Analysis of the Nexus Between Coping Strategies and Resilience to Food Insecurity Shocks: The Case of Rural Households in Boricha Woreda, Sidama National Regional State, Ethiopia

Abstract

This chapter reports on the coping strategies employed by households in the event of food insecurity shocks and the nexus between the types of coping strategies and resilience to food insecurity in one of the food-stressed woreda from Sidama National Regional State, Ethiopia. The households use various consumption-based coping strategies that run from compromising the quality of food-to-food rationing. Repeatedly occurring food shortage has also forced some households to employ resilience erosive coping mechanisms such as selling reproductive assets. Such coping strategies have an important implication on the household’s capacity to cope with the future food insecurity-related shocks, with a statistically significant relationship between the nature of coping strategies utilized in response to previous food insecurity-related shocks and the household’s resilience to upcoming shocks. Coordinating crises management based on humanitarian intervention with households’ livelihood assets protection and resilience strengthening is the major policy implication of this study.

1. Introduction

Food is the most basic need for survival, growth, and good health of human beings. Freedom from hunger is the most fundamental human right that can be attained if an individual is food secure [1]. However, a significant proportion of the world’s population still lives under the situation of food insecurity. As it is clear from the FAO et al. [2] report on the state of food and nutrition in the world, even the prospect itself is not sufficiently bright to the extent expected. Five years after the world committed to ending hunger, it has been learned that, the world is still off track to achieve this objective by 2030. Given the current pace, the world is making headway neither towards Sustainable Development Goal target 2.1, of ensuring access to safe, nutritious and sufficient food for all people all year round, nor towards target 2.2, of ending all forms of malnutrition [2].

Looking at the trends and projections of the state of global food insecurity may help to understand this claim. According to the same report, the number of undernourished people was 690 million in 2019 (60 million more than in 2014), and is expected to exceed 840 million in 2030. When it comes to Africa, the continent’s share of undernourishment prevalence for 2019 exceeds one-third of global undernourishment with about 250 million undernourished people. This figure represents about 19% of its overall population and is projected to be about26% in 2030 [2].

Various reports show that Ethiopia hosts a handful proportion of food insecure people. For instance, WFP and CSA [3] report the persistence of poverty and food insecurity despite the country’s efforts to counteract the situation. MOFED [4] reported a level of food poverty prevalence of 33.6% in 2014 versus31.8% in 2012/13. However, Ethiopia is moving in a good direction to improve the situation. A joint report of WFP and CSA [5], showed that the country has made tremendous socio-economic progress that resulted in the reduction of the prevalence of hunger and undernourishment to 25.5%. Nevertheless, the country still embraces a noticeable level of food-insecure people.

Response mechanisms to food insecurity shocks varies based on the objectives of the agents responding to it as well as the level at which they are targeted. As active actors/agents/of their own, households employ various coping strategies (response mechanisms) in the event of shocks that challenge their food security. According to Maxwell and Caldwell [6], USAID [7], and Degefa [8], such strategies are not uniform and may also not be equally sustainable, as in some cases they may erode household’s capacity to withstand future food insecurity shocks. Although, effects of households’ coping mechanisms and resilience to future shocks have been widely discussed, mainly at the conceptual level, empirical statistical evidences on the nexus are quite limited.

For instance, though Carter et al. [9] provide elegant theoretical explanation on the linkage between shock-initiated coping mechanisms and a household’s resilience, the unavailability of data on coping strategies constrains them from including this variable in their estimation model. The study of Tran [10] fails to make the distinction between positive and negative coping at the empirical level and focuses only on the immediate positive effects to recover from shocks. However, a particular coping strategy, though resilience erosive, can contribute to smooth current consumption and/or recovery from shocks. Moreover, capturing resilience only through the recovery speed proxy is also too simplistic. Thus, there is an increasing understanding of resilience as an ex antecapacity of households to withstand the effect of shocks [11, 12, 13, 14, 15, 16]. This way of conceptualizing enables to better capture the essence of resilience as absorptive (buffering), adaptive, as well as transformative capacity in addition to recognizing a futuristic nature. Considering this scenario, this chapter brings forward the linkage between resilience and coping mechanisms, focusing on Boricha woreda as a case study. For that, the following interrelated questions are discussed: (1) how do the study area’s households respond to food insecurity shocks? (2) does the resilience level of households vary based on the nature of previously employed coping mechanisms?

