Journal of Food Security, 2017, Vol. 5, No.

6, 259-272
Available online at http://pubs.sciepub.com/jfs/5/6/6
©Science and Education Publishing
DOI:10.12691/jfs-5-6-6

Food Security Status of Peri-Urban Modern Small Scale

Irrigation Project Beneficiary Female Headed
Households in Kobo Town, Ethiopia
Goitom Sisay Mengesha*

University College of Science, Department of Geography, Osmania University, Hyderabad, India

*Corresponding author: sisaygoitom@gmail.com

Abstract This study has attempted to evaluate the household food security status of peri-urban modern small
scale irrigation project beneficiaries. To attain this objective, a cross sectional survey method using structured
questionnaire was employed on randomly selected 333 households. Household Dietary Diversity Score (HDDS),
Food Consumption Score (FCS) and Per Capita Net Food Availability (NAF) were employed to evaluate the
household food security status of the sample households. Furthermore, FGT family of indices were also computed.
The HDDS result of this survey revealed that seven food groups were reported to be consumed with the mean
HDDS value of 3.42, 3.84 and 3.21 for total samples, participants and non-participants respectively. The calculated
FCS value for total samples, participants and non-participants was 42.74, 44.89 and 41.64 respectively. Furthermore,
the computed NAF value revealed that of the total samples, 198 and 135 households were found to be food secure
and food insecure respectively. This study illustrated household food insecurity, low dietary diversity and food
consumption to be far more noticeable in non-participants than participants.
Keywords: food consumption score, food security/insecurity, female headed household, household dietary
diversity, per capita food availability
Cite This Article: Goitom Sisay Mengesha, “Food Security Status of Peri-Urban Modern Small Scale
Irrigation Project Beneficiary Female Headed Households in Kobo Town, Ethiopia.” Journal of Food Security,
vol. 5, no. 6 (2017): 259-272. doi: 10.12691/jfs-5-6-6.

households are the hardest hit [4]. Thus, problems related

to increasing food availability, feeding the population,
1. Introduction improving their nutritional status and reducing poverty
levels continues to confront decision makers in many
Food insecurity has been a problem of worry to countries [5].
humanity from the beginning of time. It becomes a The latest report of [7] estimated that about 795 million
prominent policy agenda as a result of recent food crises people globally are believed to be undernourished; out of
both at a regional and global level as well as renewed which the vast majority (780 million) lived in the
commitment from donor nations to address chronic hunger developing world in the period 2014-16. The report
[1]. Since 1948 through the Universal Declaration of revealed that Africa and Sub-Saharan Africa had 232.5
Human Rights, enhancing food security and reducing and 220 million undernourished people respectively in the
under nutrition have been promised [2]. specified time.
The emergence of food security as a concept traced Despite the design and implementation of successive
back to the 1974 United Nation Food and Agricultural national food security strategies as well as considerable
Organization, World Food Conference in Rome at the development potential, transitory food insecurity and
time of global food crisis. This conference considered poverty in Ethiopia are overwhelmingly perpetuated issues
food availability as a central argument which stated “a for several decades. Many households in Ethiopia are
secure, adequate and a suitable food supply for everyone”. unable to buy or grow enough food to feed their families
Then, the 1996 World Food Summit and the United and need food aid as well as food imports each year for
Nations Millennium Declarations (2000) targeted to halve their survival [8,9].
the world’s hungry people by 2015 [3]. The history of famine in Ethiopia is thought to be as old
Despite considerable efforts and some progress made, it as the history of man who ever lived in it perhaps 250 BC.
seems that the goals are far from being achieved by many However, available literature on recorded history of
countries as chronic hunger in the world has increased famine in Ethiopia refers to the 19th and the 20th centuries
rapidly [3,4,5]. According to [6], food security is generally [8,10]. The country has faced some 44 severe famine
about to deteriorate at the aggregate level as the share of catastrophes. Drought has been occurring in Ethiopia at a
population that is food insecure is projected to rise over frequency of every 3 to 5 years which makes the country
15% by 2025. The poorest, landless and female headed to be one the world’s food aid dependent countries. Due to
260 Journal of Food Security

poor rain and effects of the El Niño, Ethiopia experienced food self-sufficiency both at individual and national level
its worst drought in 50 years where 9.7 million people through double cropping [25].
were in need of emergency assistance in August 2016 [11]. As the study area has been identified amongst the
Food insecurity exists when people lack secure access to drought prone, moisture deficit and the people are food
sufficient amount of safe and nutritious food for normal insecure, efforts have been made by KGVDP and Amhara
growth and an active and healthy life [7]. Water Works Construction Enterprise office to expand the
Ethiopia is the least urbanized nation where its degree practice of modern small scale irrigation since 2003. As
of urbanization in 2015 was 19.5 % [12] and projected to stated in [26] document, as of the commencement of the
reach 38% in 2050 [13]. On the other hand, the nation has program, out of 17000 hectare of irrigable land, only 1794
the highest rate of urbanization; between the last two hectares of land have been irrigated and benefiting 5744
censuses (1994 and 2007), urban population has increased households; of which 3844 (66.92%) were male headed
at an average annual growth rate of 4.9 % [14]. Such and the rest 1900 (33.08%) were female headed. Measuring
increase in the size and proportion of the urban population the effectiveness and efficiencies of Policies, programs
has brought with it a new challenge of widespread and and projects targeted towards improving food security
increasing urban poverty, a high unemployment rate, low is a major and challenging task for any institution and
governance capacities, weak infrastructure, poor municipal governments. Furthermore, due to the complex and
finance in cities and high demand for agricultural products multidimensional nature of food security looking for
[15]. better measures of food security still remains a major
As of the [16] report, 29.6% of the Ethiopian challenge [5].
population lives below the poverty line: of which 30.4% There is also no consensus as there are conflicting
and 25.7% lives in rural areas and urban areas respectively. evidences from different projects and regions about the
Poverty is a driving force for household food insecurity performance, sustainable and equitable role of small scale
and food insecurity again, impoverishes a household. As a irrigation to household food security. Moreover, to the
result, explanations about poverty and food insecurity are best of the writer’s knowledge, in the study area no study
inseparable as they are among the undesirable livelihood has been conducted so far targeting female headed
outcome [8,17]. The problem of food insecurity which households. The purpose of this study was therefore to
was hardship borne largely for the rural population is evaluate the household food security status of peri-urban
recently becoming a growing problem among the poor and modern small scale irrigation project beneficiary female
the disadvantaged population in Ethiopian urban areas headed households in Kobo town, Ethiopia.
[18]. Official statistics have also revealed that recently
urban poverty in Ethiopia has been growing at a rapid rate
than rural poverty. For instance, between 1995/96-1999/00, 2. Material and Methods
urban poverty has increased by 11.1% while rural poverty
has declined by 4.2% [19]. In response to soaring food 2.1. Study Area Description
prices as of 2007 and disrupted food supplies, many urban
and peri-urban poor families have no alternative than to This study was conducted in Kobo town and its
turn to urban and peri-urban irrigated agricultural surroundings. The town is the administrative centre of
activities for their livelihood and survival. Urban and Raya Kobo Wereda (district) and Kobo town administration
peri-urban food production is in many cases a response of with five kebeles (the lowest administrative unit in
urban poor as a survival strategy [20,21]. Urban and Ethiopia). It is situated in the north-eastern tip of Amhara
peri-urban irrigated agriculture, mostly practiced by National Regional State, north Wollo administrative zone,
women and female headed households have the benefit of Ethiopia [27].
market proximity and freshness. The town lies on Addis Ababa-Mekelle national highway,
To meet the food demands of its rapidly growing about 570 kilometers north of Addis Ababa (the national
population, it is expected that Ethiopia must double its capital) with a geographical coordinates between 11o 54’ 04” N
cereal production by 2025. Irrigation as one integral part and 120 20’ 56’’N latitude and between 390 25’ 56” E and
of water sector is a means by which agricultural 390 49’ 04” E longitude. The landscape of the Wereda is
production can be boosted to meet the growing food characterized by a broad fertile plain (65%) whereas the
demand of the country [22]. As Ethiopia has a significant rest 20, 6, 5, and 4% are mountainous, rugged, gorges and
potential of irrigation both in terms of the available land swampy respectively. In the study Wereda altitude ranges
and water resource, the government has prepared a water from 1400-3100 meter above mean sea level where the
sector development program to be implemented between average altitude is 1500 meter above mean sea level [27].
2002 and 2016. The estimated total irrigable land of The study Wereda has an aggregate human population
Ethiopia is 5.3 million hectares; of which potential of 239, 504 of which 120, 383 (50.26%) were men while
irrigable area, only 4 to 5% (640,000 hectares) is under the remaining 119,121 (49.74%) were women. Out of the
irrigation development [23]. In line with this, development total population, 33,135 populations (20.15%) are urban
and expansion of small-scale irrigation and rainwater dwellers; of these urban dwellers male and female
harvesting strategies particularly in arid and semi-arid population constitutes 16311 (49%) and 16824 (51%)
peri-urban and rural areas of the country become central to respectively. The study town had a population density of
Ethiopia’s policies and strategies [24]. Such schemes are 119.7 persons per square kilometer with a total area of
intended to sustainably improve the food security and 2001.57 km2 [28].
 Journal of Food Security 261

