Global Journal of Commerce & Management Perspective
Open Access

Determinants of Household Rural Poverty in Tenta Woreda, South Wollo Ethiopia

Ayisheshim Bekele Abebe^* and Mengesha Kassa ^*Correspondence: Ayisheshim Bekele Abebe, Department of Geography and Environmental Studies, Debre Markos University, Debre Marqos, Ethiopia, Email:

Author info »

Introduction

One of the most significant problems today is the uneven distribution of wealth and resulting impoverishment of certain areas of the world. Accordingly, Shah pointed out that a few get wealthy while the majority struggle in the global context as a result of the enormous external influence, behind the internal reasons, of global decisions, policies and practices by the leaders of rich countries and other global actors in the name of globalization that makes the governments of poor nations and their people often powerless.

At the current situations, there is a wide gap of wealth between the poor and rich countries of the world. For example, out of the world GDP of $ 75.5 trillion for the year 2016, about 48.44% was accounted by the major advanced economies (G7) and European Union. In contrast, 13% (954 million) people of less developed countries shares USD 952 in the same year from the above total GDP that means these regions has less than 9% GDP of the world average. Besides this inequality, the world poverty reveal a higher number of people live in poverty than previously known. For instance, world bank revealed the 1.2 billion people live on or below the measure of the new poverty line of $ 1.25, nearly half of the world population (over 3 billion) live on less than $ 2.5 a day, and at least 80% of humanity lives on less than $ 10 a day [1].

The poorest people have less access to nutrition, health, education, sanitation and water as well have little representation or voice in public and political debates/decisions. Thus, 22,000 children (UNICEF) die each day due to poverty; about 28% of all children of primary school age in the developing world were out of school, 1.1 billion in developing countries had not adequate access to water and 2.6 billion lack basic sanitation, 640 million out of 1.9 billion children (1 in 3) were without adequate shelter, 270 million (1 in 7) with no access to safe water, 2.5 billion people were forced to rely on biomass fuel wood, charcoal and animal dung to meet their energy needs for cooking [2].

According to Begna and Paul, poverty in most parts of the world remains mainly a rural phenomenon where agriculture is the dominant stay of livelihood. The large and persistent gap between the share of agriculture in GDP and employment suggests that poverty is concentrated in the agricultural and rural areas, i.e., many of the rural poor are likely to remain poor while the non-agricultural growth accelerates. This is evidenced by the fact that rural areas account for three in every four people living on less than $ 1.0 a day [3].

Regionally, poverty is more concentrated and shifting towards South Asia and Sub-Saharan Africa with 80% of people in these regions still surviving on less than US $1.25 a day, but it is more severe and worse in Sub-Saharan Africa (47% of the population) than South Asia (with 33%). In 2010, the HDI of Sub-Saharan Africa with the value of 0.460 stood the lowest as compared to South Asia and the world average of 0.697. According to the world bank estimate based on current trends, more than 0.6 billion people would live in extreme poverty in 2015 while 90% of them would be in South Asia ( with 216 million) and Sub- Saharan Africa (340 million which accounts more than 56% of the world) [4].

Literature Review

Ethiopia, being one of the least developed countries of Sub- Saharan Africa, has a population of approximately 107.53 million, up from 2015’s estimate of 98.9 million with a mosaic of ethnic group and the area of 1,104,300 million km², with complex land features, diverse climate and a wide variety of fauna and flora. About 80.08% of the population is rural and the economy is highly dependent on agriculture which has long history with relatively better success than its today’s performance. In line with this, Ethiopia had been self-sufficient in food production and was categorized as a net exporter of food grain till the late 1950’s. But from the 1960’s onwards the food gap ever widened and Ethiopia becomes a net importer [5].

By understanding this challenge the current government implemented the ADLI strategy to be self-sufficient in food production and to ensure food security in long run. According to Hussen, the government of Ethiopia gives more attention to reduce poverty through increasing productivity in the agriculture sector by adopting ADLI with the objective to match population growth and productivity as well as offer technical and institutional support to farmers. In line with this, Ethiopia stepped-up on a structural adjustment programs during the 1990’s which have changed to Ethiopian Sustainable Development and Poverty Reduction Program (SDPRP) with the aim to follow the UN millennium development goals by creating more market oriented and less state-dependent agriculture. Despite of the above strategic efforts, food insecurity as well as pervasive and structural poverty are appalling in the country due to land degradation, erratic conditions, depletion of natural resources, shortage of land and other household assets, limited opportunity for livelihood diversification, limited capacity of the government to introduce new farming technologies through adequately organized extension services and inability of agricultural production to pace with the rapid population growth [6].

