3.1. Socio-economic and demographic characteristics of sample households
3.1.1. Distributions of Sample Households by Continuous Variables
The mean age of irrigation-user and non-irrigation user household head were found to be 41.8 and 40.4 years with the standard deviations of 8.59 and 10.29, respectively. According to the results, there was no significant difference in the mean age of the household heads between irrigation-user and non-irrigation user household. This indicates that, similar distribution of age between irrigation users and non-users. When we compare the average household sizes between irrigation users and non-users, average family size for irrigation-user is 4.65 (± 1.48) persons and 4.2 (± 1.47) persons for non- irrigation users. The t-test revealed that mean difference between the two groups with regarding family size is statistically significant at 10 percent significance level. This implies the household who has large family size can best utilize irrigation than those who do not have.
The mean land holding for users of irrigation is 1.57 (± 0.54) hectares and the corresponding figure of land holding for non-users of irrigation is 1.00(± 0.44) hectares. The t-test revealed that means difference between the two groups with regarding land holding sized is statistically significant at 1 percent significance level. The mean livestock holding in Tropical Livestock Unit (TLU) for the sample households is 2.49, where the minimum is 0.00 and the maximum is 20. Mean livestock holding for irrigation user household heads is 3.13 (± 1.19) TLU and mean livestock holding for non-users of irrigation is 1.87 (± .91) TLU (Table 3). The t-test result showed that the mean comparison of the two groups with regard to livestock holding is statistically significant at 1percent significance level. The average distance of irrigation users to village is 0.64(± 0.45) hrs.; the corresponding figure of non-user household is 1.87(± 0.85) hrs. The t-test result for mean difference of the two groups with regard to distance to the irrigation scheme is statistically significant at 1percent significance level.
Table 2
Distributions of Sample Household Head by Continuous Variables
Variables
|
Total
|
Irrigation-user
|
Non-irrigation user
|
|
|
Mean
|
Std.
|
Mean
|
Std.
|
Mean
|
Std.
|
t-value
|
p-value
|
Age
|
41.1
|
9.49
|
41.8
|
8.59
|
40.4
|
10.29
|
0.52
|
0.37
|
Family size
|
4.43
|
1.48
|
4.65
|
1.48
|
4.2
|
1.47
|
1.75
|
0.08*
|
Size of cultivated land
|
1.28
|
.56
|
1.57
|
.54
|
1.00
|
.44
|
6.70
|
0.0000***
|
Livestock holding
|
2.49
|
1.22
|
3.13
|
1.19
|
1.87
|
.91
|
6.94
|
0.0000***
|
Distance to water
|
1.27
|
.92
|
.64
|
.45
|
1.87
|
.85
|
-10.5
|
0.0000 ***
|
Frequency of extension contact
|
1.62
|
1.1
|
2.22
|
1.01
|
1.05
|
.89
|
7.13
|
0.0000***
|
Source: Own survey result, 2019 *, *** means significant at 10%, and 1% significance level respectively |
3.1.2. Distributions of Sample Household Head by Categorical Variables
The survey result showed that out of the total 137 sampled households, 64.23% and 33.77% were male headed and female headed respectively. The result also indicated that 41(61.19%) and 26(38.81%) of irrigation user households and 47(67.14%) and 23(32.86%) of irrigation non-user households were male headed and female headed households respectively. This implies that sex distribution among irrigation users and non-users were similar.
The comparison by use of irrigation showed that 46 (68.66%) of irrigation users and 53 (75.71%) of non-users of irrigation are illiterate and 21(31.34%) of users and 17(24.29%) of non-users are literate in the study area. The chi-square test (χ2 = 0.85) shows that there is no significant relationship between use of irrigation and level of education.
The comparison by use of irrigation showed that 44 (65.67) users and 25 (35.71%) non-users take credit. From irrigation users of sample household heads, 34.33 percent of the sample respondents and from the non-user, 64.29 percent households said that they don’t take credit and complained for its high interest rate. The chi-square test result (χ2 = 12.28) indicated that there is statistically significant relationship between the use of irrigation and access to credit at 1 percent significance level.