Figure 1. Location map of the study area

Agriculture practiced in the suburban areas of the study To evaluate the household food security status of
town serves as the main economic stay and means of female headed households, a combination of qualitative
livelihood to the majority of the town’s people. It is and quantitative data analysis techniques were used as it
characterized by traditional mixed farming as it includes gives the chance to look the multiple dimensions of food
both crop production and livestock rearing, dependent security. The core argument for a mixed method design
mainly on rainfall. The main crops produced through was that the combination of both forms of data provides a
rainfall are cereals (Teff, sorghum and maize) and pulses better understanding of a research problem than either
(chick peas). Teff is a very fine, like cereal, scientifically quantitative or qualitative data by itself. Mixed method
known as Eragrostis tef. Furthermore, as of 2003, designs are procedures for collecting, analysing and
horticultural crops (onion, tomato, pepper) and fruits such mixing both quantitative and qualitative data in a single
as Mango and Avocado are being produced with the help study or in a multiphase series of studies [30].
of modern small scale irrigation [27].
The agro-climatic features of the district is characterized 2.2.1. Nature of Data and Method of Acquisition
by three agro-ecological zones locally known as; “Dega” Both primary and secondary sources of data were used
or Temperate (10.7%), “Weyna Dega” or Sub-tropical to address the research question. Primary data was
(61.8%) and “Kolla” or Tropical (27.5%). It experiences collected from sample female household heads (both
low and erratic rainfall with a mean annual rainfall of 670 participants and non-participants) through structured
mm where maximum amount of rainfall happens during questionnaires. This method was preferred among the
August. Rainfall distribution is bimodal where the main other techniques because it could reach to the relatively
rain season occurs between July and September while the large number of respondents. Most of the items of the
small rainy season is from January to April. The annual structured questionnaires were close ended with some
temperature ranges from the least 19°C to the highest partially open ended items. As the respondents were
33°C where 23.1°C is a mean annual temperature [27]. female heads, collecting data about the food items served
for the household members was easy as they are more
2.2. Research Design responsible for the household food preparation. The data
were collected through trained assistants and by the
Research designs are the specific procedures involved researcher after conducting the appropriate test on the
in the research process; data collection, measurement, data constructed questionnaires. Moreover, Key informant
analysis and report writing [29,30]. This study adopted the interviews, Focus Group Discussion (FGD) and personal
cross- sectional survey technique to collect primary data observation were also employed.
as a survey technique is popular and ideal mode of In-depth Key Informant Interviews were administered
observation in the social sciences. On a cross-sectional by the researcher himself with semi-structured open ended
survey design data are collected from samples at one questionnaire to five female headed households and three
specific point in time. According to [31], surveys are officials from Kobo Girana Valley Development Programme
suitable for descriptive, explanatory or exploratory studies. on issues related to irrigation and household food security.
Survey is especially ideal for studies that have individual Key informants provide detailed information on key issues
people as units for analysis. As a result, survey is ideal for that were not provided by other respondents.
this study as it centred both the individual and the Focus Group Discussions were held to gather in-depth
household as units of investigation and analysis. The head information on the concepts, perceptions and ideas of a
of the household served as the chief respondent to whom group pertaining to irrigation and Food security. Accordingly,
the study questionnaire was administered. by preparing checklists and triangulating issues, subsequent
262 Journal of Food Security

discussions were held by forming three groups (composed which reflect food supply and food access not only the
of a minimum of seven members) from the three study kebeles. food security situation but also the degree of vulnerability
As indicated in [29], to diminish subjective bias, to to food insecurity Maxwell and Frankenberger, 1992 in
relate the information obtained under mentioned methods [33]. However, because of its multi-dimensional (Diet
to what is currently happening and as it relatively quality and quantity, psychological, social and cultural
demands less of active cooperation on the part of the dimensions), broad and dynamic nature, identifying an
respondent, personal observation was also carried on appropriate indicator of food security remains a challenging
irrigation sites and homes of respondents. task [34]. As a result, the search for better measures of
For reference purpose or to use as benchmarks against food security that captures all aspects of food security still
which the findings of a study can be tested, secondary data remains a major challenge [5]. Hence, many indicators of
sources like books, articles and other unpublished reports food security had been devised and employed by different
related to the issue were also consulted. organizations; according to [33] there are approximately
450 indicators and 200 definition of food security.
2.2.2. Sample Size and Sampling Techniques Despite the progresses made in devising appropriate
Multi-stage sampling procedure was employed to select measures, the popular indicators of food security still tend
sample households. Accordingly, in the first stage, the to provide information only on one of these dimensions at
study town was selected purposively due to its familiarity a time. Accordingly, use of more than one food security
to the researcher and extensive implementation of modern indicator is a recommended remedy as one indicator
small scale irrigation projects. In the second stage, out of could not wholly explain food security [1,35]. Hence, to
41 modern small scale irrigation projects located in the evaluate the household food security status of female
study Wereda, 15 irrigation sites situated in the three headed peri-urban modern small scale irrigation project
kebeles namely Kobo Zuria, Aradum and Abuware were beneficiaries, Household Dietary Diversity Score (HDDS),
selected purposively; due to their accessibility, proximity Food Consumption Score (FCS) and per capita Net Food
to the study town and number of irrigation beneficiaries. Availability (NAF) were employed as an outcome
These irrigation sites covered 946 hectares of land and are indicator. These measures are the product of food access,
benefiting 2367 household heads of which 1619 (68.40%) availability, stability and utilization dimensions of food
are male headed and 748 (31.60%) are female headed security. Furthermore, compared to income based measures
households. In the second stage, to obtain representative of household food security, consumption based food
samples, the sample size was determined through [32] insecurity measure (HDD and FCS) were preferred for this
online sample size determination software. The size was study as they pretend to reflect household’s ability to meet
calculated using 95% confidence level and 4% margin of their basic needs, less vulnerable to measurement errors
error (confidence interval). As a result, 333 female headed and its closeness to the utility that people effectively
households (44.52%) who are beneficiaries of peri-urban extract from income [36].
modern small scale irrigation; both participants (113) The survey was held at normal or usual days (not at
and non-participants (220) proportionate to their number special occasions) immediately prior to the harvest as it
were incorporated in this study through simple random serves as a baseline for monitoring change due to
sampling technique. The Water User Association members’ an intervention. Food is relatively widely available
registry was used as a sampling frame. On the other hand, immediately after the production season and starts to
purposive sampling technique was also used to identify dwindle as the lean season approaches. Moreover, as
Key Interview Informants and Focus Group Discussion eating outside the home is not a common practice in the
members. study area, application of the questionnaire at the
household level was preferred than at the individual level.
2.2.3. Data Analysis HDDS was employed for this study as it is a good
To evaluate the household food security status of indicator of food and nutrition security for various reasons.
female headed households, a combination of qualitative First, it correlates with measures of food consumption and
and quantitative data analysis techniques were used as it are a good measure of household food access. Second, a
gives the chance to look the multiple dimensions of food varied diet is a worthy outcome in itself. Third, more diet
security. Descriptive and inferential statistics techniques variety is associated with a number of improved outcomes,
such as arithmetic mean, percentage and Standard deviation child anthropometric status, improved haemoglobin
were used to analyse the household food security situation concentration, reduced risk of mortality from cardiovascular
of sample respondents. Bivariate analysis using cross disease and incidence of hypertension. Lastly, diet
tabulations were also done to identify the relationship diversity scores can be collected through household
between participation in irrigation and household food surveys and can be used to examine Food and Nutrition
security status. Accordingly, chi-square and t-test were Security at individual and intra-household levels but it
employed to test the statistical significance of dummy and cannot necessarily indicate the quantity of food consumed
mean value of continuous variables. [2,37,38]. A study by [39] using data from Mali and [37]
Literature on household food security distinguished two in their multi-country analysis of data from 10 countries
indicators of household food security; outcome and process assessed whether household dietary diversity could be
indicators of food security. Outcome indicators are proxies used as indicator of household food security or not.
for food consumption measures either directly as food Finally, they concluded that dietary diversity holds
expenditure and caloric consumption or indirectly through promise as a means of measuring food security. An
nutritional assessment (anthropometric indicators), subsistence empirical study in Ethiopia by [40] also verified that
potential ratio or storage estimates. Process indicators households with better dietary diversity were able to have
 Journal of Food Security 263