Different studies that have been carried out in different parts of Ethiopia shows the spatial variation of poverty in the country and argued Wollo as one of the most impoverished areas. By taking these into account, the researcher was focused in Tenta Woreda, which is part of Wollo and has been repeatedly exposed to recurrent drought and livelihood crisis, to investigate the level and determinants of rural poverty at the household level [7].

The study is survey research design. Both quantitative and qualitative approaches were used symbiotically to cancel the limitations of a single one, understand social phenomena and to strengthen the finding and relevance of the study. Both primary (quantitative and qualitative) and secondary data (published and unpublished sources) was gathered using questionnaire (closed and open ended questions) in English and Amharic version languages tested by pilot study and then administered by trained data collectors. In-depth interview was administered through semi-structured schedule from Woreda level authorities and focus group discussion. Samples were selected by considering agro-climatic zones of peasant associations to account spatial variation and stratifying 31 peasant associations (rural kebeles) from sampling frame into three categories. Then, a sample of one peasant association from each agro-climatic zones selected using simple random sampling technique (Tables 1 and 2). Lastly, the total sample is determined by using the following formulae [8].

z²pq

n=E2

Where; n is the number of samples

p is assumed incidence of chronic poverty by taking value for Wollo (15% or 0.15%)

q is 1-p, i.e., 0.85

E is the allowable error (0.05)

z is the Z-score of 95% confidence (1.96)

Table 1: Sample households in each peasant associations.

Table 2: The list of independent variables.

Here, the magnitude of chronic poverty of Wollo which is 15% or 0.15% used as the p value to determine the minimum sample size since there is no other study in the particular study area on poverty. Accordingly, the number of the sample households was 196 proportionally out of the total 3108 households of the three peasant association (Table 3).

Table 3: Household demographic characteristics.

In addition eight key informants (six agricultural development agents and two head WARDO and Woreda administrator) and seven focus group discussion participant (male household head who conducts and do not conducted none/off-farm activity and a woman household heads from each agro-climatic zone) were chosen purposively [9].

The researcher used both quantitative and qualitative methods of data analysis. The researcher analyzed extent and magnitude of rural poverty by using CBN method was which is preferred over the others as it offers a monetary value of a poverty line that accounts both for the food and non-food components. The researcher also measured the level of rural poverty (incidence, depth and severity) by using Foster-Greer Thorbecke (FGT) mathematical model of poverty. The researcher also compared the magnitude of rural poverty with reference to two alternative poverty lines. One is the international standard of US $ 1.25 per day per person as estimated by the WB, the second poverty line is poverty lines consisting of 25% less than and more than the standard poverty line of the Woreda to identify severe and moderate poverty respectively [10].

The researcher also used logistic regression model like linear probability model to examine the determinants of rural poverty due to possibility of very large prediction errors and certaintyprone prediction of outcomes as it has powerful predictive power; is less restrictive and less sensitive to outliers; easy to correct a bias; and has no formal requirement for multivariate normality, homoscedasticity, or linearity of the independent variables. Finally, the result of the analysis was presented by using different statistical diagrams such as tables, graphs and piecharts; and then discussed in terms of paragraph [11].

Model specification

Logistic regression model was employed, with the dependent variable being the dichotomous variable of whether the household is poor (1) or non-poor (0) to analyze the determinants of rural household poverty. It can be used to predict a response variable on the basis of continuous, dichotomous or a mix of any of predictor variables to determine the percent of variance in the response variable explained by the predictor variables; to assess interaction effects; and to understand the impact of covariate control variables. It is also a flexible model. To this effect, this study adopted and employed binary logistic regression model (Logit model) for more than one independent variable as described here under [12].

Prob (event) = eZ ------------------------------------- (1) 1 + eZ

Or equivalently, Prob (event) = 1 ----------------- (2) 1 + e-Z

Where, e is the base of the natural logarithms, approximately 2.718

Z is the linear combination of independent variables written as

Z = Bo+B1X1+B2X2+…+BpXp ---------------------------------- (3)

Where, Bo and B1 are coefficients to be estimated from data.

