Women’s Participation in Income Generating Activities and its Effect on Household Dietary Diversity in Wolaita Zone, Southern Ethiopia

doi:10.21203/rs.3.rs-2487222/v1

Download PDF

Research Article

Women’s Participation in Income Generating Activities and its Effect on Household Dietary Diversity in Wolaita Zone, Southern Ethiopia

https://doi.org/10.21203/rs.3.rs-2487222/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Despite the existence of evidences on women’s income generating activities (IGAs) as a key way to promote their socioeconomic empowerment and gender equality that in turn has linkage to food and nutrition security, women in rural Ethiopia yet face exclusion based on the socio cultural norms of patriarchy that deprives their access and better participation in various economic activities. Therefore, we focused on factors affecting women’s participation in IGAs and its effect on household dietary diversity (HDD). Data for the study were collected from 356 women and their household dietary diversity score was constructed using 24-hour recall period. Mixed method approach, the descriptive and inferential statistics for the quantitative data and thematic content analysis for the qualitative data were used for analysis. Thus, from the totalsurveyed women, 38.4 % was engaged in their own IGAs (less participation). The result of descriptive and inferential statistics and qualitative analyses suggest that formal education, business training, business experience, initial capital, cooperative membership, cultivable land and livestock sizes positively affect women participation in IGAs whereas, distance from nearest market negatively affects their participation. Further analyses indicate that women participation in IGAs plays a significant role on increasing HDD.

Women

Income generating activities

Household dietary diversity

Mixed method approach

Wolaita

Ethiopia

Women and development issue have been the world’s agenda since the United Nations organized the first women’s conference in Mexico in 1975. Also the scholar Amartya Sen quoted that ‘Empowering women is a key to building a future we want’. Besides, within the scope of the Sustainable Development Goals (SDGs), gender equality (SDG5) is singled out as one of the 17 SDGs, and this goal is set to achieve gender equality by 2030. Hence, focusing on women’s income generating activities contributes to achieve this objective and can promote the overall development (Boserup et al, 2013).

Over the past four decades, women’s income generating activities (IGAs) and their entrepreneurship development has gained popularity in Africa with a growing number of females to start and run their own business (Vouf, 2017; Endalew, 2020). As stated by this source, now a day’s feminist studies in the academic arena and has led to a desire to learn more about the role of women in this 21st century is not limited to domestic activities or merely housewives. Thus, giving room for women to show up their talents as entrepreneurs, as farm and non-farm laborers, in family businesses and as self-employed to generate income can bring positive change starting from the family levels to the community development (Haile and Tekletsadik, 2015).

In Ethiopia, women are an imperative part of the workforce and development pretended by them cannot be isolated from the economic perspectives (FMSDA, 2016). Particularly, in rural Ethiopia women have a trend of performing various on-farm and off/non-farm income generating activities (Yusuf, 2015, Gebru et al., 2019). Efforts have been also made by government and private organizations to empower women through accessing credit service, establishing women’s associations and self-help groups, and participating political positions (FAO, 2012). Despite these facts, the economic activities in which women engaged in and their equal benefit from economic growth have been challenged by different factors like the restriction of work outside the home, poor women education and skills, multiple responsibilities, honor associated with the women’s sexuality, underdeveloped enterprise culture, an inadequate support system for businesses and poor infrastructures (Siyum, 2015; Desta, 2018; Salma et al., 2020).

Household dietary diversity (HDD) is also the part of this study, which is an important indicator for food and nutrition security and help to achieve SDG 2 (“zero hunger”) and SDG 3 (“good health and well-Being”). However, poor dietary diversity that has direct relation to under nutrition has been the major problem in Ethiopia (Kalkuhl et al., 2013; Pieters et al., 2013). Though there is growing national and global commitment to reduce this problem in the country, a significant proportion of the population suffer from macro and micro-nutrient deficiencies (hidden hunger) and undernourishment (FAO et al., 2014); which mainly affects the rural households where there is limited source of income (FAO et al., 2019). Furthermore, in southern Ethiopia, most of the low-income households consume monotonous diets which are of low quality, cereal based, and lacking in vegetables, fruit, and animal-source foods, thereby increasing the risk of micro-nutrient deficiencies, child stunting and underweight (FAO, IFAD & WFP, 2014).

The position of women within their households determines the household dietary diversity, food and nutrition security (Lepine and Strobl, 2013; Imai et al., 2014). Because, women can play an important role in food production and consumption decision of their family if they have their own source of income (Schutter, 2013; De Brauw 2015). It is also well demonstrated that when women have their own source of income, they favor to control over assets and decision-making power, and they prefer to expend their income to family food than men (Schutter, 2016). Thus, a diversified diet is highly linked to the economic ability of a household to access a variety of foods by obtaining a number of different food groups consumed during a determined period (FAO, 2013).

The previous studies in different parts of the Ethiopia (Arethun, 2013, Abdi, 2014; Derejaw and Sundara, 2016; Fami et al., 2020) stated that the rural household’s participation in IGAs was determined by the interaction of multiple socioeconomic factors like their educational status, business training, credit utilization, market access, ownership of the productive resources like land and the others. These studies have two gaps from the current researchers’ point of view. First, we argue that, these studies did not give particular attention to the particular gender (women) rather than households’ IGAs. Thus, merging both genders in such type of studies may result in difficulty to understand the gendered effect of socioeconomic, institutional and environmental factors on women’s participation in IGAs. Because, women in rural Ethiopia still need particular focus on research and academic arena; since, there has been trend of women exclusion based on the sociocultural norms of patriarchy that deprived their access to development and high involvement in economic activities. Furthermore, the current study analyzes the linkage between women’s IGAs and household dietary diversity.