better diet quantity. Moreover, dietary diversity is also meats and fish, milk, sugar and honey, oil and fats with a
recommended as an objective to be included in each weight of 2,3,1,1,4,4,0.5 and 0.5 respectively), second
country’s food based dietary guidelines [41]. On the other sum all the consumption frequencies of food items of the
hand, the major challenge of employing Dietary Diversity same group and recode the value of each group above 7 as
Score is absence of international consensus regarding the 7, thirdly multiply the value obtained for each food group
number of individual foods or food groups used to by its weight that is based on its nutrient content to create
calculate the dietary diversity score and the threshold points new weighted food group scores and at last we sum the
of low, medium and good diversity scores which affect the created new weighted food group scores. Finally,
comparability and generalizing of findings [4,42]. employing the standard and calibrated FCS thresholds,
Dietary Diversity represents the number of different sample households were categorized in to three groups;
foods or food groups consumed by a household over a poor, borderline and acceptable food consumption with a
given reference period. It is measured by summing the value of 0-28, 28.5- 42, and >42 respectively [48]. These
number of foods or food groups consumed over a threshold points were employed by [14] in the household
reference period. The reference period usually ranges from food security study of Ethiopian urban areas. Moreover,
1 to 3 days but 7 days is also often used and periods of up this method was also proved to be applicable in Ethiopian
to 15 days have been reported [42]. For this study, food security studies [38,49].
following the recommendation of [43] and findings of [38],
= Astaple Xstaple + A pulse X pulse + A veg X veg +
FCS
the recall period of 24 hours was chosen. It is less subject
to recall error, less cumbersome for the respondent and A fruit X fruit + A animal X animal + Asugar Xsugar +
also conforms to the recall time period used in many
A dairy X dairy + A oil X oil
dietary diversity studies. It was also proved to be very
consistent in the case of Ethiopia [38]. Where: FCS=is Food Consumption Score
With regard to the choice between food items and food Xi= is Frequencies of food consumption = number of
groups a study by [39] using data from Mali testify that days for which each food group was consumed during the
food group diversity was a stronger predictor of dietary past 7 days
quality than the simple count of individual food. Ai= is weight of each food group
Accordingly, following the work of [39] and because of its The Net Available Food (NAF) for the sample
simplicity, for this study I therefore preferred food groups households was computed using a modified form of a
instead of individual foods for the analysis of diet simple equation known as Household Food Balance
diversity. Furthermore, the selection of food groups was Model adopted from FAO’s Regional Food Balance
done based on the prior knowledge of common household Model by Degefa in 1996 [3]. The choice of Household
dietary patterns and food systems among the study Food Balance Model for this survey was inspired by the
area as well as the guidelines of Food and Agriculture scientific work of many studies such as [3,10,50,51].
Organization of the United Nations [43]. Food group Employing this model they also estimated 5% and 10% of
refers to a collection of food items that have similar the total produces for seed reserves and post-harvest losses
caloric and nutrient content. respectively. In developing countries post-harvest loss is a
To estimate the HDDS, household heads were asked major constraint in achieving food security and it occurs
whether their family members had eaten the listed food in the levels of pre-processing, storage, packaging and
groups or not within the last 24 hours prior to the survey. marketing [5].
Thus, a consumption of each food groups was given the Data used for the computation of NAF for this study
value of 1 if it was consumed by the household members was generated from the field survey embarked on to assess
and a value of 0 if not. As a result, seven major food the period from October 2015 to September 2016.
groups namely (1) main staples, (2) pulses and legumes, Household Food Balance Model is specified as;
(3) dairy products, (4) meat, (5) oils and fats, (6)
vegetables and (7) fruits were reported to be consumed by NGA = ( GP + GB + FA + GB + GG )
the sample households in the reference period. Finally, the − ( HL + GU + GS + GV )
alternatives were summed and the value ranges from 0 to
the maximum of 7. Where, NGA= Net grain available/year/household
Following the thresholds of [44,45,46,47], the cut-off GP= Total grain produced/year/household
points of seven or more food groups, four to six food GB= Total grain bought from market/year/household
groups and less than four food groups were employed to FA= Quantity of food aid obtained/year/household
segment the calculated HDDS as high, medium and low GB=Total grain borrowed /year/household
dietary diversity scores respectively. Furthermore, the GG= Total grain obtained through gift or
cut-off points of medium and high HDDS were used to remittance/year/household
designate nutritionally adequate dietary diversity while HL= Post harvest losses/year (10%)
low HDDS was used to indicate nutritionally inadequate GU=Quantity of grain reserved for seed/year/household
dietary diversity [43,46,47]. (5%)
FCS is a composite score calculated based on dietary GS=Amount of grain sold/year/household
diversity, food frequency and relative nutritional importance GV=Grain given to others within a year
of different food groups consumed by household members. Following the empirical works of [51] and [52], the
Using standard seven day food frequency data, FCS was computed NAF was therefore compared against 2.25
calculated first by grouping all the food items into specific quintals of food grain (225 kg) per Adult Equivalent per
food groups (Main staples, pulses, vegetables, fruits, year, which is roughly cereal equivalent of the
264 Journal of Food Security