2. Linkage between household’s resilience to food insecurity and coping mechanisms

Maxwell and Caldwell [6] identify four coping strategies that households employ when they face food shortages or do not have the resources to purchase food. They include taking action on the quality of food to eat, looking for options that help increase food supply, reducing the number of household members that they have to feed through such mechanisms like sending some of them to neighbors’ houses, and managing the deficit through mechanisms such as food rationing. Conceptually, these strategies are consumption-based ones having a lesser impact on the households’ capacity to cope with future food insecurity shocks.

Carter et al. [9] put the households’ actions to cope with shock-induced food security challenges in a certain rational decision-based logical order. As per this source, initially households choose to depend on the markets and other institutions that they have access to. To maintain their consumption standard without further asset depletion, households with financial market access or access to informal finance might borrow against future earnings. Resorting to insurance arrangements, seeking for and receiving disaster aid as well as working for long hours are also coping options that they can exercise before taking action against their productive assets. Households without access to such options may opt to sustain their consumption by drawing down on their assets: the decision which they argue can further increase the sensitivity of assets and weaken the future. Finally, households may cope by reducing consumption. This coping strategy can be the last option for those lacking other assets or options and may also be pursued by households who are reluctant to increase their future vulnerability due to depletion of the stock of assets. However, coping by reducing consumption is regarded unfavorably as it does have multiple costs, i.e., immediate hunger as well as the long-term effect on children’s growth and development [17].

To the linkage between coping mechanisms and shocks, it is postulated that adverse events (shocks) may cause a decline in assets and incomes in the short-run and might have negative effects on household livelihoods in the longer-run [10]. However, the extent of the effects, depends on the nature of the shocks, the asset dynamics, as well as on the coping strategies employed. Carter et al. [9] opine that when a given shock happens, it will have both direct and indirect impact on households’ resilience to future shocks. Firstly, the shock itself brings direct harm to the quality of households’ asset. As households’ respond to shocks using their assets and resources, the indirect impact comes via such responses to a particular shock. The whole idea here is that the coping mechanisms used in response to food insecurity-related shocks at a given point can cause a decline in the household’s ability to cope with future shocks depending on the strategies employed in between two time periods.

The origin of the concept of resilience is linked to the field of ecology. According to Holling [18], in ecology, the term resilience is used as a measure of systems persistence and capacity to absorb changes and disturbances and still retain the same relationship with state variables. To a household’s food security, resilience has been conceptualized as the ability of the household to maintain its food security withstanding shocks and stresses, depending on the options available and its ability to handle risks [11]. Accordingly, resilience is a multifaceted capacity: absorptive, adaptive, and transformative. While explaining the linkage between the nature of coping strategies and resilience, Frankenberger et al. [19] sustain those certain strategies may have negative and permanent consequences to resilience. Positive coping strategies are those based on available skills and resources, to face, manage and recover from shocks and that do not compromise resilience. On the other hand, negative coping strategies, if employed, undermine future options making it more difficult to cope with the next shock or stress [20]. Hence, it can be argued that the resilience status of a household at a particular time point (resilience to future food insecurity shocks) is partly a reflection of the type of coping strategies previously employed. Figure 1 represents this conceptualization.

Figure 1.

Conceptual representation of food insecurity shocks-coping strategies-resilience nexus.

3. Illustrative case

3.1 Description of the study area

The illustrative case is based on the data collected from one of the food-stressed woredas fromthe Sidama National Regional State called Boricha woreda. As per the CSA [21] report, Boricha woredahas a total population of 250,260 inhabitants, of whom 125,524 are men and 124,736 women. Yirba is the administrative capital. The area has two rain periods a year: the short rainy months (the belg rain-from March to May) and the long rainy months (the kiremntrain from June to October). The remaining months constitute the dry season when both humans and animals face water shortages. Besides that, Boricha woredais known for unreliable rainfall patterns (both in amount and periodicity) for a couple of years and associated food stresses. Mixed subsistence agriculture supports the livelihood of the population. Enset and maize are the two dominant food crops grown at the household level. Khat, coffee, and livestock are also part of the household’s economy in the area through their concentration is not uniform across all kebeles.Complete dependence on rain-fed farming for subsistence together with rainfall variability exposes people to high risks of harvest loss that easily translates into food insecurity [22]. There are 39 kebeles (the lowest administrative unit)in Boricha woreda. Of these, three are urban and 36 are rural. According to SNNPR [23] livelihood profile report, these Kebelesare classified into three livelihood zones: Sidama Coffee Livelihood, Sidama Maiz Belt Livelihood, and Agro-pastoralist Livelihood.