Moreover, though there are empirical evidences on relationship between women empowerment and household dietary diversity (Schmidt 2012; Lepine, 2013; Sraboni et al. 2014; Feiruz and Fanaye, 2015; Thomas, 2020), there are still less frequently peer-reviewed literature from Ethiopia on this particular theme (Bayeh, 2016). and it hardly exists when we come specifically to the study area. Also methodologically, the above studies used instrumental variable (IV) approach to estimate the linkage and found that ordinary least squares (OLS) and IV models led to similar estimates, except that OLS result was underestimated. We used these earlier methods to argue the selection bias problem as a gap, and we Heckman two step approaches to solve the above econometric problem that is lead to robust inference on outcome variable.

More specifically, based on report data of Wolaita Zone Women and Children Affairs Office Trade and Industry Development office (WZWCAO and WZTIDO 2021/22), though there has been some improvements in women’s socioeconomic status in the area, they still suffers from the traditional belief of that, they consider women as merely housewives and men are the only responsible person to earn household income and fulfill their food requirement. Consequently, in most households only husbands struggle to fulfill the income and food requirements of their family. Therefore, this study adds knowledge and empirical evidence to the existing literature by identifying the major factors that affect women’s involvement in IGAs and examining its causal relation to HDD.

Conceptual Framework

Literatures argue that women’s participation in different income generating activities is the function of demographic, socio economic, institutional and physical factors. Because, the ownership and accessibility of these factors enables them to have power to say and decide especially in resource allocation in country like Ethiopia, where there has been trend of patriarchal power relationship between male and female (Alsop et al., 2005; Alibhai et al, 2015). Moreover, income controlled by women particularly in Africa is more likely to be spent to produce and purchase household food for their family (UNICEF, 1990, WHO, 2015; Malapit et al., 2015; Zereyesus, 2017; Sariyev et al., 2020). Thus, based on concepts raised above, the conceptual framework for current study was developed as shown in Fig. 1.

Study area

Wolaita zone is one of the 13 administrative zones in Southern Nations, Nationalities and Peoples' Region (SNNPR) of Ethiopia and geographically located between 7° 00' North latitude and 37° 45'East longitude (Fig. 2.). Wolaita Sodo is the administrative center of the zone, which is located at 312 km SW to the Addis Ababa, the capital city of Ethiopia. The zone has total area of 4512 km² and consist 12 districts (locally termed as woredas). An estimated population of the zone is 5,385,782, of whom 2,698,261 and 2,687,021 are females and males respectively (CSA projection, 2020). The average annual temperature and rainfall of the zone is 21.86°C and 1350 mm/year respectively. Livestock and crop production with some off/non -farm activities are an integral part of the livelihood system and source of income in the area. However, majority of the women particularly, resource poor and household headed women in the area are poorly engaged in different IGAs. Moreover, the area is characterized by food and nutrition insecurity and plenty of households, specifically and women and children suffer from consuming inadequate dietary diversity (Getu and Mihretu, 2017).

Research design and sampling technique

The study used cross-sectional research design consisting both qualitative and quantitative data of 2021/22. This design is chosen by considering research objectives and time horizon (Robson, 2002 in Saunders et al., 2007)

The study used multistage (four stage) sampling techniques to select the final respondents. First, Wolaita zone was selected purposely based on the researchers’ previous exposure that has created an opportunity to be aware of the existing problem on this study theme. Then, three districts were selected randomly from different agro ecologies, and from each district two kebeles (kebele is the least administrative unit in Ethiopia) were randomly selected. In the third stage, sampling frame (women headed households and women under men headed households) was obtained from respective kebele offices and categorized under two stratum (women who perform and manage their own IGAs and women who do not (housewives)). Finally, systematic random sampling technique which can be taken as an improvement over a simple random sampling (Kothari, 2004) was applied to select the sample respondents from each stratum.

The sample size for this study was determined by using Kothari formula in Eq. 1 below.

$\text{n} = \frac{{\text{z}}^{2}\text{p}\text{q}\text{N}}{{\text{e}}^{2}\left(\text{N}-1\right)+ {\text{z}}^{2}\text{p}\text{q}} = \frac{{\left(1.96\right)}^{2}\left(0.5\right)\left(0.5\right)\left(4747\right)}{{\left(0.05\right)}^{2}\left(4747-1\right)+ {\left(1.96\right)}^{2}\left(0.5\right)\left(0.5\right)}\approx 356$ 1

Where; n is the sample size; Z is the standard cumulative distribution that corresponds to the level of confidence with the value of 1.96; e is desired level of precision; p is the estimated proportion of an attribute present in the population with the value of 0.5 as suggested by Kothari (2004)to get the desired minimum sample size of households at 95% confidence level and ± 5% precision; q = 1-p; and N is the size of the total population from which the sample is drawn. Then, n is redistributed to each kebele via probability proportion to size ($\text{n}\text{i} = \frac{Ni*n}{\text{N}}$) (Table 1). The study also applied the same sampling procedure of selecting study zone, sample districts and kebeles to reach the final focus group discussants and key informants. Then, finally a total of 3 FGDs in three districts, each consisting 12 discussants were held. The composition of FGDs included 2 kebele leaders, 2 elder women and 2 development agents who were purposely selected from respective kebeles and four women who were randomly selected from two heterogeneous groups (2 from IGAs participants and 2 from non-participants). Also by following the same sampling procedure of selecting the study zone and sample districts, the final

Table 1

Sample households taken from sample kebeles
Target districts	Sample kebeles	Sampling frames	Samples taken
DamotWoyde	Galcha Sake	804	60
DamotWoyde	Adecha	1012	76
Abala Abaya	AbalaLongena	740	55
Abala Abaya	AbalaKolshebo	435	33
Damot Gale	Woshi Gale	903	68
Damot Gale	Wandara Gale	853	64
		4747	356
Source: Computed based on data from Kebeles.

Participants of KIIs were purposely selected. These include 2 experts from each district’s Women and Children Affairs Office, 2 from each district’s Trade and Industry Development Office and 2 from each district’s Agriculture Office.