X is the independent variable.

Dependent variable (y) and Explanatory variables (Xi)

Rural household poverty was the dependent variable which was dichotomous and designated by a value of zero for the poor households and one for the non-poor households. The lists of explanatory variables-which include both categorical and continuous.

Results

Respondents background

The demographic characteristics of the Woreda includes 70.41% male headed and 29.59% female headed household with average family size minimum 1 and maximum 12 children. From the same table, 63.27% of the respondent cannot read and write and the rest respondents 36.73% are literate.

Extent and magnitude of poverty

With regard to extent and magnitude of poverty (computing the three indices of poverty i.e. incidence, depth and severity), about 66.1% of the households live in absolute food poverty [13]. The incidence was found to be significantly higher than 42% for kersa kondaltity Woreda in Oromia region and 63% for Zeghe Peninsula. The food poverty gap of the Woreda was found to be 35.39%. Accordingly, the average consumption shortfall required to bring the poor to the food poverty line was found to be 35.39% of the food poverty line [14]. The gap was also found to be higher compared to 27.6 and 31.52% for Zeghe peninsula and Kara Kondaltity Woreda respectively. Food poverty severity was found to be 17.25% which is significantly higher than that of 14.73% for Zeghe Peninsula and 11.54% for Kers Kondaltity Woreda (Table 4).

Table 4: Extent and magnitude of food poverty.

The extent and magnitude of total poverty

The total poverty incidence of the Woreda was found to be 67.3%, which is higher than those of Zeghe Peninsula, 40.5% for Kersa Kondaltity Woreda respectively. Poverty gap of the Woreda was found to be 37.43% which is significantly higher than 27.18% of Kersa Kondaltity Woreda, 12.2% of the national average and 32.8% of Zeghe Peninsula. Poverty severity index of Tenta Woreda was found to be 22.39%, which exceeded significantly 4.6% of the national average, 18.7% of Zeghe Peninsula and 9.89% of Kersa Kondaltity Woreda (Table 5).

Table 5: The extent and magnitude of total poverty.

Households’ demographic characteristics and poverty

The study examines the existence of any systematic relationships between certain important demographic characters such as sex, age, educational level of household heads, dependency ratio and household size and household poverty.

Age and poverty: In this study, it was hypothesized that families with younger household heads were assumed to be poorer than families headed by elders. This is because of the fact that older households heads presumed to have accumulated more assets, better farmland holding and more than those young. As shown in Table 6, out of the total 196 household heads 37.25% and 31.63% household heads were found to be between the age category of 36-45 and 46-55 respectively. The mean age for the entire sample size was 45.80 years with standard deviation of 10.955. Besides, the age of the household heads was examined among the three PAs based on agro-climate and chi-square test was run to see the association [15]. The chi-square result shows that there is no significant association of the age of the household heads among the three agro-climatic zones with the x² value of 13.229 and p value of 0.211.

Table 6: Household heads’ age and poverty.

Accordingly, the cross-tabulation of age and poverty shows that 3.79%, 11.36%, 36.36% out of the total 132 (100%) poor category whereas 1.56%, 10.94%, 39.06% out of the total 64 (100%) non-poor households were found to be within the age category of ≤25, 26-35, 36-45 respectively. This indicates that there is not much difference of poverty status (poor and non-poor) of households in different age groups. Hence, t-test was run to test the relationship between the age of the poor and non-poor and the result shows that the mean age of the poor and non-poor households was found to be 45.55 and 46.31 with the standard deviation of 11.417 and 10.001. This age variation does show statistically insignificant relationship between the two groups with the t value of -0.459 and p value of 0.647.

This is consistent with the data obtained through interview and FGD about the incidence of poverty in relation to household age in their areas through open ended question. The key informants and FGD participants affirms that poverty situation does not vary with age of household heads’ age due to the activeness of heads in lower age to participate in different activities (high labor productivity) with small family size although they have low assets; high productivity and better assets than the lower ages but large family size in medium age; and relatively higher accumulation of wealth, low family size and better work experience but low labor productivity in higher age.