3.2 Methodological briefing

Based on insights from literature and the resulting framework presented in Figure 1, it was assumed that the coping strategies employed by households in response to food insecurity shocks that happened at time (T0), can have an influence on the resilience level at a time (T1) in a way that households with negative coping strategies at T0scoreless on resilience at T1. As the households’ coping mechanisms are the response actions to shocks, data can be captured usually ex post(or retroactively). Accordingly, the linkage between the level of resilience and household coping mechanisms was examined based on surveys before time T1 in response to various stressors/shocks challenging their food security situation. Conceptually, the study examined the relationship between the nature of coping mechanisms employed at time (T0) and the resilience status of households at the time (T1), the proxy of households’ capacity to effectively respond to future food insecurity shocks.

The selection of the illustrative study was based on a cross-sectional survey conducted by using structured questionnaires and key informants’ interviews. It involved 420 randomly selected households from three randomly selected kebeles(one kebelefrom each livelihood zone). As resilience is a multi-dimensional concept that is not directly observable, it has to be measured through a proxy. To this end, the study adopted the FAO’s Resilience Index Measurement Analysis Model (RIMA) originally proposed and used by [11, 12]. The model quantitatively assesses household resilience through latent variable modeling. Accordingly, in the study, resilience was treated as a latent variable to be estimated by using seven indicators (dimensions): agricultural assets, agricultural technology adoption, access to basic services, social capital, social safety nets, adaptive capacity, income and food access. Each of these seven indicators of resilience is a latent variable to be estimated using observable household-level variables. Using the Principal Component Analysis (PCA), the estimation of resilience score (index) was done hierarchically. First, an index for each of the above dimensions of resilience was done separately using observable variables. Then, the resilience score for each household was estimated with PCA based on the indices of those resilience dimensions (indicators) (see Figure 2). All the seven indicator variables were strongly loaded on the first component and the component scores were used as resilience index for each household. The following path diagram (Figure 2) has been adapted from [12], in order to visually depict this estimation procedure.

Figure 2.

Household’s resilience estimation procedure.

At the household level, the resilience index was estimated using the Eq. (1) below, which was further transformed using the weighting mechanisms and applying the Bartlett method of component scoring. The Bartlett method was selected as it generally produces latent variable scores that are unbiased and univocal [24].

$${\mathrm{R}}_{\mathrm{i}}={\mathrm{w}}_{\mathrm{AA}}{\mathrm{AA}}_{\mathrm{i}}+{\mathrm{w}}_{\mathrm{ATA}}{\mathrm{ATA}}_{\mathrm{i}}+{\mathrm{w}}_{\mathrm{ABS}}{\mathrm{ABS}}_{\mathrm{i}}+{\mathrm{w}}_{\mathrm{SC}}{\mathrm{SC}}_{\mathrm{i}}+{\mathrm{w}}_{\mathrm{SS}}{\mathrm{SS}}_{\mathrm{i}}+{\mathrm{w}}_{\mathrm{AC}}{\mathrm{AC}}_{\mathrm{i}}+{\mathrm{w}}_{\mathrm{IFA}}{\mathrm{IFA}}_{\mathrm{i}}$$E1

where:

$R_i$= resilience of household i, ${\mathit{AA}}_i$= agricultural assets, ${\mathit{ATA}}_i$= agricultural technology adoption, ${\mathit{ABS}}_i$=access to basic services, ${\mathit{SC}}_i$= social capital, ${\mathit{SS}}_i$= social safety nets, ${\mathit{AC}}_i$= adaptive capacity, ${\mathit{IFA}}_i$= income and food access, $w=$Weight for each indicator of resilience.