Data sources, types and collection methods

The study used both qualitative and quantitative data collected from primary sources (respondents) and secondary sources (reports of woreda and zonal governments, statistical information, published theses and dissertations and journal articles). The data collection tools used in the study includes semi-structured questionnaire (for household survey) which was administered by well-trained enumerators (2 from each kebele) and the researchers. The qualitative data of the study were collected using FGDs and KIIs. Here, open ended questions were prepared specifically for in-depth understanding of the qualitative responses that supplement and triangulate quantitative data. The collected data had detailed information on household demographic and socioeconomic characteristics, institutional information, households’ assets, location, types of IGA undertaken by women, reasons for non-involvement of women in IGAs and household dietary diversity data.

Household dietary diversity data

The household dietary diversity is measured by computing the Household Dietary Diversity Score (HDDS), increasingly used as an approximate measure of energy consumption per capita of the household and proxy for nutrient adequacy (Jones et al., 2014). The assessment of the HDDS could be during the period of food shortage, such as immediately prior to the harvest or immediately after emergencies or natural disasters (Kennedy et al., 2011). The twelve food groups included in HDD questions are: (1) Cereals (2) Root and tubers (3) Vegetables (4) Fruits (5) Meat, Poultry, (6) Egg, (7) Fish and other sea food, (8) Pulses/legumes/nuts (9) Milk and milk products (10) Oil/fats (12) Spices, condiments and beverages. The questions were applied for two groups of women; 137 IGAs participants and 219 non-participants. Here, the responsible for food preparation was asked the question, or if that person is unavailable, another adult person who consumes food in the recall period was asked. The interviewer first determined whether the previous 24 hour is "usual" or "normal" for the household. If it was a special occasion, such as a funeral or feast another day would be chose.

Data analysis

The descriptive statistics include frequency, percentage, mean and standard deviation were used to describe and summarize the demographic and socioeconomic characteristics of the respondents. And the inferential statistics, T-test was applied to compare the mean difference of continuous explanatory variables and household dietary diversity between the two groups (IGAs participant and non-participant group), while χ²-test were applied to compare the proportion difference of dummy and categorical explanatory variables between two groups.

The econometrics approach used to analyze women’s participation in IGAs and its effect on HDD was Heckman two steps, established by Heckman (1979). Though there is another competing model like PSM to analyze these objectives, it still assumes that selection is based on observable factors and can lead to biased result when there are non-observable factors that influence both women’s participation in IGAs and HDD. Hence, to circumvent this potential pitfall, we used the Heckman two-step approach which corrects the problem of selection bias and simultaneity (Marchenko and Genton, 2012). Thus, women’s participation in IGAs and its effect on HDD involves two steps. The first step has to do with the probability of participation by using the Probit estimation. In this step, the decision to participate in the IGAs is not randomly assigned or it’s based on self-selection whether to participate in IGA or not. In other words, participant women may be systematically different from non-participants, and these differences may obscure the true effect of women’s IGAs on HDDs. The first step (selection equation) is given as:

$$Fi\text{*} = \beta 0 + \beta 1Xi+\epsilon i$$

Where F* is an unobserved latent variable representing woman IGA participation decision, Xi is a vector of explanatory variables, β is a vector of parameters to be estimated, and εi is an error term distributed with mean 0 and variance 1. The observed dummy variable can be expressed as:

F_i= 1 if F_i* > 0 (for IGA participants) (3)

F_i= 0 if F_i* ≤ 0 (for non IGA participants) (4)

The second stage (i.e participation effect on HDD, measured by household dietary diversity score (HDDS)) was estimated by using Ordinary Least Square (OLS). For this stage a substantive equation is given as:

$$Yi=\alpha 0 +\alpha 1Zi+\mu i$$

It should be noted that for the correction for self-selection biases in the substantive equation (5), an Inverse Mills Ratio (IMR) represented by the symbol λ as an extra explanatory variable is added. The IMR computed from first step probit regression provides a corrected selection of OLS estimates (Greene, 2003). Adding IMR translates Eq. (2) into Eq. (6) as:

$$Yi=+ \alpha 0 + \alpha iXi + \delta i\lambda i + \mu i$$

Where; δi is the coefficient of the IMR (λi). If lambda (λ) is statistically significant, sample selection bias is a problem and, therefore, Heckman’s two-stage sample selection model is appropriate for the estimation (Marchenko and Genton, 2012). The formulation process of IMR is given by:

${\lambda }\text{i} = \frac{\varphi \left(Xi\alpha \right)}{\varphi \left(Xi\alpha \right)}$ (7) Where; φ and ϕ are normal probability density function and cumulative density function, respectively of the standard normal distribution, and ϕ≡ (ωiχ). µi is a two-sided error term with N(0,σ2 v). In general, the model computes the inverse mills ratio from the Probit regression and uses it as additional explanatory variable to explain the outcome variable (Heckman, 1976).

The qualitative data of the study was recorded and transcribed and afterwards analyzed through thematic content analysis which emphasizes on identifying, analyzing and interpreting patterns of meaning or themes argued by FGDs and KIIs (Alhojailan, 2012).

3.6. Description of variables

The explanatory variables in Table 2 were included in the study to explain the dependent and outcome variables based on its theoretical and empirical background and the study area contexts.