According to the survey result 49 (73.13%) of users and 29(41.43%) of non-users get training and the corresponding figure that shows have no access to training of users and non-users is 18(26.87%) and 41(58.57%) respectively. There was statistically significant difference at 1 percent between irrigation users and non-users in terms of the distribution of households who received training on irrigation practice.
Table 3
Distributions of Sample Household Head by Categorical Variables
Variables
|
|
Total
|
Irrigation-user
|
Non-irrigation user
|
|
|
Category
|
N
|
%
|
N
|
%
|
N
|
%
|
χ2-value
|
P-value
|
Sex
|
male
|
88
|
64.23
|
41
|
61.19
|
47
|
67.14
|
0.52
|
0.468
|
Female
|
49
|
33.77
|
26
|
38.81
|
23
|
32.86
|
Education
|
Literate
|
38
|
27.74
|
21
|
31.34
|
17
|
24.29
|
0.85
|
0.356
|
illiterate
|
99
|
72.26
|
46
|
68.66
|
53
|
75.71
|
Access to training
|
Yes
|
29
|
41.3
|
49
|
73.13
|
29
|
41.43
|
14.03
|
0.000***
|
No
|
59
|
43.07
|
18
|
26.87
|
41
|
58.57
|
Access to credit
|
Yes
|
69
|
50.36
|
44
|
65.67
|
25
|
35.71
|
12.28
|
0.000***
|
No
|
68
|
49.64
|
23
|
34.33
|
45
|
64.29
|
Access to market
Information
|
Yes
|
104
|
75.91
|
62
|
92.54
|
42
|
60.00
|
19.82
|
0.000***
|
No
|
33
|
24.09
|
5
|
7.46
|
28
|
40.00
|
Source: Own survey result, 2019 **, *** means significant at 5 and 1% level respectively |
3.2. Determinant Factors Affecting Farmer’s Participation on Small-scale Irrigation
Results presented in Table 7 below showed that the estimated model appears to perform well for the intended matching exercise. The likelihood ratio chi-square value 73.50 revealed that the overall fitness of the model was found significant at 1% significance level. This indicates the model’s estimate fit the data at an acceptable level. Moreover, the small value of Pseudo-R2 (0.77) showed that irrigation user households do not have much distinct characteristics over non-user households and as such finding a good match between irrigation user and non-user households becomes easier.
Out of the total 12 independent variables only six variables were significant such as, Households with large livestock holding, large cultivable land and, and frequency of extension contact were identified using probit regression found to have positive relationship and significantly affecting probability of participation in small-scale irrigation. However, age of the household, distance to FTC and irrigation water sources was found to influence participation in small-scale irrigation negatively and significantly.
Table 4: Probit Model for the Determinants of Participation in Small-scale Irrigation
Variable
|
dy/dx
|
Std. Err.
|
z
|
P>z
|
Agehh
|
-.0379603
|
.01182
|
-3.21
|
0.001***
|
educlehh*
|
.0467409
|
.28352
|
.16
|
0.8690
|
Famiz
|
.0804611
|
.05952
|
1.35
|
0.176
|
Ttlansize
|
.7467513
|
.18006
|
4.15
|
0.000***
|
Tlu
|
.2320531
|
.07712
|
3.01
|
0.003***
|
credit*
|
.0896396
|
.19173
|
0.47
|
0.640
|
Freqextncont
|
.2298971
|
.09145
|
2.51
|
0.012 **
|
Disftc
|
-.0135482
|
.00607
|
-2.23
|
0.026 **
|
Diswars
|
-.8531831
|
.15865
|
-5.38
|
0.000***
|
partsa~g*
|
.2510201
|
.18769
|
1.34
|
0.181
|
annino~t
|
.0000327
|
.00003
|
1.01
|
0.313
|
accmar~o*
|
-.0121527
|
.22704
|
-0.05
|
0.957
|
Number of obs. = 137
|
Wald chi2(12) = 73.50
|
Prob > chi2 =0.0000
|
Log pseudo likelihood = -21.494
Pseudo R2 =0.7740
|
Source: Own survey result, 2019 **, *** means significant at 5% and 1% significance level respectively
Age of the Household Head (AGEHH)
Consistent with that of our prior expectation the age of household head influences the probability of participation in small-scale irrigation negatively and significant at 1% level of significance. The marginal effect of age indicated that as a farmer’s age increases by 1 year from the mean value, the probability of participation in small-scale irrigation decreased by 3.7 percent. This may be because the use of irrigation is labor intensive and exhaustive work that the older household heads cannot tolerate this challenge.