recommended national average daily caloric requirement subsistence level. In other words, it will provide the
for a moderately active adult (2100 kcal/person/day set by possibility to estimate the required resources to eradicate
WHO, FAO and Ethiopian Government). Finally, the household food insecurity through proper targeting.
difference between net food grains available and food Giving more weight to the household food security
grains demanded by a household was used to determine severity among the most food insecure households is
the food security status of the sample household. Thus, equivalent to assuming that α > 1. Accordingly, setting
households whose available average net annual per capita FGT (α = 2 ) , squared insecurity gap is mathematically
food grain greater than or equal to 2.25 quintals of food
written as:
grain were regarded as food secure, whereas households
whose net available per capita food grain had fallen below
1 q
Z − Yi 2
the threshold were labelled as food insecure. For this Food Severity Index ( P2 ) =
N
∑( Z
)
purpose, the family size of each household was converted i =1
into Adult Equivalent family size which considers age and
Finally, to evaluate the relationship between Household
sex of each family member of the household. Furthermore,
Dietary Diversity Score, Food Consumption Score, and
the procedure of [53] was also employed to compute the
Net Available Food, Pearson Correlation Coefficient was
incidence, depth and severity of household food insecurity.
employed.
The Foster-Greer-Thorbecke (FGT) measure is given as:
Qualitative data mainly opinions and perceptions
q
Z − Yi α obtained from open ended questionnaires, Key Informant
1
Pα =
N
∑( Z
) Interviews, Focus Group Discussions and personal
i =1 observations were also analyzed qualitatively by using
common expressions and similar opinions.
Where: N= Total number of sample households under
Microsoft Excel and SPSS version 23 software was
the study.
used to organize and analyze the collected primary data.
Yi= is the measure of per adult equivalent annual net
food grain available to the ith household
Z = represents the cut-off point between food secure
and food insecure households (2.25 quintals of food grain
3. Hypothesis
per annum per adult)
It is hypothesised that other variables being constant,
q = is the number of food insecure households and
participant households will have better Household Dietary
α = is the weight attached to the severity of food Diversity Score, Food Consumption Score and Per Capita
insecurity Net Food Available than non-participant households.
In FGT index, Yi ≥ Z mean the specified household is
food secure.
Following [33], using FGT family of indices, the head 4. Results and Discussion
count ratio, food insecurity gap and squared food
The concept of food security is understood and used
insecurity gap were computed to have in-depth insight
differently depending on the context, time frame and
among sample food insecure households.
geographical region in question [5]. The term originated in
Head count ratio estimates the percentage of sample
1974 when the World Food Conference first gave an
households whose available annual per capita food grain is
official definition of food security as “availability at all
falling below the minimal requirement (2.25 quintals of
times of adequate world food supplies of basic foodstuffs
food grain per adult per annum). Giving no weight to the
to sustain a steady expansion of food consumption and to
severity of food insecurity is equivalent to assuming that α
offset fluctuations in production and prices” ([54], pp. 27).
= 0, becomes the ratio between number of food insecure
This definition focused exclusively on the ‘availability’
households with the total sample household size.
aspect of food security. This definition was a reflection of
q
Accordingly, the formula collapses to P(0) = . the problems caused by the worldwide food crisis in the
N 1970s. In 1983, FAO extended this concept to embrace
Giving equal weight to the severity of food insecurity food access by vulnerable people, implying that the
among all food insecure households is equivalent to demand side of food security is at least as important as its
assuming that α = 1 . Summing the numerator gives the supply side. This version suggests that food security
food insecurity gap. The food insecurity gap index should “ensure that all people at all times have both
measures the mean depth of food insecurity among the physical and economic access to the basic food that they
food insecure female headed households. It is the mean need” ([54], pp. 27). In 1986, the World Bank report on
proportion by which the food security status of the food “Poverty and Hunger”, further explained the concept of
insecure households falls below the minimum level of food security. As of this report, food security is defined as
food grain requirement. It is mathematically expressed as: “access of all people at all times to enough food for an
active, healthy life” ([54], pp. 27). This concept also
1 q
Z − Yi
∑(
embraces malnutrition, poverty and food safety issues.
Food Insecurity Gap Index ( P1 ) = )
N Z The most widely accepted definition of food security,
i =1
generated in 1996 at the World Food Summit emphasized
The food severity index (or squared insecurity gap) the multidimensionality of food security. As of this
characterizes the amount of resources that will be required Summit “food security, at the individual, household,
to bring all the food insecure households to the national, regional and global levels is achieved when all
 Journal of Food Security 265

people, at all times, have physical and economic access to Stability of food supplies captures the ability of people
sufficient, safe and nutritious food to meet their dietary to access adequate food at all times. Hence, it refers to
needs and food preferences for an active and healthy life”. both the availability and access dimensions of food
Furthermore, this definition of food security has four security [55].
dimensions; availability, access, utilization and stability. Measuring food security at the national, regional,
On the other hand, “food insecurity exists when people do community, household and individual level is vital for the
not have adequate physical, social and economic access to development of appropriate policy and program options
food as defined above” ([54], pp. 28). [5]. In this study, the focus was on household (a unit of
Food availability reflects the availability of sufficient people living together and headed by a household head)
quantities of food of appropriate quality, either by food security as it is a basic social unit in a society. At the
domestic production or food imports. It is a measure of the household level, a household is labelled as food secure
amount of food that is and will be physically available in a when it has access to the food needed for a healthy life for
population during a certain period of time. Food all of its members (adequate with regard to quality,
availability corresponds to, not just physical but also quantity and safety as well as culturally acceptable) and
energy supplies of food, implying a strong connection when it is not at unwarranted risk of losing such access
with diets [2,55]. Diet quantity available to a household (UN ACC/SCN, 1991) in [56].
can be measured by daily food energy consumption per For this study household food security implied whether
capita or per adult equivalent, and percentage of households the household can produce sufficient food from their own
or people that are food energy-deficient. The second diet production or purchase food grains of the right quality
quantity indicator is the percentage of households in a and quantity in the local market which clearly implies
population group that do not consume sufficient dietary availability of enough food and the capacity of the
energy. If the estimated total energy in the food that the household to acquire it respectfully. Therefore, the survey
household acquires daily is lower than the sum of its result of Household Diet Diversity Score, Food Consumption
members’ daily requirements, the household is classified Score as well as the Net Available Food Grain values of
as food energy-deficient or food insecure and vice versa. the sample households are presented below.
Food access captures whether people have enough
resources to acquire nutritious food for a healthy diet. It 4.1. Households Diet Diversity Score (HDDS)
covers both economic access to food at the household
level and physical access to food in the market, and is The higher the Diet Diversity Score, the more food
measured in terms of income and food prices for the groups were consumed, the more varied the diet and
former, and infrastructure to market outlet for the latter. therefore, the higher the nutritional quality of the diet. For
Consequently, poverty rates and purchasing power parity this study the researcher simply generated HDDS by
indices are strongly linked to food access [55]. A summing up all the food groups reported to be consumed
household’s ability to spend on food is a good indicator of by the sample household members within 24 hours prior
food access at the household level. Household food access to the survey.
is measured through food or nutrient intake at the The effect of commercialization on food consumption
household level reported in adult equivalent [5]. and nutrition is a complex and controversial subject;
Food utilization which emphases on assuring the opponents of commercialization insist that if the resources
biological utilization of food has two elements; one is that are used to produce agricultural export crops were
based on indicators representing under nutrition for used instead to produce food for the local economy,
children under the age of five, while the other reflects food undernourishment can be minimized. On the other hand,
quality and hygiene conditions. Therefore, utilization of advocates argue that exploiting of comparative advantage;
food can be reached through adequate diet, clean water, commercialization could raise farm incomes and improve
sanitation and health care all of which are necessary for nutrition [5]. The empirical study of [40], found that
nutritional well-being [55]. This dimension of food households with irrigation water access were encouraged
security gives due attention for non-food inputs in food to produce cash crops more commanded in the market and
security. The consumption of foods both in quantity and the risk in the marketing of these crops was ultimately
quality that is sufficient to meet energy and nutrient damaging their nutritional position. Irrigator households
requirements is a basic measure of food utilization. allocate relatively more land to more risky perishable cash
According to Frankenberger et al. (1997, pp.1) in [2] “a crops and less land to produce staples (food crops).
person is considered nutrition secure when she/he has a Depicted in Table 1, the survey result of this study
nutritionally adequate diet and the food consumed is revealed that, seven food groups were reported to be
biologically utilized such that adequate performance is consumed with the minimum, maximum, mean and
maintained in growth, resisting or recovering from standard deviation value of 3, 6, 3.42 and 0.71 HDDS
disease, pregnancy, lactation and physical work”. Data on respectively. Moreover, the mean HDDS for participant
nutrient requirements and recommended intakes are vital and non-participant households was computed to be 3.84
to estimate food utilization. Food intake comprises Protein and 3.21 respectively. Comparing irrigation participants
and other nutrients, but energy intake is one of the main from non-participants in terms of mean HDDS, the
parameters and is extremely important in improving food independent sample test was associated with a statistically
utilization. The National Academy of Sciences (1995) has significant effect t (134.759) =6.738, p<0.01 at 99% level
arrived at a figure of 2100 kcal per day for use in food of significance. This significant value of the t-test states
emergency situations, which is based on an assumption of the existence of HDDS discrepancy between sample
light activity [5]. participant and non-participant households. Hence, it can
266 Journal of Food Security