Table 2

Description of variables, measurements and expected signs
No	Variables	Variable type	Measurements
Selection and outcome variables
1	Women’s participation in IGAs	Dummy	1 participated, 0 otherwise
2	Household dietary diversity	Continuous	HDDS
Explanatory variables
1	Age	Continuous	Year	-&-
2	Education status	Continuous	Year of schooling	+ & +
3	Marital status	Categorical	1 married, 2 divorced, 3 widowed	+ & +
4	Labor size	Continuous	AE	+&+
5	Business experience	Continuous	Years	+ &+
6	Initial capital	Continuous	Birr	+ &+
7	Livestock in TLU	Continuous	TLU	+& +
8	Cultivable land size	Continuous	Hectare	+ & +
9	Remittance	Dummy	Received 1, 0 otherwise	+ & +
10	Credit use	Dummy	1 used,0 otherwise	+ &+
11	Extension contact	continuous	Frequency	+ & +
12	Business training	Continuous	1 attended, 0 otherwise	+& +
13	Cooperative membership	Dummy	1 member, 0 not member	+ & +
14	Distance to the nearest market	Continuous	Hour	-&-
15	Distance to the main road	Continuous	Hour	-&-
16	Irrigation water availability	Dummy	1available, 0 otherwise	+ &+

Statistical summary of explanatory variables in relation to women’s participation in IGAs

The descriptive output the study depict that, out of the total surveyed women households, the proportion of participants in IGAs were 137 (38.4%) and non-participants in IGAs were 219 (61.6%); meaning that in the study area, the existing status of women’s involvement in different income portfolio was low, and it needs to investigate on the problem. During the survey time, women were also requested to inform the types of IGAs in which they were engaged. Accordingly, there were various IGAs in which women undertake and manage in the area. The major one include: selling livestock products (butter, yoghurt and cheese (23.5%), rearing poultry, cattle and small ruminants (23.5% ), retailing grains and root crops (21.3%), home gardening (18. 8%) and selling local beer (Areke, Tella and Bushbush) (12.9%).

The statistical summary of continuous variables in Table 3 below indicates the following results: The mean year of formal education for participant women was significantly higher (2.18 years) than that of non-participants’ (1.53 years). This result depicts that, without indicating any causal effect, women with the better education status are more likely to participate in IGAs as compared to women with less education status. With respect to business experience, the mean business experience in years for participant women was significantly longer (3.21 years) than that of non-participants’ (0.84 years). This finding indicates that, women who previously experienced more time in business had better chance of involving in IGAs as compared to women who had less experience in business. The mean initial capital for participant women was significantly higher (3549.41 Birr) than the non-participants’ (1736.86 Birr); implies that, owning a more initial capital initiated women to involve in IGAs as compared to women who own less initial capital. The mean value of cultivable land for participant women was significantly higher (1.88 ha) than non-participants’ (0.76 ha), and also the size for IGAs participant women is significantly higher (2.15 TLU) than the non-participants’ (1.24 TLU). This result implies that women who have more cultivable land size and livestock were highly initiated to involve in agriculture related IGAs than those women who have less size of cultivable land and livestock in their household.

Table 3

Respondents’ characteristics (continuous variables)
Variables	Participants (137)		Non-participants (219)		t-values
Variables	Mean	SD	Mean	SD
Age	44.90	6.07	46.06	9.01	-0.608
Education years	2.18	1.40	1.53	0.89	1.343**
Business experience	3.21	1.94	0.84	0.65	3.751***
Household labor size	3.25	1.25	2.79	1.07	-0.059
Initial capital	3549.41	2103.12	1736.86	1416.66	0.842**
Cultivable land size	1.88	0.82	0.76	0.61	2.297**
Total livestock size in TLU	2.15	1.02	1.24	1.09	1.361**
Distance from main paved road	0.76	0.49	1.48	0.73	-2.656**
Distance from closest market	1.60	0.75	2.54	1.01	-0.863**
Frequency of extension contact	1.55	1.03	0.85	0.71	3.159***
“”, “” and “**” represent statistical significance at 10, 5 and 1% levels respectively.

The participant women walk less time to arrive the main paved road (0.76 hour) than the non-participants (1.48 hour) and the mean value of this variable was significantly different between the two categories. Moreover, the distance from closest market for participant women takes less time (1.60 hour) than the non-participants (2.54 hour) and its mean difference is statistically significant. This indicates that, without causal relationship market and main road distance matters the participation of women in the IGAs. In terms of the frequency of extension contact, the mean value in frequency of contacting extension agents was significantly higher (1.55) than the non-participants (0.85) per month. This depicts that, the more frequent visit of women to the development agents in the area enabled them to participate in the IGAs than women who had less frequent visit.

The statistical summary of discrete variables in Table 4 below indicate the following results: The proportion of credit users in IGAs participant category was significantly higher (31.4%) as compared the proportion of credit users in non-participants (23.5%). In terms business training, majority of participant women received business training (67.1%) than the non-participants (46.7%) and the proportion difference was statistically significant. Moreover, the percent of cooperative members in the participant category was significantly higher (52.9%) than the percent of cooperative members in non-participant category (32.1%); meaning that, without any causal relationship, being a member of cooperative increased the participation of women in the IGAs as compared to non-members.

Table 4

Respondents’ characteristics (discrete variables)
Variables	Categories	Participants (137)	Non-participants (219)	χ2
Variables		Percent	Percent
Marital status	married	88.2	58.4	0.304
	widowed	9.4	30.7
	divorced	2.4	10.9
Credit	not used	68.6	76.5	0.315*
Credit	used	31.4	23.5	0.315*
Remittance	not received	65.4	69.3	0.036
Remittance	received	34.6	30.7	0.036
Business training	not trained	32.9	53.3	31.794***
Business training	trained	67.1	46.7	31.794***
Cooperative membership	not member	47.1	67.9	28.002***
Cooperative membership	member	52.9	32.1	28.002***
Irrigation water availability	not available	52.0	73.7	0.032
Irrigation water availability	available	48.0	26.3	0.032
“”, “” and “**” represent statistical significance at 10, 5 and 1% levels respectively.

Comparing HDD between IGAs participant and non-participant women

The descriptive result in Table 5 presents that, the dietary diversity score of participant women was significantly different (5.25) than that of the non-participants’ (4.08). This shows the substantial contribution of women’s participation in IGAs on HDD in the area.

Table 5

Comparing household dietary diversity between participant and non-participant women
Household dietary diversity	Participants (137)		Non-participants (219)		t-value
Household dietary diversity	Mean	SD	Mean	SD
Dietary diversity score	5.25	3.89	4.08	2.71	0.534**
“”, “” and “**” represent statistical significance at 10, 5 and 1% levels respectively.