Educational status and poverty: The study cross tabulated and examined the educational status of household heads among the three PAs on the basis of agro-climate by using chi-square test in order to show the association. The result shows that there is no association of literacy status among the three PAs with the chi value of 2.378 and p value of 0.305 [16]. Based on the disaggregation, the illiterates constitute 63.64% and 36.36% of the poor and non-poor while the literates (ranging from only read and write to second cycle primary education) constitute the remaining 36.36% and 37.50% of the poor and non-poor households, in order. Chi-square test was run to test whether households with illiterate heads are more prone to poverty than literate heads. The result showed that there is no significant statistical difference between the literacy status of the poor and non-poor households with the chi value 0.024 and p value of 0.877. This tells that educational status and poverty do not have relationship in the study area (Table 7).

Table 7: Household heads’ educational level and poverty.

The key informants and focus group discussion participants also made clear that the more educated household heads, although they are small in number, passed their time in education rather than accumulating farming assets like livestock and small ruminants although they try to implement better way of production. Besides this, household heads grant less farm assets like farmland for the educated newly separated members to establish a new household than the non-educated.

Household head’s gender and poverty: In this study, female-headed households’ accounts for 29.59% while the male-headed households account 70.41%. The cross tabulation of sex and PAs by using chi-square test demonstrated that there is no significant contingency of household heads’ sex among the three PAs with the chi value of 1.290 and p value of 0.525. The disaggregation also shows that the female-headed households found to account over one-third (35.61%) of the total poor, only less than a quarter (17.19%) of the total 64 non poor sample households. On the other hand, the male-headed poor households account for 64.39% of the total 132 poor households and 70.41% of the total 64 non poor households. Hence, the test by chi-square shows that there is significant systematic contingency relation between sex of the household head and their poverty status with the x² value of 7.018 and p value of 0.008 (Table 8).

Table 8: Household heads’ sex and poverty.

The interview and discussion with the focus group contend that there is variation of the poverty status of households on the basis of household heads’ sex. The tradition of the Woreda discourages women’s productivity in different ways. For instance, women expected not to be able to plow farmland as a result women plowing is strictly forbidden by the norm. They are also deprived in productive agricultural resources like land. Accordingly, women could only have access to land through marriage in the Woreda’s culture unlike giving land for the new established male household; and even a widow's land still belonged to the husband. Women, most often, do not obtain farmland from their ex-husbands at the time of divorce.

Household size and poverty

About 47.45% household heads were found to be within the family size category of 4-6 with the mean family size for the total households of 6.14 and the standard deviation of 2.533. The average family size of the study area was significantly higher than that of the regional and national average. Chi-square test was run to see if there is significant difference in family size among the three sample PAs and the result showed that there is no systematic contingency difference with chi value of 8.934 and p value of 0.177. The disaggregation also indicates that more than four-fifth (92.19%) of the total non-poor households was found to have six and less members. Contrarily, more than two-third (41.32%) of the total poor households were found to have seven and more members of a family. The t-test, which was run to test the hypothesis that households with small family size tend to be better off than those with large family size, shows that there is significant statistical difference between the household size of the poor and non-poor (t=4.953 and p=0.000) (Table 9).

Table 9: Household’s family size and poverty.

The focus group discussion participants and key informants also contend that most of the households in the study area give birth to children not only in search of their labor but also due to their perception children are gift of god and assets by themselves. However, households with large family size were highly prone to poverty due to lack of other productive assets mainly land. In addition, frequent birth has a negative impact on the economic productivity of mothers by prohibiting them from conducting agricultural and other activities in home on the one hand and it leads them to cost money for medication on the other hand. This in turn affects the wellbeing of the whole members of the household. Consequently, poverty status of households is aggravated by large family size.

Age dependency burden and poverty: The study used the year categories of ≤15, 15-64 and 65 ≤ as the age of young, adult and old respectively. Dependency burden of the Woreda was calculated by the non-working population in the young and old age groups, on the working age population. The result indicates that 27.55%, 64.29% and 8.16% out of the total 196 households were found to be in the dependency burden categories of 0-75, 75.1-150 and 150.1-225. In this study, the minimum and maximum age dependency of households was found to be 0.00 and 200.00.