be inferred that compared with non-participants, participant Lack of Dietary Diversity is a severe problem in
households who cultivate their irrigable land enjoy relatively developing world where diets are predominantly starchy
a diversified diet. On the other hand, non-participant staples (cereals high in carbohydrates, but low in nutrients
households who share crop out their farm land and forced and vitamins) with little or no animal products and
to lose half of their farm produces with limited decision vegetables [4,5]. Nationally, 58 and 30% of the Ethiopian
power on preference and quality of crops to be produced, households’ consume four or fewer and three or fewer out
were exposed for low dietary diversity. of seven food groups respectively [58].
Table 1. Household Dietary Diversity Score of the sample households
Households Household Diet Diversity Score t –value
Minimum 3.00
Maximum 6.00
Total
Mean 3.42
Standard deviation 0.71
Minimum 3.00
Maximum 6.00
Participants 6.738***
Mean 3.84
Standard deviation 0.95
Minimum 3.00
Maximum 5.00
Non-participants
Mean 3.21
Standard deviation 0.42

*** refers significant at 0.01 level

Source: computed from field survey, 2016
Figure 2. Sample Households’ Household Dietary Diversity status
(Source: Computed from field survey, 2016)
Besides to calculating the mean HDDS, an attempt has
also been made to indicate which food groups were Setting the HDDS of seven or more food groups, four
reported to be predominantly consumed by the sample to six food groups and less or equal to three food groups
households. Accordingly, as shown in Table 2 all the as threshold points of high, medium and low HDDS
sample households reported to consume cereals and edible respectively, as presented in Figure 2, 68.5% of the sample
oils used for cooking during in the last 24 hours preceding households had low HDDS (nutritionally inadequate dietary
the survey. The second most common food group diversity). The remaining, 31.5% of the sample households
contributed to the daily meals was pulses and legumes had medium HDDS (nutritionally adequate dietary diversity).
(86.5%), followed by rarely consumed but rich in Furthermore, 46.9 and 53.1% of participant households as
micronutrients vegetables (18.0%), dairy products (15.9%), well as 79.5 and 20.5% of non-participant households had
animal protein (13.2%) and fruits (8.7%). Thus, cereals, low and medium measure of HDDS respectively.
oils as well as pulses and legumes were the dominant food As per the Focus Group Discussants’, Key Informant
groups reported to be consumed within the 24 hours recall Interview participants’ opinion and personal observations,
period prior to the survey. In line with this study, the the problem of low HDDS of the sample households was
survey of [57] report reveals that, three food categories related with lack of knowledge and awareness about
(cereals, edible oil and fat, and legumes) dominate the nutrition, cultural preferences, participation in irrigation
Ethiopian meal. Furthermore, a study by [44], revealed which resulted for the practice of mono-cash cropping
that a cereal flour made baked bread was eaten wrapping a (mainly onion), household’s dependence on crop production
sauce from a pulse crop or a vegetable crop or meat alone as a means of employment, food and cash income, and the
or less frequently mix of two or more of these. A mix of prevalence of severe drought in the survey year due to
two or more of these was a common consumption habit in inadequate rain and effects of the El Niño. According to
Ethiopia. the officials of KGVDP, during in the last consecutive
production years the production of onion and tomato in
Table 2. Food Groups reported to be consumed by sample the study area contributed for more than 90 and 8% of the
households in 24 hours recall period
irrigated land respectively. In line with this, [59] also
Total Participants Non-participants found that 89% of the sample farmers in Northern
Food groups
Yes No Yes No Yes No Ethiopia produce onion employing deep well irrigation.
Main Staples 333 0 113 0 220 0 The finding of this study exposed the need for
Pulses and tremendous effort for creating awareness among the
288 45 82 31 206 14
legumes nutritionally inadequate sample households about the need
Dairy products 53 280 39 74 14 206 to diversify their food baskets to ensure balanced diets so
Animal protein 44 289 30 83 14 206 as to improve their nutritional status.
Oils 333 0 113 0 220 0
Vegetables 60 273 37 76 23 197 4.2. Food Consumption Score (FCS)
Fruits 29 304 20 93 9 211
The problem of food insecurity (Nutritional insecurity)
Source: Computed from field survey, 2016 may occur under adequate availability and accessibility
 Journal of Food Security 267

