 2022

Table 3. Descriptive statistics of input and output variables

Variables	Mean	SD	Minimum	Maximum
Labor	4.12	1.94	2.15	9.25
Number of oxen	4.01	1.99	1	8
Wheat farm size	1.28	0.86	.25	5
Amount of DAP used	138.84	119.65	19	525
Amount of urea used	153.81	121.93	25	550
Amount of seed used	154.41	129.66	20	600
Amount of wheat produced	3801	2591	900	12,500

Source: Survey data 2022

Table 4. Stochastic frontier model result of wheat production

Variables	Coefficients	Standard error	T-statistic	P-value
Lnlabor	0.035	0.026	1.350	0.177
Lnoxen	0.118***	0.034	3.500	0.000
Lnlfarm	−0.214***	0.074	−2.890	0.004
LnDap	0.076	0.129	0.590	0.552
Lnurea	0.523***	0.159	3.290	0.001
Lnseed	0.374***	0.060	6.240	0.000
Constant term	−0.018	0.056	−0.330	0.742
Gamma				0.833
Sigma square				0.091
Log-likelihood				206.25***
Returns to scale				0.912
Actual wheat production				38.01
Potential wheat production				56.98
Wheat production gap				18.97
Monetary value of wheat production gap				72, 109.44

Source: Survey data 2022; *** denotes significance level at p < 0.01

Figure 2. Technical efficiency score distribution of smallholder farmers (N = 400).

Table 5. Descriptive statistics results of the technical efficiency level of wheat production in the study area

Descriptive statistics	Technical efficiency
Minimum	0.318
Maximum	0.969
Mean	0.77
Standard deviation	0.213

Source: Survey data 2022

Table 6. Beta regression model output on sources of wheat production technical inefficiency

Variables	Coefficients	T-statistic	P-value	Marginal effect
Sex of respondents	−0.015	−0.260	0.799	−0.001
Educational status	−1.684*	−1.840	0.066	−0.099
Farm experience	−0.066***	−6.340	0.000	−0.004
Dependency ratio	0.980***	4.500	0.000	0.057
Livestock ownership	−0.233	−0.310	0.755	−0.014
Distance to all weather road	0.073	0.400	0.688	0.004
Distance to the local wheat market	2.580***	7.400	0.000	0.151
Household income	−0.048	−1.440	0.149	−0.003
Distance to the extension office	0.025***	3.610	0.000	0.001
Credit access	−0.077	−1.270	0.204	−0.005
Wheat demand information	−0.153	−0.830	0.408	−0.009
Wheat price information	−1.040***	−5.620	0.000	0.061
Wheat supply information	−0.071	−1.180	0.238	−0.004
Constant	−10.806	−14.970	0.000
Number of observations				400
Log-likelihood value				404.88
Chi-square value				184.50***

Source: Survey data 2022; *** and * denote significance level at p < 0.01 and 0.1, respectively