The surveys to analyze the coping mechanisms measurements included two sets of questions: consumption-based (strategies employed in the last 7 days before the date of the survey) and non-consumption based (strategies used in the last 2 years preceding the survey date). The analysis of data on coping strategies was done descriptively using percentages. The linkage between households’ resilience and the previously employed coping mechanisms was examined using contingency table and chi-square tests as well as using the odds ratio. In the analysis, households were categorized into two groups: those who previously employed negative (resilience erosive) coping mechanisms and those who did not employ such coping strategies over the past 2 years. In the current study, such categorization was done based on insights from conceptual literature such as [7, 19]. Hence, based on these conceptual works, coping strategies such as selling of reproductive animals, oxen used for farming, and land, land rental, withdrawal of children from school, borrowing money at the high interest rate, and diversion of loans from MFIs were treated as resilience erosive or negative strategies. Accordingly, households who did use any of these coping strategies over the past 2 years were classified under the negative coping category. Based on Guyu and Muluneh [15] and considering the relative location of the surveyed households on the latent variables (resilience scores,) the study households were categorized into resilient and none- resilient groups.

4. Findings and discussion

4.1 Coping strategies adapted

Literature indicates that the response of households to food insecurity challenges include different coping strategies. These may involve the modification of consumption habits (consumption-based coping strategies) and/or use of the available resources (non-consumption-based strategies). For instance, Christiaensen and Boisvert [25] contend that when they anticipate food shortage people start to consider changing their consumption habits rather than waiting until food is completely exhausted. Though such change in the consumption habits is generally believed to be a short-term adjustments, it could go long as a normal habit even in the situation where non-consumption-based strategies too are activated. This is mainly true in the situation where a given community lives under long standing food stress in terms of availability and/or access. The point here is that though non-consumption-based strategies such as selling key productive assets are used, foods obtained through such actions could still be subject to consumption-based coping such as rationing. This can lead us to safely argue that the two sets of coping strategies, consumption and non-consumption based, should not be seen as completely isolated and mutually exclusive as they appear in the literature. Notwithstanding the complexity here, the analysis of the household’s coping strategies was done in light of the general assumption that households are rational decision-makers and thus, the first options are those with the least impact on livelihood or future food security.

Consumption-based coping strategies constitute short-term alteration of consumption patterns. Writers like Watts [26], Corbett [27], and Devereux [28] consider them as easily reversible strategies that do not jeopardize long-term prospects as they mostly do not require a commitment of domestic resources. The households’ responses summary (Table 1) indicates that 60.2% (253) of the households rely on less preferred foods at least once in a week. 45.5% (191) reported that the consumption of adults was restricted in favor of children. According to one of the elderly key informants “during food shortage, usually mothers take the burden of not having to eat giving priority to children and father”. Similarly, a total of 181 (43%) and 141 (33.6%) households limited portion sizes and reduced the number of meals. The proportion of households who reported that they borrowed food or relied on the help from a friend/relative and purchased food on credit was 39.5% (166) and 32.4% (136), respectively. All the remaining coping strategies summarized in Table 1 were utilized by a small proportion of the households. Only 7.6% (32) of the surveyed households indicated that they relied on wild foods and/or immature crops. Probably, this could be due to the timing of the survey, as it was conducted just after the harvesting period (dry season). Similarly, only a small number of households, 17.9% (75), gave priority to working members at the expense of non-working members, and only 1.7% (7) fastens the entire day. Again, a relatively small proportion of total households, 13.6% (57), consumed seed stocks held for the next season at least once a week. The proportion of households who engaged in the coping behavior of sending family members to eat elsewhere and begging was12.1% (51) and 2.6% (11), respectively. Such findings could be because the experienced level of food insecurity might not be of the extent that forces households to engage in such behaviors or due to the strong local culture that discourages such practices.

Coping strategy	Number/proportion of households employed
	Count	Percentage (%)
Relied on less preferred foods	253	60.2
Borrowed food or relied on help from a friend/relative	166	39.5
Purchased food on credit	136	32.4
Relied on wild foods, hunt, or immature crops	32	7.6
Consume seed stock held for next season	57	13.6
Household members sent to eat else where	51	12.1
Household members sent to beg	11	2.6
Portion size at mealtimes limited	181	43
Consumption by adults restricted in order for small children	191	45.5
Priority given for working members of household at the expense of non-working members	75	17.9
Meals eaten in a day reduced	141	33.6
Entire days skipped without eating	7	1.7

Table 1.