due to problems related to consumption [10]. Hence, revealed that cereals (rice, sorghum, barley and wheat),
food consumption patterns are vital indicators of the edible oil and fats, and legumes characterize the Ethiopian
food security status of households; households who often meal in the last seven days prior the survey.
consume a wide variety of food items or food groups are
more food secure than households who infinitely consume
very alike food stuff [49]. To estimate food consumption
pattern of female headed households, FCS was calculated
at the household level over the period of a week preceding
the survey. The score shows dietary diversity, food frequency
and relative nutritional importance of different food
groups consumed by household members. The calculated
household FCS of this study displayed in Table 3 stated
that the minimum, maximum and mean household FCS
for the sample households was 37.5, 54.0 and 42.74
respectively. Moreover, the calculated mean household
FCS also showed variation among participant (44.89) and
non-participant households (41.64). The independent
sample test was associated with a statistically significant
effect t (124.024) = 7.438, p<0.01 at 1% probability level.
This significant value of the t-test assured the existence of
Figure 3. Percentage distributions of sample households based on FCS
noticeable household FCS variation between participant thresholds (Source: Computed from field survey, 2016)
and non-participant households. Therefore, based on
the survey result, it can be said that compared with Employing the standard and calibrated thresholds of
non-participants, participant households enjoy better FCS. [14] FCS, Figure 3 vividly revelled that a majority, 225
(67.57%) of the sample households were found to be in
Table 3. Food groups reported to be consumed by sample households the borderline food consumption group, while the remaining
in a week period 108 (32.43%) were clustered under the acceptable food
Number of days a given food group was consumed by: consumption category. Moreover, 57.52 and 19.55% of
Food groups Total Participants
Non- participant and non-participant households respectively
participants fall under the acceptable FCS category, while 42.48 and
Mean SD Mean SD Mean SD 80.45% of the sample participants and non-participants
Main Staples 7.00 0.00 7.00 0.00 7.00 0.00 respectively were grouped under the border line FCS.
Vegetable 0.52 0.69 0.84 0.90 0.35 0.48 The computed Household Dietary Diversity and household
Fruits 0.16 0.46 0.34 0.68 0.07 0.25 Food Consumption Scores of this survey are snapshots of
Animal the economic ability of a household to access a variety of
0.14 0.35 0.27 0.44 0.07 0.26 foods during the time of data collection and cannot
protein
Dairy
0.23 0.58 0.58 0.84 0.05 0.22 represent households’ annual food consumption pattern.
products Accordingly, it also appears important to estimate the
Oils 7.00 0.00 7.00 0.00 7.00 0.00 average annual per capita net food availability of the
Sugar 5.06 1.27 5.39 1.33 4.89 1.20 sample households.
Pulses and
6.86 0.35 6.72 0.45 6.93 0.26
legumes
FCS 42.74 3.26 44.89 4.52 41.64 1.45
4.3. Net Available Food (NAF)
t-value 7.438*** Major dependence on rain-fed agriculture makes food
*** refers significant at 0.01 level, SD-Standard Deviation security in Ethiopia to be highly sensitive to climate risks.
Source: Computed from field survey, 2016 For instance, the 2015/16 severe drought had an adverse
effect on food availability and consequently on food
All the sample households reported regular consumption security in the country in general and the study area in
of main staples, edible oils, as well as pulses and legumes particular. This worst drought resulted in the minimal crop
almost on a daily basis over the period of a week prior to harvest, massive livestock death, very low livestock
the survey. Sugar with low nutrient value was also production, staple food price soaring and limited food
reported to be frequently consumed, on an average of 5.06 availability.
days a week. Vegetables (0.52 days per a week), dairy Household Food Balance Model (HFBM) was employed
products (0.23 days per a week), fruits (0.16 days per a to estimate the annual NAF of sample households. The
week) and animal proteins (0.14 days per a week) were the model considered food grain available from households’
least frequently consumed food groups. Therefore, the own production, purchased and gifted grains on one hand
survey result of this study clearly demonstrated the and grain sold, post-harvest loss and seed reserve (for
existence of inadequate diet for a healthy active life as participant households only) on the other hand. To this end,
animal proteins, fruits, dairy products and vegetables with NAF was employed as an indicator of food availability and
moderate and high nutrient value were reported to be household food security situation of the sample households.
consumed least frequently over a period of a week prior to The estimated annual NAF value for each respondent
the survey. Consistent with this survey finding, the household was therefore compared against 225 kg of food
national Food Consumption Score survey of [57] also grain, which is approximately cereal equivalent of the
268 Journal of Food Security

recommended average daily kilocalorie of 2100 for a annual NAF grain volume, 73.31, 26.65 and 0.04% was
healthy adult person. The variation between the available generated from the respondents’ own production, through
grain and the recommended grain was used to determine purchase and obtained as a gift respectively. This in turn
the household food security status of sample households. clearly designated that households’ own production was the
Thus, households whose annual NAF per capita food grain most important dietary source of energy and food availability
was greater than the recommended demand were regarded in the survey period. In subsistence economies, household
as food secure households, while those experiencing a food food security was largely linked to availability of food
grain deficit were labelled as food insecure households. from household’s own production than consumption of
To determine household level grain equivalent requirement, market purchased food. Thus, it can be said that during in
the average household size in Adult Equivalent was the survey year sample households had subsistence economies.
multiplied by the minimum acceptable weighted average
food requirement of 2.25 quintals (225 kg) of food grain Table 5. Sample households’ total annual food available from all
food sources
per adult person per year. The sample households’ average
household size measured in Adult Equivalent was 2.28. Food grain Total Participants Non-participants
Hence, the total annual minimum food requirement for a Own production 2086.5 1266 820.5
sample household was estimated at 5.13 quintals (513kg Grain purchased 758.45 321.7 436.75
of food grain equivalent). On the other hand, the survey Obtained as a gift 1 0 1
result revealed that the total annual NAF grain for the Grand total 2845.95 1587.7 1258.25
sample total 760.09 Adult Equivalent household members
Grain sold 712.5 555.5 157
from all sources for the surveyed year was 1769.47
quintals. This grand volume gave the average annual Reserved for seed 79.385 79.39 0
per capita NAF food grain volume of 2.52 quintals Post-harvest losses 284.595 158.77 125.83
(5.75 quintals per household) which in turn was a bit Total deduced 1076.48 793.66 282.83
greater than the minimum annual food grain requirement. NAF 1769.47 794.05 975.43
Having this computed crude mean annual per capita NAF
Source: Computed from field survey, 2016
grain volume; it can then be inferred that all the sample
households could be regarded as food secure given the
Employing the Adult Equivalent annual per capita
fact that the annual average available per capita food grain
threshold of 225 kg of food grain volume, as depicted in
volume was greater than the recommended food grain
Table 6 out of the sample households, 198 households
volume. Furthermore, the computed average annual Adult
(59.46%) with an average annual per capita Adult
Equivalent per capita NAF grain volume for participant
Equivalent NAF grain volume of 2.98 quintals were found
and non-participant households illustrated in Table 4 was
to be food secure and the remaining 135 households
2.6484 and 2.4556 quintals respectively. Considering this
(40.54%) were food insecure with a mean annual per
estimated annual NAF grain volume, it can be understood
capita Adult Equivalent NAF grain volume of 1.85
that participant and non-participant households had
quintals. The annual NAF survey result also portrayed the
variation in average annual Adult Equivalent per capita
existence of household food security status disparity
NAF grain amount. The independent sample test was
between participant and non-participant households.
associated with a statistically significant effect t (331)
Where 65.49% of participants with an average annual per
=2.265, p<0.05 at 5% probability level. This significant
capita Adult Equivalent NAF grain volume of 3.01
test value assured that there was relatively a significant
quintals and 56.36% of non-participant households with a
mean annual Adult Equivalent per capita NAF grain
mean annual per capita Adult Equivalent NAF grain
volume difference between sample participant and non-
volume of 2.95 quintals were found to be food secure.
participant households. Therefore, based on the t-test result,
Thus, as per the annual threshold of 225 kg of food grain,
it is possible to infer that compared with non-participants,
it can be inferred that participant households were more
participant households enjoy a relatively better NAF grain
food secure than non-participants. However, the Chi-
quantity for the survey year.
square test statistical association result showed no
Table 4. Descriptive statistics summary of per capita annual NAF significant differences between participant and non-
food grain participant households in their food security status at all
NAF/ADE/year (Quintals) probability levels.
Households t-value
Mean SD Table 6. Food Security Status of Sample households as per
Total 2.5210 0.74016 NAF/ADE/year

Participants 2.6484 0.84145 2.265** NAF/ADE/year Chi-

Households (Quintals) square
Non-Participants 2.4556 0.67508 Mean Std. Dev value
** refers significant at 0.05 level, SD-Standard Deviation Total 2.98 0.63
Source: Computed from field survey, 2016. Food Secure Participants 3.01 0.83
Non-participants 2.95 0.47
According to the [60] study, in times of good weather, 2.578
Total 1.85 0.16
75-80% of the annual agricultural produce in Ethiopia is Food
Participants 1.96 0.12
estimated to be consumed at the household level. The insecure
survey result of the Household Food Balance model Non-participants 1.81 0.16
publicized in Table 5 also revealed that out of the total Source: Computed from field survey, 2016
 Journal of Food Security 269