Generally the age dependency ratio of the Woreda was indicates 98.4293 economically inactive persons per 100 exceeding national dependency ratio. On the basis of the disaggregation, the dependency burden categories of 0-75 constituted 21.21% out of 132 total poor and 40.62% out of 64 total non-poor households respectively. The result showed that there is no statistically significant difference between the age dependency burdens of the two groups with the t value of 0.831 and p value of 0.407. On average, the mean age dependency burden was found to be higher for poor households than the non-poor ones, i.e. 100.42 and 94.3191 with the standard deviation of 44.54161 and 55.05658 respectively. This might be due to lower contribution of labor than the other assets on the poverty status of households (Table 10).

Table 10: Household’s age dependency burden and poverty.

The response of key informants and focus group discussants was consistent with the survey result and they justify this as due to the participation of youngsters and elders in farming activities like looking after cattle, sheep and goats; harvesting crops; and undertaking other domestic activities like cooking wot (curry). They also added that the influence of age of household members on the poverty status of the household is little because most of households in the area have adequate labor needed to conduct activities on their resource.

Households’ productive asset ownership characteristics and poverty

Households’ asset possession affects the farmers’ economic performance which in turn affects the poverty status of households. Farmers' productive assets ownership affected not only the quantity and quality of food available but also other non-food consumption. It is likely that this was because the levels of agricultural outputs were determined by the adequate and timely availability of productive resources such as land and livestock.

Households’ landholding size and poverty: The land holding size of households was analyzed. About 4.59% and 5.61% were found to have landless and 1.51 hectares-2.00 hectares of land with the minimum, maximum, mean and standard deviation of 0.00, 2.00, 0.43169 and 0.5995 respectively. This indicates that about 95.41% of households had their own land although more than four-fifth (84.70%) of the households were having one and less hectares of land. About 27 households (13.78%) were found not cultivating for different reasons, which include shortage of plowing oxen, other agricultural inputs like seed, and labor accounting for 100%, 7.41% and 14.82% respectively (by allowing multiple responses). Accordingly 77.78% of non-cultivators (by allowing multiple responses) preferred and decided to share-cropping. Not much less than half (46.94%) of the farming households cultivates half a hectare and less. Thus, the study examined land holding size of the sample households in the three peasant associations on the basis of agro-climate, but the result shows that there is no significant difference among these three PAs with the chi value of 4.305 and p value of 0.829.

As it is evidenced from many empirical research findings, access to land is one of the critical factors determining poverty. The participants in the focus group discussion also indicated that land size was the most important factors of agricultural production and wealth disparities between households. Based on the disaggregation, about 6.82%, 51.51% of the total poor households were found to be categorized in landless, below 0.50 hectares of land with the mean and standard deviation 0.4678 and 0.32310 respectively. On the other way about 26.56% and 14.06% of the 64 total non-poor households were found to be categorized 1.01-1.50 and 1.51 hectares-2.00 hectares of land with the mean and standard deviation of 0.9375 and 0.48795 respectively. The study also demonstrates that 6.82% of the total sample poor households did not have farmland while none of the non-poor households were landless. However, the land holding size of the majority of the sample households is severely diminished due to population density. This shows that the percentage of poor becomes lower and lower than the corresponding for the non-poor households when the size of farm land increases. To test whether households with smaller farmland are more vulnerable to poverty or not t-test was run and the result indicates that the average farmland holding for poor was found to be 0.4223 ha with the standard deviation of 0.27824 and the corresponding for non-poor households is 0.9648 ha with the standard deviation of.46314. This difference in farm size between poor and non-poor households is found to be statistically significant with the value of t-10.200 and p-value of 0.000 (Table 11).

Discussion

The age of household head (due to accumulated farming experience) provides a household with wider chance of escaping poverty is not significant. The result is found to be consistent with the result of the focus group discussion and key informants as well as the findings of Bigste, et al., and MoFED but inconsistent with the findings of Ayalneh, et al., Bashaasha, Ayalneh and Korf and Sepahvand. Female headed households are more likely trapped in poverty due to their deprivation to many important productive resources. The result of logistic regression revealed that the coefficient for headship of the household was found to be positive and significant at 0.01 significance level and is a determinant of rural poverty. The odds ratio shows that female headed households are 30.471 times prone to poverty than their counterpart of male headed households. The result is found consistent with the findings of Bigsten, et al., and Bigsten, et al., and inconsistent with the findings of MoFED, Ayalneh Maru and Metalign.