Consumption based coping strategies.

Complementary and non-consumption-based coping strategies (Table 2) included, selling reproductive animals at least once within the last 2 years period (42.6%), and renting (10%) or selling (2.1%) their lands (10%). About 20.7% (87) and 21% (88) of the households had removed their children from school and borrowed money at high-interest rates respectively. A total of 37.6% (158) households reported that they coped by selling small animals and about 19% (80) migrated to nearer areas in search of wage labor. Almost none, 1.9% (8), of the households had engaged in the coping behavior of diverting loans from Monetary Financial Institutions (MFIs) to consumption and only 4.8% (20) households had drawn on financial savings to respond to the food insecurity problem. This could be due to a lack of cash savings to draw from and/or limited access to MFIs both of which are common in the rural context. Nearly half, 51.7% (217), reported that they have appealed for food aid to overcome food insecurity within the last 2 years. One-third of the households, 33.3% (140), reported that they used selling firewood as a coping mechanism (see Figure 3).

Coping strategies	Number of households adopted
	Count	Percentage (%)
Sold reproductive animals	179	42.6
Sold oxen used for farming	98	23.3
Sold land	9	2.1
Rented out land	42	10
Removed children from school	87	20.7
Borrowed money at high interest rate	88	21
Sold small animals	158	37.6
Migrated to nearer areas to wage labor	80	19
Drawing on savings	20	4.8
Selling fire wood	140	33.3
Diverting loans from MFIs to consumption	8	1.9
Appealed for aid	217	51.7

Table 2.

Non-consumption based coping strategies used by households.

Figure 3

Household members taking fire woods collected form forests to market centers.

According to the key informants, they collect fire wood from the forest around Bilate River towards the border of Loka Abaya woredaand supply to Dila Anole and Balela towns. From our discussions, we further learned that due to persistent food stress, poor people have made collecting and selling fire wood as a regular source of income for food purchase. However, the issue of concern exists. That is, if left unchecked, such a heavily reliance on forests could wipe out the only left over of the ancient forests in the area. Almost all elderly key informants stressed that in the past most of the woredahad been covered by dense forests that hosted many wild animals until the downfall of the emperor regime. But, the increasingly growing demand for farm land since then has resulted in the clearance of forests to its demise.

4.2 Relationship between previously employed coping mechanisms and resilience status (level) of the households

As referred above, several authors such as Frankenberger et al. [19], Carter et al. [9], Tran [10], and USAID [7], pinpoint that the types of coping mechanisms employed by households in response to previously happened shocks can affect their resilience to future shocks.

Based on these conceptual backdrops, we have endeavored to understand how the previously used coping strategies of households relate to their resilience status. To this end, households were asked if they experienced one or more shocks challenging their food security situation in the last 2 years preceding the survey and the responses are summarized in Table 3. Most of the surveyed households, 79.3% (333), experienced one or more types of shocks that they believe affected their food security situation. Households have also identified a set of coping strategies employed in the past 2 years to cope with food insecurity problems/shocks (Table 2).

Variables	Response	Count	Percentage (%)
If shocks affecting ability to feed HHs occurred within the last 2 years	Yes	333	79.3
	No	87	20.7
Number of shocks experienced*	Only one	99	29.7 (23.6% of total)
	More than one	234	70.3 (55.7% of total)

Table 3.

Previously experienced food security situation threatening shocks.

^*

List of shocks include crop failure, household member death, livestock death, and illness.

When it comes to identifying negative coping strategies (erosive resilience), it seems that literatures lack perfect unanimity. With the argument that they undermine future options making it more difficult to cope with next shocks, Pasteur [20] considers strategies such as delaying medical treatment, exploiting natural resources, taking children out of school, eating less, eating less nutritious food, and eroding productive assets as resilience erosive coping strategies. However, some of the strategies considered as negative coping here are consumption-based (temporary adjustments on eating) that are considered by others as easily reversible. Specially, stage 2 and stage 3 coping strategies from the list identified by Watts [26] and Frankenberger [29] are generally treated as erosive coping mechanisms. Based on the literature and on study area’s context, selling reproductive animals, oxen, and land, or renting land, taking children from school, borrowing money at high-interest rates, and diversion of loans from MFIs to consumption were considered as negative (resilience erosive) coping in this illustrative case. Accordingly, households were classified into two coping categories (Table 4): those who used negative coping in the past 2 years and those who did not. As indicated in the table, 59.5% (250) of the households employed one or more negative (erosive) coping strategies in the last 2 years preceding the date of the survey.