Table 6

Heck man’s first-stage (selection) result
Factors	Coefficient	dy/dx	Standard error	P>\|z\|
Age of the respondent	− .108489	− .0039226	.0608623	0.75
Marital status	1.820352	.0658182	1.168196	0.119
Education level	.562653**	.0203438	.3996895	0.016
Business experience	1.402217***	.0506998	.4872587	0.004
Household labor size	.1003826	.0036295	.2746041	0.715
Initial capital	1.001039***	.0000376	.0003856	0.007
Cultivable land size	.8038291**	.0290639	.5492927	0.014
Livestock size in TLU	.9561667**	.034572	.3934258	0.015
Credit use	.3909624	.0213673	.7743469	0.445
Remittance	2.482814	.0897708	1.098908	0.24
Distance from paved main road	-3.349351	− .121102	1.035992	0.12
Distance from closest market	-1.631502***	− .05899	.4963625	0.001
Frequency of extension contact	1.89134	.0683849	.7516686	0.12
Attending business training	1.097687**	.0396889	.7791052	0.016
Cooperative membership	.8490648**	.0306995	.7710213	0.027
Irrigation water availability	.4014518	.0181309	.9177991	0.585
-cons	1.353979		.808508	0.063
“”, “” and “**” represent statistical significance at 10, 5 and 1% levels respectively.

Determinants of women’s participation in IGAs

The overall goodness of fit for the Heckman-two step selection model was statistically significant at 1% significance level (χ2 =, 93. 23***). Also the coefficient of Inverse Mills Ratio (lamda) was found to be significant at the probability of less than 5% which implies the existence of sample selection bias. The lamda is used as the correction factor to capture sample selectivity bias.

In the first stage (probit) analysis, from the total of 16 explanatory variables included in the regression, 9 variables were found to be significantly affecting women’s participation in IGAs and discussed in the following way ().

Formal education status of women had positive and significant influence on probability of women’s participation in IGAs. The estimation result of marginal effect indicates that, a 1 year increase in women’s education status increases their likelihood of participation in IGAs by 2.03%, ceteris paribus. The possible reason from KII depicts that, women who spent more years in the formal education easily understand their gender equality, aware to use the available opportunities that help to start business and that in turn increases their motivation to engage in different income portfolios. The focus group discussants also argued that educated women are aware of the costs and benefits of participating in a particular business. The result is consistent to empirical evidence of Minot et al. (2006);

Business experience had a positive and significant influence on women participation in IGAs. The result of marginal effect indicates that, as women's business experience increases by 1 year, the probability of their participation in IGAs increase by 5.07%, ceteris paribus. The focus group discussants also argued that women who had prior experience in their own income generation were better skilled on techniques to conduct a particular IGA, and able to differentiate the profitable IGAs and risks they can face. Consistently the study of Halkias (2011) and Selig (2014) found that business experience and participation in the IGAs were positively associated.

Initial capital was one of the important factors that had positive and significant influence on participation of women in the IGAs. For 1 additional Birr increase of initial capital, ceteris paribus, the women would be 0.0037% more likely to involve in IGAs. The focus groups also argued that, the financial source for majority of women in the area were their husbands and hence they hardly have cash on their hand. So, the availability of cash on their hand initiates them to be involved in IGAs. This contradicts the finding from the similar theme by Solomon et al. (2016).

Cultivable land size owned by the household had a positive and significant influence on participation of women in IGAs. Thus, owning 1 hectare additional land, ceteris paribus, increases the probability of women’s participation in IGAs by 2.9%. Moreover, the evidence from KII verified that, availability of cultivable land motivated women to be engaged in home gardening and rearing small ruminants, and also on other hand household headed women allow their land for share farming and renting as income source. The study findings of Feiruz and Fanaye (2015) and Sariyev et al. (2020) were consistent to the current result.

Livestock size in TLU was positively and significantly associated to women’s participation in the IGAs. An increase in 1 TLU increases the likelihood of women’s participation in IGAs by 3.45%. Thus, as mentioned by focus group discussants, it is the responsibilities of women to sell livestock products like milk, butter, cheese and egg in the market. This helped them to get the chance to collect market information about IGAs and as a main source of initial capital to start new IGAs. The present finding contradicts the study finding of Sariyev et al. (2020) which indicated that increased number of livestock limits the time of women’s involvement market oriented businesses.

Distance to closest market had a negative and significant influence on participation of women in IGAs. For each 1 hour additional journey to the closest market, ceteris paribus; the women would be 5.9% less likely to participate in IGAs. Also, as argued by focus group discussants, the possible reason for this finding was, when women reside at the long distance from the market area, they could not access both input’s and output’s market information; they could not meet their peers who had previous experience on market participation, and also as women have a lot of household related burdens they are not much interested to go a long distance and waste their time. The result is consistent to the study result of Sariyev et al. (2020), reasoned out the negative relationship between market distance and women’s participation in IGAs.

Business training had a positive and significant influence on participation of women in IGAs. With considering of ceteris paribus, participating in business training increases the likelihood of women’s participation in the IGAs by 3.96%. Evidence focus group discussants revealed that in the study area, business oriented training sometimes provided for women by Government bodies and NGOs had narrowed the gaps in their understanding, knowledge and skill towards undertaking different activities, and this in turn increased their motivation towards participation in the IGAs. This confirms the study result of Netsaalem (2011).

Cooperative membership had positive and significant influence on women’s participation in IGAs at 5% significance level. Being a member in cooperative, ceteris paribus, increases the probability of women’s IGA participation by 3.07%. The evidence from key informants depicted that women’s involvement in cooperatives particularly, savings and credit cooperatives and Ekub had improved the business information sharing among the co-members and helped them as source of initial capital to start IGAs. Besides, the cooperative member woman who was one of the focus group discussant informed that, her participation in Ekub was the reason to start the vegetable farming which she has been working in and increased her income.