Household size was found to be a significant determinant of rural poverty. Family size was found to be significant at 0.01 level of significance and has a negative logit coefficient and odds ratio of 0.232 implying that the probability of a household to be poor increases by a factor of 0.232 as family size increases by a unit. The result was consistent with the findings of Ayalneh and Korf, Metalign, Maru, MoFED and Getaneh, but not with Ayalneh and Korf Dawit, et al., and Sepahvand. The model shows that education is not significant at 0.01 confidence interval. The result was consistent with the findings of Dawit, et al., and Maru but inconsistent with the result of Ayalneh, Dercon MoFED, Tassew.

Land holding size is found to be negative and significant impact on the probability of being poor at 0.01 level is significance. The household with the smaller the land holding is the greater the chance of falling into poverty. The probability of households from being poor reduced by a factor of 51.896 as land holding increases by one hectare. The descriptive statistics also revealed that the poor have either no land or have little. The contribution of landholding size was consistently and unanimously confirmed by the focus group discussion participants. The result was also consistent with the findings of Dawit, et al., Sepahvand, Metalign and Dercon, but not with Bigsten, et al., Bgsten, et al. and MoFED.

Oxen holding size found to be significant at 0.01 levels of significance in determining the poverty status of the households. The coefficient was found to be negative and the odds ratio was 242.306 which imply that a unit increase in oxen holding size decreases the probability of households from being poor by a factor of 242.306. The significant role of owning oxen in escaping poverty was consistently confirmed by the focus group discussions and key informants. The result was also found to be consistent with the findings of Bigsten, et al., Bigsten, et al., MoFED and Metalig. Regarding with livestock (other than ox, sheep and goats) holding size, it was expected that those households with more of such livestock as cattle and equines have better opportunity of smoothing their income overtime and hence of escaping out poverty. The coefficient for the variable was found to be negative and significant at 0.01 level with the odds ratio of 1.910 which indicates that owning these livestock was found to reduce poverty by a factor of 1.910. The key informants and focus group discussion participants also assured the model result. The result was consistent with the findings of Dercon and Krishnan, Dercon, Maru Metalign and Dawit, et al.

The coefficient of small ruminant holding size was found to be negative and significant at 0.01 level of significance. The odds ratio also indicates that a unit increase in the number of small ruminants decreases the probability of households to be poor by a factor of 1.910. The result also shows that its role beehive ownership was not significant in reduce poor and diversify their livelihood, generate extra income. The result was consistent with the finding of Metalign, but not consistent with the result from the key informants and focus group discussion participants as well as the finding of Maru. The use of modern agricultural inputs like fertilizer, high yielding varieties, pesticides, insecticides and herbicides is among others vital contributors to boost crop production. The coefficient for applying modern agricultural inputs was found to be positive and insignificant. The reason might be related to the fact that farmers who utilize modern agricultural inputs get indebted to the credit giving organizations and individuals because they get either the inputs or the money to buy such inputs in credit and they have to pay back the loan right after harvest when prices for crops fail. This situation forces them to sale large amount of their produce for cheaper prices and remains with small amounts of their produce, which makes them to be poor.

The coefficient for household’s involvement in non/off-farm activities was found to be positive although not statistically significant. This indicates that engagement of households in non/off-farm activities has the role of pulling, instead of pushing out, households in poverty. The reason given by the key informants and focus group discussion participants for this is that non/off-farm activities were not be practiced as a way of accumulating more wealth, which are heavily constrained by lack of skill and working capital, as a coping mechanism. As a result, households tend to concentrate on low return activities that have low risk. The result was found to be consistent with the findings of MoFED and Bigsten, et al., but not with Getaneh and Maru, Ayalneh and Korf and Dawit, et al.

The model revealed that the coefficient for credit access was found to be positive although not significant. This was also affirmed by the focus group discussion participants and justified this as due to diversion of loan for unintended purpose and high interest rate. The coefficient of saving habit is negative although not statistically significant. The insignificant relation could be due to the small number of the households who practice saving. Accordingly, the focus group discussion participants affirmed that the poor do not have the habit of saving whereas the more well off or non-poor households practiced, but their number is very small. The result was consistent with the finding of Maru, but not with Metalign. In general, family size, sex, land holding size, oxen holding size, livestock holding size and small ruminant holding size found to significantly correlate with poverty and emerged as the determinants of poverty in the study area.