Distance of Residence from Irrigation Scheme (DISWARS)
Consistent with a priori expectation this variable influence small-scale irrigation participation decision negatively and statistically significant at 1% significance level and. The negative relationship tells us that when the household head’s residence is far from the irrigation scheme, the household heads have less participation in irrigation. The marginal effect of this variable shows that as the distance from the farmers’ residence to the water source increases by 1hour from the mean value, the probability of participation in small-scale irrigation decreases by 85.3 percentage points, keeping all other variables constant. The finding of this study is similar to the findings by Kinfe et al. (2012) that household’s residence to water sources is found to have a significant and negative relationship to the probability of participation in small scale irrigation.
Distance to Farmers Training Center (DISFTC)
as expected, coefficient of this variable has negative relationship with irrigation participation and it is statistically significant at 5% significance level. The marginal effect of this variable indicated that as the walking distance to the farmer training center from the residence increases by 1hour from the mean value, the probability of participation in small-scale irrigation decreases by 1.3 percent. This indicates that the longer the distance from home to the farmer training centers and/or development agent offices, the lower is the probability to start and use irrigation. This is consistent with Moll (2004, found that the farmer training center is a source of information for the Tibia resident.
Size of Cultivated Land (TTLANSIZE)
Consistent with a priori expectation farm size influences the probability of participate in small-scale irrigation positively and significant at 1% significance level. The marginal effect of this variable indicates that as the size of cultivated land increases by one hectare from the mean value, the probability of participation in small-scale irrigation increases by 74.6 percentage points, keeping all other variables constant. This means that households who have more land are more likely to participate in small-scale irrigation as compared to households who have less land. This result is consistent with the finding of Mohammed and Jema (2013) as well as Martey et al (2013) who also obtained that farm size influenced the household heads decision to participate in agricultural projects.
Livestock Holding (TLU)
Consistent with a priori expectation livestock holding influences the probability of participation in small-scale irrigation positively and significant at 1% level of significance. The marginal effect shows that as the number of livestock in tropical livestock unit increases by one unit from mean value, the probability to participate in small-scale irrigation increases by 23.2 percentage points, keeping all other variables constant. This relationship implies that household with more livestock possession might have the capacity to generate cash income to purchase input and could be able to take more risk associated with use of irrigation. The same result was reported by Desale (2008) and Yenetila (2007) that livestock holding has positive influence on participation decision in small-scale irrigation.
Frequency of Extension Contact (FREQEXTNCONT)
This variable is consistent with the original hypothesis it has positive association with irrigation participation and significant at 5% level of significance level. The marginal effect shows that as frequency of extension contact increases by one unit from mean value, the probability to participate in small-scale irrigation increases by 22 percentage points, keeping all other variables constant. The possible explanation is that development agents are the major sources of information, provides technical support, advice on use of the availability of technology, facilitate the use of input to farmers, encourage for extension package participation and other development issues for rural farmer which may have positive aggregated impact for enhancing production and productivities. This result is consistent with that of reported by Ransom et al.(2003) and Ouma, (2002).