Dietary diversity effect of women’s participation in the IGAs

In the outcome estimation result, the value of IMR was significant and positive rho, which means that the error term of both the selection and outcome equation were correlated positively. This shows that, the explanatory variables included in the selection model also explained household dietary diversity. Thus, in outcome estimation, five variables were found to be significant and interpreted as follows (Table 7).

Table 7

Heckman’s second stage (outcome) result
Factors	Coefficient	Standard error	P>\|z\|
Age of the respondent	.0069036	.0144093	0.632
Marital status	− .3820604	.2176385	0.79
Education status	.06835003***	.2281369	0.003
Initial capital	.0000948**	.0000233	0.031
Cultivable land size	.3474354***	.1015298	0.001
Livestock size in TLU	.071886**	.0904206	0.042
Credit use	− .1890805	.1674034	0.259
Remittance	.0483659	.0574367	0.400
Distance from main road	− .4308208	.1915082	0.24
Distance from closest market	− .318067**	.1351021	0.019
Frequency of extension contact	.1774456	.1186833	0.135
Irrigation water availability	.3467818	.2230582	0.120
Mills (lambda)	0.6251475 **	.2457084	0.011
-cons	1.344154***	.7620155	0.000 4
Number of observation = 356, Censored observation = 219, Uncensored observation = 137, Wald chi²(16) = 93.23, Prob > chi² = 0.000, Rho = 0.86340 and Sigma = 0.72405048
“”, “” and “**” represent statistical significance at 10, 5 and 1% levels respectively.
Source: Survey, 2021.

As prior expectation, formal education status had a significant and positive effect on household dietary diversity at 5% significance level. A 1 grade increase in formal education of women, ceteris paribus, the score of dietary diversity (DDS) increases by 6.83%. The FGD also supplemented this result, that as women who had better education status in the area are more able in sharing the allocation of resources and household decision thereby reducing the dominance of male their husband. Besides, they have understanding on importance of consuming variety of food as compared women who are not educated. This finding supports the notion that educating a female member is educating the household (Fanzo et al., 2013; Fanzo, 2017).

Women’s initial capital to start business had positive and significant effect on household dietary diversity at 1% significance level. Although the effect was very small, one additional Birr on their initial, ceteris paribus, could increase the score of dietary diversity by 0.0095%. The result was supported by the evidence from FGD, that women who had more initial capital are better initiated, think for the business opportunity and start some IGAs, thereby increase their own income. This in turn enables them to access variety of food particularly, fruit and vegetable food groups from market.

Cultivable land size had significant and positive effect on household dietary diversity at 1% significance level. A 1 hectare increase in cultivable land size, ceteris paribus, would increase the score of dietary diversity by 34%. The possible justification found from FGD and field observation depicted that, in the study area households with large land size are more tended to variety of crops and also they have low sense of fearing for the risks. In particular, women are more tended to participate in home gardens and poultry, small ruminant production, if they own large land size. %. This result contradicts the finding of other authors (Adimasu and Huluka, 2019) and confirms the finding of (Nkegbe et al., 2017).

Livestock size had also positive and significant effect on household dietary diversity at 5% significance level. The coefficient value shows that, owning one additional livestock in TLU, ceteris paribus, the score of dietary diversity increases by 9.04%. Evidence from key informants also verified this result; thus, women who owned more livestock at household level had an opportunity of consuming livestock products like milk, butter, cheese and egg, and also accessing these products. The FGDs also argued the benefit of owning ox for plowing their own land and other’s land for share farming; thereby they can produce variety food. Similarly, Abera et al. (2020) corroborated the positive effect of livestock ownership on HDD.

On other hand, distance from closest market had negative and significant effect on household dietary diversity at 5% significance level. The coefficient value indicates that, as it take 1 more hours for women to arrive the closest market, ceteris paribus, the score of dietary diversity decreases by 31%. The focus group discussants argued that, women who reside near to the market are more accessible to the market information, low transaction costs in terms of time and money, and these enabled them to diversify their income source. Further, women spend more proportion of their income on family food requirement as compared to the income earned by their husbands. This was also reasoned out by the findings of Kassie et al. (2015) and Thomas (2020).

In this study we analyzed the determinants of women’s participation in IGAs and its effect on household dietary diversity. Our analysis verified that, the participation of women in different IGAs was low in the in area. Thus, women’s participation or non-participation in the IGAs was the function of demographic, socioeconomic, institutional and physical factors.

The women’s formal education level, business experience, amount of initial capital, cultivable land size, livestock size, attending in business training and cooperative membership had positive and statistically significant contribution whereas, distance from the closest market had negative and statistically significant contribution to women to be engaged in IGAs like selling livestock products, poultry production, retailing grains and root crops, home gardening and selling local brew.

Furthermore, based on this econometrics evidence, the women’s participation in the IGAs had positive contribution on increasing the availability and accessibility of various food items; and is an indication for improving dietary diversity at household level. As verified by the result, women’s formal education, their initial capital, cultivable land size and livestock size had positive and statistically significant effect whereas, distance from closest market had negative and statistically significant effect on household dietary diversity.

The study result has strong implications for any initiatives and interventions working to enhance the women‘s IGAs, their economic empowerment, and thereby improve its contribution to HDD.

Therefore, starting from the policy makers to the development actors at grass root levels in the country have to strengthen the adult literacy with main of women, their participation in business training, develop and support their initial capital and membership in cooperatives, access market opportunities and improve their land and livestock ownership. The particular focus on these interventions should be undertaken by the local level actors, such as zonal and woreda level Women Affairs, Agriculture, Cooperative, Trade and Industry offices in collaboration with the Aid Organizations working in the study area.

This study also verified the positive effect of women’s IGAs on HDD, which is one of the proxy indicator for the food and nutrition security should. Hence, policy and development actors working to improve dietary diversity, food and nutrition security should not only focus on men farmers but also on women’s empowerment through IGAs.