		Frequency	Percent	Valid percent	Cumulative percent
Valid	Non resilience erosive (positive) coping	170	40.5%	40.5%	40.5%
	Resilience erosive (negative) coping	250	59.5%	59.5%	100.0
	Total	420	100.0	100.0

Table 4.

Households by coping type.

The households’ resilience position (status) was determined based on their relative resilience scores and using the criteria of [15]. Based on relative resilience score (index) achieved by households, Guyu and Muluneh [15] classify four resilience categories: Vulnerable (resilience index (RI) < 0.100). Moderately Resilient (0.100 ≤ RI < 0.250), Resilient (0.250 ≤ RI < 0.500) and Highly Resilient (RI ≥ 0.500). Using the resilience scores estimated through the Bartlett method in PCA and applying these cutoff schemes, households are categorized into four categories (Table 5). A very significant proportion of the surveyed households (61%) was not resilient (or vulnerable to food insecurity shocks) and only 39% was resilient at different levels. With these pieces of information on the nature of previously employed coping and resilience status, now the discussion turns to examine the relationship between the nature of coping mechanisms and the relative resilience position (status) of the households. Our analysis proceeds with the proposition that the nature of previously used coping strategies can affect the predictive resilience of households (estimated at time T1) in the form that those with prior negative coping strategies scoreless on resilience. Contingency Table and chi-square test statistic, and the odds ratio were employed to analyze and test this proposed relationship of the two variables. Table 6 presents cross-tabulation of previously employed coping types and households’ resilience levels. About 59.5% (250) of the households used one or more types of erosive resilience (negative) coping strategies within the last 2 years. From this group, only 19.6% (49) was found to be resilient (scoring relatively high on resilience index) at time T1 (time of the survey). Most households, 80.4% (201), that adapted one or more negative coping strategies were found to be non-resilient. On the other hand, out of the total households who did not previously use negative coping strategies, 67.6% (115) was found to be resilient at time T1 (scoring relatively high on resilience index) against 32.4% (55) scoring relatively low on resilience (non-resilient).

Measurement	Households by resilience category				Total
	Non resilient	Moderately resilient	Resilient	Highly resilient
Count	256	22	18	124	420
Percent	61%	5.2%	4.3%	29.5%	100

Table 5.

Distribution of household resilience status.

			Coping type		Total
			Non-resilience erosive (positive coping)	Resilience erosive (negative coping)
Resilience level	Resilient	Count	115	49	164
		% within resilience level	70.1%	29.9%	100.0%
		% within coping type	67.6%	19.6%	39.0%
		% of Total	27.4%	11.7%	39.0%
	Non-resilient	Count	55	201	256
		% within resilience level	21.5%	78.5%	100.0%
		% within coping type	32.4%	80.4%	61.0%
		% of Total	13.1%	47.9%	61.0%
Total		Count	170	250	420
		% within resilience level	40.5%	59.5%	100.0%
		% within coping type	100.0%	100.0%	100.0%
		% of Total	40.5%	59.5%	100.0%

Table 6.

Cross-tabulation of households’ resilience level and previously used coping strategy.

The Chi-Square test was run as a way of checking if the observed frequency (or percentage) differences in the contingency table (Table 6) were statistically significant. In statistical terms, it tests the implicit null hypothesis that there is no relationship between types/nature of previously employed coping strategies and the resilience status of the households. That is, it tests the hypothesis that the household’s resilience score (status) at time T1 is independent of types of coping methods employed by a household in response to shocks that occurred before time T1. The result of the Chi-Square test (Table 7) revealed high significance for χ2 (1) = 98.149, P < 0.001 indicating an association between household’s resilience status and types of previously employed coping strategies. Besides the association between these two variables, it does not show the strength of the relationship that has been detected. Therefore, the Phi test for 2 by 2 contingency table, was also performed [30] giving a noticeable level of association between the household’s resilience level and types of coping strategies previously employed (Table 8). The sign of the relationship is also as expected as the two variables were coded similarly.

