4.3.1.
Variables in the equations
Table 4.25, variable in the equations
Exp(B) 95% C.I. for
B S.E. Wald Df Sig. EXP(B)
Odds Ratio Low Upper
Ste Sex 2.363 .923 6.562 1 .01 10.626 1.74er 64.806
p 1a 0 2
Age .146 .051 8.127 1 .00 1.157 1.04 1.279
4 7
Marital status 2.819 2 .24
Marital status(1) - .883 2.793 1 .094 .229 .040 1.290
1.476 5
Marital status (2) -.701 .956 .538 1 .46 .496 .076 3.229
family size -.907 .423 4.605 1 .033 .404 .176 .924
2
Educational 4.357 1.190 13.407 1 .00 77.987 7.57 803.126
background
land size 13.747 6 .030 3
land size (1) -.482 1.867 .067 1 .793 .618 .016 23.976
6
land size (2) 3.488 1.898 3.379 1 .06 32.715 .794 1348.752
6
land size (3) 5.286 1.903 7.716 1 .00 197.500 4.74 8227.126
land size (4) 3.914 2.552 2.352 1 .125 50.116 .3371 7453.985
5
land size (5) 5.098 8.962 .324 1 .56 163.675 .000 6961868640.
land size (6) - 99.959 .002 1 .969 .008 .000 9.771E+ 566
4.824 2 82
Membership of 1.803 .856 4.432 1 .03 6.067 1.13 32.505
cooperative 5 2
Distance to 13.796 3 .00
financial institution 3
Distance to - 1.132 2.478 1 .11 .168 .018 1.548
financial institution 1.782 5
(1)
Distance to - 2.841 13.106 1 .00 .000 .000 .009
financial in 10.28 0
Distance to - 4.536 3.406 1 .06 .000 .000 1.680
financial institution 8.370 5
(3)
Training(1) 1.749 .820 4.549 1 .03 5.751 1.15 28.698
Annual income .001 .000 17.300 1 .003 1.001 1.002 1.001
0 0
Annual expenditure -.001 .000 16.531 1 .00 .999 .999 1.000
0
Constant - 3.297 14.628 1 .00 .000
12.60 0
Source: SPSS version 26 binary logistic regression model outputs.
The result of the binary logistic model in the above table 4.25 shows that, ten variables were
significantly influence household saving behavior: sex, age, family size, educational
background, land size, membership in a cooperative, distance to a financial institution,
training, annual income, and annual expenditure have been significant. marital status was not
significant influence on household saving behavior of the study area. Each of the variables is
discussed below.
Sex of the household head
Sex is one of the variables that can affect rural household saving behavior. The econometric
result shows a positive and significant impact on household saving behavior. The odds ratio
indicates that the odd of male saver household heads is 10.626 times more likely to be a saver
than the odd of female saver household heads, controlling for other variables in the model.
(Coff 2.363, OR 10.626, p=.010) This result shows that households headed by males had a
positive significant impact on household saving behavior. Male individuals had better-saving
behavior than females because male household heads are in a decision-making position in
economic terms in the study area. This findings supported the conclusion of previous studies
(Kibet et al., 2017; Obayelu 2012; Zone & Borko 2018) sex of household head significant
influence on the saving behavior of households.
Age of the household head
Age is a positive and significant factor that influences saving behavior. The odds ratio of this
variable indicates that the age of household head increases by one year, 1.157 times more
likely to the probability of saver than its counterparts, holding all other variables constant in
the model. (Coff 0.146, OR 1.157, p=.004, CI 1.047-1.279). This result indicates that age
increase the probability of household saving also increase. Thus, household heads with
advanced ages chance to saver than households with less advanced ages. This result similar in
(Markos2015; Zone & Borko 2018)
Family size
Family size had a significant negative impact on household saving behavior. As the family
size of the household head increase the probability of savers accordingly decreases. The odds
ratio indicates that for each one number increase of family members, 0.404 times less likely
to be the probability of a saver than its counterparts, the remaining variables constant in the
model. (Coeff-.907, OR 0.404, P = 0.032, CI 0.176, 0.924). This indicates that households
with a large number of family members are less likely to save than those with a small number
of family members. There is an inverse relationship between family size and saving behavior.