Funding The authors declare that no funds were received during the preparation of this manuscript.

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

Acknowledgments We would like express our gratitude to the Target Districts’ Women and Children Affair offices and Trade and Industry development Offices. Our gratitude also extends to enumerators and all respondents from zone to kebele levels.

Abera, A., Tesfaye, W., Aklilu, (2020). Determinants of Women Participation in Income Generating Activities in Ethiopia.Research Square. https://orcid.org/0000-0003-0763-3866.
Alhojailan,M.I.(2012).Thematic content analysis:A critical review process and evaluation. West east journal of social science.Vo.1, No.1.
Alibhai, S., Buehren, N. and Papineni, S. (2015) ‘Female entrepreneurs who succeed in male-dominated sectors in Ethiopia.’ Gender Innovation Lab Policy Issue 12. Washington, DC: World Bank.
Alsop, R., Heinsohn, N. and Somma, A. (2005) ‘Measuring empowerment: An analytic framework’, in R. Alsop (ed.) Power, rights and poverty: Concepts and connections. Washington, DC: World Bank.
Bayeh, E. (2016). The role of empowering women and achieving gender equality to the sustainable development of Ethiopia. Pacific Science Review A: Natural Science and Engineering, 2, 37–42. https://doi.org/10.1016/j.psrb.2016.09.013
Boserup, E., Fei, S.T. & Toulmin, C. (2013), “Woman's Role in Economic Development”, Routledge publishers, New York:NY.
CSA (2020). Population projection of Ethiopia for all regions at Woreda level from 2014-2017 . Addis Ababa, Ethiopia.
Derajew, F. and and M. Sundara (2016).Factors Influencing Farmers’ Participation On Off Farm Activities: A Case Study in East Gojjam Zone, Ethiopia.International Journal of Life Sciences. Vol. 5 No.2. Pp. 112-119
Desta, S.(2018). Desk Review of Studies Conducted on Women Entrepreneurs in Ethiopia Addis Ababa Chamber of Commerce and Secretarial Associations.
De Brauw, A. (2015) ‘Gender, control, and crop choice in northern Mozambique’ Agricultural
Economics 46(3): 435-448.
Endalew, T.A. (2020). Determinants that Influence Women Entrepreneurs Performance
in Ethiopia”, Vol. 9, No. 24,pp.1-20.
Fami, A.Lemma, Z.Danie, B.(2020). Determinant of Rural Household Participation in Non-Farm and Level of Generated Household Income from the Activities: The Case of Haramaya Woreda, East Hararghe Zone, Ethiopia.Journal of Economics and Sustainable Development, Vol.11, No.19.
Fanzo, J., Marshall, Q., Wong, J., Merchan, R. I., Jaber, M. I., Souza, A., & Verjee, N. (2013). The integration of nutrition into extension and advisory services: A synthesis of experiences, lessons, and recommendations. Lindau, Switzerland: Global Forum for Rural Advisory Services.
Fanzo, J. C. (2017). Decisive decisions on production compared with market strategies to improve diets in rural Africa. The Journal of Nutrition: Commentary, 147, 1–2. doi:10.3945/jn.116.241703.
FAO (2012). The state of food insecurity in the world: Economic growth is necessary but not sufficient to accelerate reduction of hunger and malnutrition Food and Agriculture Organization of the United Nations, Rome, 2012. 65pp.
FAO (2013). Resilient livelihoods: disaster risk reduction for food and nutrition security framework program. Rome.
FAO, IFAD, & WFP. (2014). The state of food insecurity in the World 2014. Strengthening the
enabling environment for food security and nutrition. Rome: FAO.
FAO, IFAD, UNICEF,& WHO. (2019). The State of Food Security and Nutrition in the World 2019. Safeguarding against economic slowdowns and downturns. Rome, FAO. Licence: CC BY-NC-SA 3.0 IGO.
Feiruz, Y. and Fanaye, T. (2015). Women's Empowerment in Agriculture and Dietary Diversity in Ethiopia. Working Paper, Ethiopian Development Research Institute. Addis Ababa
FMSDA (2016). Micro and small enterprises development annual report, Federal Micro Enterprises Addis Ababa, Ethiopia.
Gebru, K. M., Leung, M., Rammelt, C., & Zoomers, A. (2019). Vegetable Business and Smallholders ’ Food Security : Empirical Findings from Northern Ethiopia. 1–28. https://doi.org/10.3390/su11030743.
Getu Gamo Sagaro, Mihiretu Alemayehu (2017). Dietary Diversity and Associated Factors Among Infants and Young Children in Wolaita Zone, Southern Ethiopia. Science Journal of Clinical Medicine. Vol. 6, No. 4, pp. 53-59. doi: 10.11648/j.sjcm.20170604.12 .
Greene, W. H. (2003). Econometric analysis. Prentice-Hall.
Hagos, F., Makombe, G., Namara, R., & Awulachew, S. B. (2008). Does access to small scale irrigation promote market-oriented production in Ethiopia?International Water Management Institute. (No. 614-2016- 40817).
Haile.and Tekletsadik, M. (2015). Women's Entrepreneurship Development In Ethiopia A
Case Study Of Women in Self Employment (WISE).
Halkias (2011). Challenges facing women entrepreneurs. Management Research Review in Nigeria; 26.
Heckman, J.J. (1979). Sample Selection Bias as a Specification Error. Econometrica 47(1):153-162.
Huluka, Beneberu A., Admassu, T., Wondimagegnhu (2019). Determinants of household dietary diversity in the Yayo biosphere reserve of Ethiopia: An empirical analysis using sustainable livelihood framework, Cogent Food & Agriculture, 5:1, 1690829, DOI: 10.1080/23311932.2019.1690829.
Imai, K. S., S. K. Annim, V. S. Kulkarni, and R. Gaiha (2014) ‘Women’s Empowerment and Prevalence of Stunted and Underweight Children in Rural India’ World Development 62: 88-105.