Following the procedures of FGT family of indices that the depth of household food insecurity amongst
explained in the methodology, head count index, food non-participants was fairly higher than their counterparts.
insecurity gap as well as severity of household food The limitation of household food insecurity gap
insecurity was computed. Moreover, as there was a mean measure is however is that it ignores the effect of
annual per capita Adult Equivalent NAF grain quantity inequality (deeper below or slight closer to the threshold)
variation between sample food insecure participant and among the food insecure households. As a result, to
non-participant households, separate value of FGT family identify the most food insecure sample households among
of indices were also calculated. Consequently, the survey the food insecure households, severity of food insecurity
result revealed that the estimated head count ratio or (Squared food insecurity gap index) was computed by
prevalence of food insecurity for the total food insecure assigning a higher weight (α = 2). Accordingly, as shown
sample households was 0.4054; of which 0.3451 and in Table 7, 0.0329, 0.0052 and 0.0277 was the calculated
0.4364 was for the food insecure participant and non- severity of household food insecurity for the sample total
participant households respectively. This implies that food insecure households, food insecure participant and
40.54, 34.51 and 43.64% of the sample households, non-participant households respectively. These figures
participant and non-participant households respectively further implied that the severity of food insecurity among
were not able to meet the stipulated minimum requirement. the sample food insecure households, food insecure
Hence, it can be deduced that the prevalence of food participants and food insecure non-participant households
insecurity problem was fairly higher for non-participant was 3.29, 0.52 and 2.77% respectively. Moreover, the
households than participants. However, the Chi-square test survey result also showed the existence of discrepancy
statistical association result showed no significant differences among the food insecure households; where, participant
between participant and non-participant households in households’ severity of food insecurity was estimated to
their food security status at all probability levels. be 2.25% lower than non-participant households’ severity
of food insecurity. The independent sample test was
4.4. Extent of Household Food Insecurity associated with a statistically significant effect t (116.462)
=6.593, p<0.01 at 99% level of significance. This
To have an in-depth insight on how these food insecure significant value of the t- test declared that the severity of
sample households were far below the recommended food household food insecurity amongst participants was fairly
grain volume; household food insecurity gap was also lower than the severity of household food insecurity
computed. The rationale behind the calculation of food amongst non-participants.
insecurity gap was to estimate the resources required to
alleviate the problem of household food insecurity through Table 7. Food insecurity indices of the sample food insecure
households
proper targeting. Generally, as the food insecurity gap
value gets larger, more resources are needed to lift them Household Food Food insecure Households (Ratio)
Test-value
out of the food insecurity situation. Publicized in Table 7, Insecurity indices Total Participants Non-participants
the calculated food insecurity gap value for the total food Incidence of food 2.578
0.4054 0.3451 0.4364
insecure sample households, food insecure participant and insecurity (α = 0) (Chi-square)
non-participant households was found to be 0.0727, Depth of food 5.793***
0.0727 0.0153 0.0561
0.0153 and 0.0561 respectively. These estimated values insecurity (α = 1) (t-value)
further indicate that the problem of household food Severity of food 6.593***
0.0329 0.0052 0.0277
insecurity was not as such severe as the value tends insecurity (α = 2) (t-value)
towards zero; the degree of household food insecurity ***refers significant at 0.01 level
diminishes and requires less resource to lift the sample Source: Computed from field survey, 2016.
food insecure households out of food insecurity trap.
Accordingly, if the Wereda officials mobilize and Though minimal in level, the result of this survey
distribute resources that can meet and sustain 7.27, 1.53 testified the prevalence of severe food shortage problem
and 5.61% of the food grain amount requirement of the amongst the sample irrigation beneficiary households. In
sample total food insecure households, food insecure line with this empirical finding, as per the Focus Group
participant and non-participant households respectively, Discussants’, Key Informant Interview participants’
theoretically the problem of household food insecurity can opinion and personal observations, because of their lower
be eliminated. In other words, 0.1636, 0.0344 and 0.1262 poverty level and high level of food production and stock
quintals of annual per capita food grain was required to available, sample households were not beneficiaries of any
bring all the sample total food insecure, participant and food aid programmes in the survey year. As a result,
non-participant households to the rank of food secure participation in non-farm income generating activities,
households respectively. As the estimated average food remittance from abroad, sell-off ruminant animals, and
grain amount required to bring non-participant households consumption of less expensive grain foods were
out of food insecurity was estimated to be 4.08% higher mentioned as the common coping strategies employed by
than participant households, it can be said that the depth of the sample transitory food insecure households to mitigate
non-participant households’ food insecurity was slightly their food shortage problem. As food insecurity is a
higher than participant households. The independent seasonal phenomenon, pronounced food shortage problem
sample test was associated with a statistically significant was reported to happen between the wet agricultural
effect t (90.903) =5.793, p<0.01 at 99% level of seasons (July) to the beginning of harvesting season
significance. This significant value of the t- test confirmed (September) of the survey year. The 2015/16 integrated
270 Journal of Food Security

surveys on agriculture Ethiopia Socioeconomic Survey of sample food secure and food insecure households had a
[57], also exposes that in all parts of Ethiopia, June, July, computed 3.64 and 3.11 mean HDDS respectively. The
August and September were flagged as particularly independent sample test was associated with a statistically
pronounced months of food insecurity. significant effect t (307.043) = 7.914, p<0.01 at 1%
probability level. This significant value of the t- test
4.5. Correlation between HDDS, FCS, and ascertained that sample food secure households enjoyed a
relatively diversified meal than their counterparts.
NAF
The computed mean household FCS value displayed in
Various indicators of household food security had been Table 9 also showed variations among sample food secure
employed by different organizations and researchers as (43.78) and food insecure households (41.21). The
one indicator at a time could not wholly explain household independent sample test was associated with a statistically
food security. Likewise, it is also important to evaluate the significant effect t (329.291) = 8.317, p<0.01 at 1%
comparative performances of these indicators. To examine probability level. This significant value of the t- test
the validity of alternative measures of food security, discovered that sample food secure household had a
measures of correlation such as Pearson or Spearman relatively higher FCS than food insecure households.
correlation coefficients can be employed [33]. For this
Table 9. Relationships between household food security status,
study, I employed Pearson Correlation Coefficient to HDDS and FCS
evaluate the consistency of household food security
indicators used for this study. Accordingly, comparison Indicators Food Security Status Mean SD t-value
analysis between indicators was made at cut-off points of Food insecure 3.11 0.40
HDDS 7.91***
less or equal to three food groups for the HDDS, less than Food secure 3.64 0.80
42 (poor and borderline food consumption) for FCS and Food insecure 41.21 2.19
FCS 8.32***
less than 225 kg of annual food grain for Adult Equivalent Food secure 43.79 3.46
NAF value. The strength of the relationship, the direction
*** refers significant at 0.01 level, SD-Standard deviation
and the level of significance is presented below. Source: Computed from field survey, 2016
Table 8. Correlation between HDDS, FCS and NAF

Indicators
Households
below the cut-
Pearson correlation value 5. Conclusion
HDDS HDDS FCS and
off point (%) and FCS and NAF NAF
Attaining food security for all people at all times
HDDS 68.5
remains a major challenge for many developing countries
FCS 67.57 0.83*** 0.40*** 0.40*** including Ethiopia. The current government of Ethiopia
NAF 40.54 has been expanding and prioritize modern small scale
*** refers significant at 0.01 level irrigation projects as a means for achieving food security
Source: Computed from field survey, 2016 and reducing poverty at household level. Hence, this
study has attempted to evaluate the household food
The Pearson correlation matrix of this survey security status of peri-urban modern small scale irrigation
demonstrated in Table 9 shows that HDDS is significantly project beneficiary households. To achieve this objective,
correlated (0.828) with FCS, whereas NAF had moderate a combined household food security measures; Household
correlation (0.402) with FCS as well as HDDS. Dietary Diversity Score (HDDS), Food Consumption
Accordingly, the choice between the above mentioned Score (FCS) and Per Capita Net Food Availability (NAF)
indicators of household food security can be done based were employed.
on purpose, time, and resource availability. These The computed Household Dietary Diversity Score
correlation results on the performance of specific (HDDS), Food Consumption Score (FCS) and Per Capita
indicators are found to be consistent with the previous Net Food Availability (NAF) measures of household food
empirical findings. [37] In their multi-country analysis of security exposed household’s low household dietary
data from 10 countries found that correlation between diversity, food consumption score as well as food security
number of food groups and energy consumption ranged status of the studied households. Furthermore, the
from 0.085 to 0.329. [61] In their comparative evaluation comparative analysis also illustrated that the prevalence of
of dietary indicators used in food consumption assessment household food insecurity, low household dietary diversity
in Mozambique also found the correlation between food and food consumption was far more noticeable among
items and energy consumption to be 0.243. Furthermore, a non-participant households than participants. However,
comparison of the Household Dietary Diversity and Food despite their significance difference in their household
Consumption Scores by [62] in Burkina Faso, Lao People’s dietary diversity and food consumption scores, the Chi-
Democratic Republic and northern Uganda showed significant square test statistical association result showed no
correlation between these indicators and concluded that significant differences between participant and non-
the choice of indicator of food security assessment and participant households in their food security status at all
surveillance will vary depending on user’s need. probability levels. On the other hand, the FGT family of
Employing the t-test of association, an attempt was also indices; food insecurity gap as well as severity of
made to identify the linkage between households’ food household food insecurity measure as well as the
security status and computed mean HDDS as well as FCS. independent statistical t-test values of this survey
Accordingly, as illustrated in Table 9, it was found that indicated that the depth and severity of household food
 Journal of Food Security 271