Conclusion

The study analyzed the data obtained through in-depth interview and FGD in order to confirm whether there is divergence and consistency between quantitative techniques of identifying the determinants of poverty at household level with the qualitative responses of the community members. As a result, 11 (out of the total 14) hypothesized variables in the study were consistently confirmed by the binary logistic model, indepth interview and focus group discussion.

Availability of Data and Materials

The data analyzed in the study is available from the corresponding author on reasonable request.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Ethical Approval

This is an observational study. The university research ethics committee has confirmed that no ethical approval is required.

Author Contributions

All authors contributed to the study conception and design.

Consent to Participate

Informed consent was obtained from all individual participants included in the study.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Conflict of Interest

There are no conflicts of interest.

Funding

This research did not receive any specific grant from funding agencies.

Acknowledgement

We would not forget the tireless guidance, assistance and support of Dr. Kibemo Detamo (PhD) who generously devoted his time and tirelessly marked the manuscript.

References

1. Enquobahrie A. Understanding poverty: The Ethiopian context. The Gambia AAPAM roundtable conference, Addis Ababa, Ethiopia, 2014.

   [Google Scholar]

2. Bogale A, Korf B. Analysis of poverty and its covariates among smallholder farmers in the Eastern Hararghe highlands of Ethiopia. International association of agricultural economists, Beijing, China, 2009;51469.

   [Google Scholar]

3. Bogale A, Hagedorn K, Korf B. Why does poverty persist in rural Ethiopia? Food security and poverty, Durban, South Africa, 2003;25857.

   [Google Scholar]

4. Bogale A, Hagedorn K, Korf B. Determinants of poverty in rural Ethiopia. Q J Int Agric. 2005;44(2):101-120.

   [Google Scholar]

5. Begna B, Paul I. An assessment worded, poverty, the case of selected kebeles of Shashemone Worded, Ethiopia. J sustain Dev Afr. 2010;12(4):125-143.
6. Bellu LG, Liberati P. Impacts of policies on poverty. Absolute poverty lines, 2005;1-22.

   [Google Scholar]

7. Bevan P, Pankhurst A. A Sociological perspective on the causes of economic poverty and inequality in Ethiopia, 2001..

   [Google Scholar]

8. Bidani B, Datt G, Lanjouw JO, Lanjouw P. Specifying poverty lines: How and why. In Asia and Pacific forum on poverty: Reforming policies and institutions for poverty reduction, Asian development bank, Manila 2001;5-9.

   [Google Scholar]

9. Bekele A. Determinants of household rural poverty in Tenta Woreda, South Wollo Zone, Ethiopia, 2022.

   [Google Scholar]

10. Bigsten A, Kebede B, Abebe A, Mekonnen T. Poverty and welfare in Ethiopia: Profile and determinants. Ethiop J Econ. 2002;8(1).

    [Crossref] [Google Scholar]

11. Bigsten A, Abebe S, Mekonnen T. Changes in welfare and poverty: An application of stochastic dominance criteria. Ethiop J Econ. 2002;8(1):21-36.

    [Crossref] [Google Scholar]

12. Bigsten A, Kebede B, Shimeles A, Taddesse M. Growth and poverty reduction in Ethiopia: Evidence from household panel surveys. World Dev. 2003;31(1):87-106.

    [Crossref] [Google Scholar]

13. Boccanfuso D. A conceptual framework for approaches to poverty. Overview paper workshop, Dakar-Senegal, 2004.
14. Coudouel A, Hentschel JS, Wodon QT. Poverty measurement and analysis. A sourcebook for poverty reduction strategies. 2002;1:27-74.

    [Google Scholar]

15. Dawit A, Woldeamlak B, Gete Z, Yemisrach A, Trutmana P. Extent and determinates of household poverty in rural Ethiopia: A study of six villages. EASSRR. 2011;27(2): 21-49.

    [Google Scholar]

16. Berisso O. Determinants of consumption expenditure and poverty dynamics in urban Ethiopia: Evidence from panel data. Poverty and well-being in East Africa: A multi-faceted economic approach. 2016;139-64.

    [Crossref] [Google Scholar]