	Value	Df	Asymp. Sig. (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Pearson Chi-Square	98.149a	1	0.000	0.000	0.000
N of valid cases	420

Table 7.

Tests of association between resilience status and coping type.

^a

0 cells (0.0%) have expected count less than 5. The minimum expected count is 66.38.

^b

Computed only for a 2 × 2 table.

		Value	Approx. sig.	Exact sig.
Nominal by Nominal	Phi	0.483	0.000	0.000
N of valid cases		420

Table 8.

Test of the strength of association (resilience level and coping type).

^a

Not assuming the null hypothesis.

^b

Using the asymptotic standard error assuming the null hypothesis.

Both association (Chi-Square) and strength of association (Phi test) tests highlighted the existence of meaningful relationships between the two variables under consideration. To further check the strength of association between the two variables the odds ratio was used as a supplement to the Phi test. The odds ratio here refers to the ratio of the odds that a household will be resilient to future shocks with no prior use of negative coping strategies to the odds that a household will be resilient through it previously used some kind of negative coping strategies. Based on frequencies in Table 6, the odds ratio was computed as:

$$\begin{array}{rcl}\text{Oddsratio}& =& \text{Odds of being resilient with}\mathrm{no}\text{prior}\mathrm{use}\text{of negative coping}\\ & & ÷\text{Odds of being resilient with prior}\mathrm{use}\text{of negative coping}\hfill \end{array}$$E2

$$\begin{array}{l}\text{Odds of being resilient with}\mathrm{no}\text{prior}\mathrm{use}\text{of negative coping}\\ =\begin{array}{l}\text{Number of resilient households}\mathrm{who}\text{didn}’\mathrm{t}\mathrm{use}\text{negative coping}\\ ÷\text{Number of nonresilient households}\mathrm{who}\mathrm{did}\text{not}\mathrm{use}\text{negative coping}\end{array}\\ =115÷55=2.0909\end{array}$$E3

This ratio shows that the number of households who are resilient with no prior use of negative coping is as twice as those who are non-resilient though they did not employ negative (erosive) coping before. It is also possible to be resilient or non-resilient to future shocks without prior negative (erosive) coping. However, it is more likely to be resilient than non-resilient given the initial state (previous experience in terms of coping type) is that of no negative (erosive) coping strategy.

$$\begin{array}{l}\text{Odds of being resilient with prior use of negative coping}\\ =\begin{array}{l}\text{Number of resilient households who did use negative coping}\\ ÷\text{Number of nonresilient households who did use negative coping}\end{array}\\ =49/201=0.24378\end{array}$$E4

The ratio here shows that the number of resilient households experiencing previous negative coping is about four times less than the number of non-resilient households.

Given the two pieces of information (odds ratios presented above) and referring to the first equation, the odds ratio of interest here (the odds that a household will be resilient to future shocks with no prior use of negative coping strategies to the odds that a household will be resilient through it previously used some kind of negative coping strategies) can be computed as follows:

$$\mathit{\text{Oddsratio}}=2.09090÷0.24378=8.57$$E5

The odds ratio indicates that households who did not previously use negative coping strategies were 8.57 times more likely to be resilient to future shocks. So, the clear implication of this finding is that the type of coping mechanisms used in response to given food insecurity-related shocks at a particular point in time can have an impact on households’ ability to respond to the upcoming shocks. This finding is in line with the Chi-Square test result above and the extant theoretical literature discussed in the chapter.

5. Conclusion

Depending on the initial state of the households, some of the coping strategies can lead to the poverty trap and erode the ability to cope with similar problems in the future. If left uncontrolled, even the coping mechanisms with no immediate individual impact, like selling firewood, may not be environmentally sustainable. This is especially true in the case of the study area as the source of firewood collection is, mostly, the single leftover of the ancient forest, which is confined to marginal areas around the Billate River. Additionally, the coping mechanisms utilized currently by the households can have important implications on their capacity to cope with future shocks, depending on their resource base. Hence, well-targeted interventions that go beyond saving lives (humanitarian emergency) and focusing on livelihood assets protection and capacity building to future shocks is the recommended policy option.