This finding is consistent with previous studies (Samantaraya & Patra 2014; one & Borko
2018)
Educational background
Educational background is an essential personality that determines the farmer’s ability to
communicate, acquire information and undertake new technology. This variable has a
positive and significant impact on saving behavior. The odds ratio indicates that literate
household heads are 77.987 times more likely to odd of saver than illiterate household heads,
holding all other variables constant in the model. (Coff 4.357, OR 77.987, p=.000) This result
implied that educated household heads had a better chance to savings than non-educated
households. Because educated households would have greater access to information,
households can simply understand the value of saving. This result support the previous
studies (Hailu 2020; Zone & Borko 2018).
The land size
Land size has a significant and positive impact on household saving behavior. The odds ratio
of land size indicates that households that have 1.6-2.0 hectares of land were 197.500 times
more likely to have a chance of saver than households that have less than 0.5 hectares of land,
keeping the other variables constant. (Coff 5.286 OR 197.500, P=0.005). The other
categories of land size contrast are fell because of statistical insignificance. This result
indicates that households with large land sizes have better saving behavior than households
with small land sizes. This finding consistent with the previous studies (Duressa & Ejara
2018; Samantaraya & Patra 2014),
Membership of cooperative
Membership of a cooperative has a significant positive impact on household saving behavior.
The odds ratio of this variable indicates that a household’s membership of a cooperative was
6.067 times more likely to be a saver than a non-membership of a cooperative household,
keeping all other factors constant. (coeff 1.803 OR 6.067, P=0.035). The result revealed that
cooperatives play an important role in households and in developing strategies for
encouraging or mobilizing saving. Membership of a cooperative improves the saving
behavior. This result is similar in previous studies (Obayelu 2012).
Distance from the financial institution
Distance from financial institutions is found to have a negative and significant effect on
saving behavior. The odds ratio of this variable indicates that a distance of 4-7 km from
financial institution has 10.3 present lower probability of saver than a distance of less than 3
km from financial institution, keeping the other variables constant. (coeff -10.286, P=0.003).
The other contrast is fell because there are no statistical differences. This finding in line with
previous studies( Zone & Borko 2018; Duressa & Ejara 2018)
Training
Training helps to enhance human capabilities by improving the way people think and live.
The variables were found to be a positive and significant impact on household saving
behavior. The odds ratio indicates that household heads that have attained training are 5.751
times more likely to have a probability of saver than those households’ heads that have not
attained training, all other variables being constant in the model. (Coff 1.749, OR 5.751,
p=0.033) This result indicated that the household head's training helped households to obtain
information, correct misconceptions about savings, and improve the savings behavior of rural
households. This result supported previous studies (Teshome et al., 2013; Dires, 2021)
Annual income
Annual income has a statistically significant and positive impact on saving behavior. The
odds ratio of income indicates that a household's income increases by one unit, 1.001 times
more likely to probability of saver than its counterparts, all other remaining variables being
constant. (Coff 0.001, p=0.000)As a result, the household head's annual income determines
the household’s decision either to save or not to save, and households with higher incomes
are more likely to saver than households with lower incomes. That is, as household income
increases, the probability of savings also increases. This study suggests that income is an
important factor in influencing saving behavior. This result similar to pervious empirical
studies (Saqib & Ali, 2016;Obayelu 2012)
Annual expenditure
Annual expenditure was found to be has a statistically significant and negative impact on
saving behavior. Household expenditure odds ratio indicates that household expenditure
increasing by one unit 0.999 times reduce the likelihood being a saver when compared to
counterparts controlling other variables in the model. . (Coff -.001, OR .999, p=0.000)This
implies that households with higher expenditures are less likely to saver than households with
lower expenditures. These findings similar to (Goshim Azeref 2018; Abiyot Alemu 2021)
Marital status of household head
The Marital status of the household heads is one of independent variables expected to affect
their saving behavior. According to the results of the study, marital status was no significant
impact on household saving behavior. This result is inconsistence with previous research
(Zone & Borko 2018; Ansori 2015; Markos 2015)