Jones, A. D., Shrinivas, A., & Bezner-Kerr, R. (2014, June). Farm production diversity is associated with greater household dietary diversity in Malawi: Findings from nationally representative data. Food Policy, 46, 1–12.
Kalkuhl, M., Kornher, L., Kozicka, M., Boulanger, P., and Torero, M. (2013). Long-Term Drivers of Food and Nutrition Security. Food secure working paper no.06. The Hague: LEI Wageningen UR.
Kassie M, Stage J, Teklewold H, Erenstein O (2015) Gendered food security in rural Malawi: why is women’s food security status lower? Food Sec. 7:1299–1320. https://doi.org/10.1007/s12571-015-0517-y
Kennedy G, Ballard T, Dop MC (2011) Guidelines for measuring household and individual dietary diversity. Food and Agriculture Organization of the United Nations, [Rome]
Kennedy, G., A. Berardo, C. Papavero, P. Horjus, T. Ballard, M. Dop, J. Delbaere, and I. D. Brouwer. (2013) ‘Proxy measures of household food consumption for food security assessment and surveillance: comparison of the household dietary diversity and food consumption scores’, Public Health Nutrition 13(12): 2010-2018.
Kothari C. (2004). Research Methodology: Methods and Techniques‟. 2^need. New Delhi:New Age International (P) Ltd.
Lepine, A., and E. Strobl (2013) ‘The Effect of Women’s Bargaining Power on Child Nutrition in Rural Senegal’ World Development 45: 17-30.
Malapit, H. J. L. and A. R. Quisumbing (2015) ‘What dimensions of women’s empowerment
matter for nutrition in Ghana?’, Food Policy 52: 54-63.
Marchenko, Y. V., & Genton, M. G. (2012). A Heckman selection-t model. Journal of the American Statistical Association, 107(497), 304–317. https://doi.org/10. 1080/01621459.2012.656011.
Minot, N., Epprecht, M., Anh, T. and Trung, L. (2006)”Income diversification in the upland of Vietnan”. International Food Policy Research Institute, Washington D.C.
Netsaalem, B. (2011). Factors affecting the performance of micro and small enterprises: The case of Harar, Ethiopia.
Nkegbe, P. K., Abu, B. M., & Issahaku, H. (2017). Food security in the Savannah accelerated development authority zone of Ghana: An ordered probit with household hunger scale approach. Agriculture & Food Security, 6(35), 1–11. doi:10.1186/s40066-017-0111.
Pieters, H., Guariso, A., and Vandeplas, A. (2013). Conceptual framework for the analysis of the determinants of food and nutrition security Food secure Working paper no. 13. The Hague: LEI Wageningen UR.
Salma, J. ,Sanam, H., Arif.,J ., Abid, H. , Muhammad.A. ,Jinsoo, H. ,and Abida, J. (2020). Impacts of Rural Women’s Traditional Economic Activities on Household Economy: Changing Economic Contributions through Empowered Women in Rural Pakistan” Sustainability, Vol.12, No.2731, pp. 2-23.
Sariyev, O., Loos, T.K., Zeller, M. et al. Women in household decision-making and implications for dietary quality in Bhutan. Agric Econ 8, 13 (2020). https://doi.org/10.1186/s40100-020-00158-0.
Saunders,M., Lewis, P. and Thornill,A. (2007). Research Methods for business students.4^th edition,Financial timesprentice Hall, Edinburgh Gate, Harlow
Schutter, O. (2016). Gender equality and food security: women’s empowerment as a tool against hunger. Asian Development Bank, Mandaluyong City Metro Manila Philippines.
Selig G. (2014). Critical success factors for winning entrepreneurs and the role of an incubator. Proceedings of the 2014 Zone 1 Conference of the American Society for Engineering Education-Engineering Education: Industry Involvement and Interdisciplinary Trends. Asee Zone 1 ; 2014.27.
Siyum, (2015). Role of Micro and Small Scale Business Enterprises in Urban Poverty
Alleviation: a case study on cobble stone paving sector in Addis Ababa city.
Solomon, T. Tadele, F., Shiferaw K., Daniel B. (2016). Determinants of growth of micro and small enterprises (MSEs): Empirical evidence from Ethiopia.
Thomas, L., Euan, Ph., Deborah, R. (2020). From Farm to Kitchen: How Gender Affects Production Diversity and the Dietary Intake of Farm Households in Ethiopia Journal of Agricultural Economics, Vol. 72, No. 1, 268–292 doi: 10.1111/1477-9552.12404.
UNICEF (United Nations Children’s Fund). 1990. Strategy for Improved Nutrition of Women and Children in Developing Countries. New York: UNICEF.
Voufo, B. T. (2017). Women Empowerment and Intra-household Dietary Diversity in Nigeria.
Yusuf, H. A., Nuhu1, K. J., Shuaibu1, H., Yusuf, H. O. and Yusuf, O. (2015), “Factors Affecting
the Involvement of Women in IGAs in Sabon-Gari Local Government Area of Kaduna State, Nigeria”, American Journal of Experimental Agriculture,Vol.5, No 1, pp54-59.
Zereyesus, Y. A. (2017) ‘Women’s Empowerment in Agriculture and Household-LevelHealth in Northern Ghana: A Capability Approach’, Journal of International Development 29: 899-918.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Women’s Participation in Income Generating Activities and its Effect on Household Dietary Diversity in Wolaita Zone, Southern Ethiopia

Status:

Version 1

Abstract

Figures

Introduction

Conceptual Framework

Methodology

Study area

Research design and sampling technique

Data sources, types and collection methods

Household dietary diversity data

Data analysis

3.6. Description of variables

Results And Discussion

Statistical summary of explanatory variables in relation to women’s participation in IGAs

Comparing HDD between IGAs participant and non-participant women

Determinants of women’s participation in IGAs

Dietary diversity effect of women’s participation in the IGAs

Conclusion and policy implications

Declarations

References

Additional Declarations

Status:

Version 1