insecurity were relatively higher for non-participant [4] FAO (2008). Water for the rural poor; Interventions for
households than participant households improving livelihoods in Sub-Saharan Africa. In; Faures, J. M. and
Santini, G. (eds). Food and Agriculture Organization, Rome.
Based on the empirical evidences of this survey it can [5] Babu S. and Sanyal P. (2009). Food security, poverty and nutrition
be inferred that, the impact of peri-urban modern small policy analysis; statistical methods and applications, Elsevier Inc.
scale irrigation projects on achieving household food security [6] Rosen S, Meade B, Murray A. (2015). International food security
is minimal. This finding further questions the performance assessment, 2015-2025. GFA-26, U.S. Department of Agriculture,
of irrigation schemes as a means of sustainable poverty Economic Research Service, June 2015.
[7] FAO, IFAD and WFP (2015). The state of Food insecurity in the
reduction and food security attainment. world 2015. Meeting the 2015 international hunger targets: taking
The computed Pearson correlation result showed that stock of uneven progress. Rome, FAO
Household Dietary Diversity Score is significantly [8] Degefa Tolossa (2005). Rural livelihoods, poverty and food
correlated with Food Consumption Score, whereas Net insecurity in Ethiopia, A case study at Erenssa and Garbi
communities in Oromiya Zone, Amhara National Regional State.
Available Food had moderate correlation with Food
Unpublished PhD Thesis, Department of Geography, Norwegian
Consumption Score as well as Household Dietary University of Science and Technology, NTNU, Trondheim.
Diversity Score. Moreover, the t-test statistical association [9] UNDP (2009). Human development report 2009, New York:
also assured that sample food secure households had Palgrave Macmillan.
relatively better Household Dietary Diversity Score and [10] Guyu Ferede (2015). Household Vulnerability to Green Famine;
Component Based Analysis of Indicators in Belo-jiganfoy District,
Food Consumption Score than food insecure households. Benishangul-Gumuz Region, Ethiopia. Applied Science Reports,
Cognizant to this, the finding of this study also supports 9(3), 139-156.
and suggests the use of more than one indicators of [11] UNICEF (2017). Ethiopian Humanitarian Situation report.
household food security while assessing the food security [12] UNDP (2016). Human Development Report 2016. New York,
status of households. USA.
The per capita NAF grain analysis of a single year [13] UNDESA (2015). World Urbanization prospects: The 2014
Revision, (ST/ESA/SER.A/366), United Nations, Department of
intake for a household tells more regarding seasonal Economic and Social Affairs, Population Division, New York,
household food shortage than showing chronic food USA.
insecurity in terms of poverty and its persistence over [14] WFP (2009). Summary of food security and Vulnerability in
years and very unfortunately, in the survey year, the area selected urban centres of Ethiopia. Addis Ababa, Ethiopia.
was suffering a severe food shortage crisis due to too [15] Samson Kassahun and Alok Tiwari (2012). Urban Development in
Ethiopia: Challenges and Policy Responses. The IUP journal of
little rain and effects of the El Niño. Furthermore, the Governance and public policy, Vol. 7 no.1.
Chi-square test statistical association result showed no [16] MoFED (2012). Ethiopia’s progress towards eradicating poverty:
significant differences between participant and non-participant an interim report on poverty analysis study (2010/11). Development
households in their food security status at all probability planning and research directorate, FDRE, Addis Ababa.
[17] Degefa Tolossa (2010). Some realities of the urban poor and their
levels. This in turn inquires to further identify determinants food security situations: a case study of Berta Gibi and Gemechu
of household food security and assess the impact of Safar in the city of Addis Ababa, Ethiopia. Environment and
modern small scale irrigation projects on the beneficiary Urbanization, 22(1), 179-198, IIED.
households’ livelihood assets as commercialized agriculture [18] Dessalegn Rahmato (2013). Food security and safety Nets:
Assessment and challenges, in Dessalegn Rahmato, Alula
has complex linkage with food security and livelihood.
Pankhurst and Jan-Gerrit Van Uffelen (eds), Food Security, Safety
This parcel of information is important for policy Nets and Social Protection in Ethiopia. Addis Ababa: FSS.
makers in the intervention areas of food insecurity and [19] MoFED (2002). Poverty profile of Ethiopia, Addis Ababa,
poverty reduction. Ethiopia.
[20] FAO (2001). Urban And Peri-Urban Agriculture; A briefing guide
for the successful implementation of Urban and Peri-urban
Agriculture in Developing Countries and Countries of Transition,
Acknowledgements Rome, Italy.
[21] Veenhuizen R. V. and Danso G. (2007). Profitability and
Research for this dissertation was supported in part by Sustainability of Urban and Peri- urban agriculture, Agricultural
Open Society Foundations (OSF). Moreover, the author Management, marketing and finance occasional paper, FAO,
Rome.
also gratefully acknowledges the financial support from [22] Awulachew, S. B., Merrey, D.J., Kamara, A. B., Van Koopen, B.,
Wollo University (WU) and thanks the interviewed Penning de Vries, F and Boelee, E., and Makombe, G. (2005).
households. The opinions expressed herein are my own Experiences and Opportunities for Promoting Small Scale/Micro
and do not necessarily express the views of OSF and WU. Irrigation and Rainwater Harvesting For Food Security in
Ethiopia. Colombo, Sri Lanka: IWMI. V.86p. (Working paper 98).
[23] Awlachew, S. B., Erkosa, T. and Namara, R. E. (2010). Irrigation
Potential in Ethiopia: Constraints and Opportunities for
References Enhancing the System, International Water Management Institute.
[24] Nugusse Zeweld (2013). Food Security through Small Scale
[1] Maxwell D., Jennifer C. and Bapu V. (2013.) How Do Different Irrigation; Case Study from Northern Ethiopia, Masters
Indicators of Household Food Security Compare? Empirical unpublished thesis Ghent University, Belgium.
Evidence from Tigray, Feinstein International Centre, Tufts [25] Solomon Cherre (2006). Irrigation policies, strategies and
University; Medford, USA. institutional support conditions in Ethiopia in Awulachew S. B.,
[2] Pangaribowo E. H., Gerber N. and Torero M. (2013). Food and Menker M., Abesha D., Atnafe T., Wondimkun Y. (eds) (2006).
Nutrition Security Indicators; A Review, FOODSECURE working Best Practices and Technologies for Small Scale Agricultural
paper 04. Water Management in Ethiopia. Proceeding of a
MoARD/MoWR/USAID/IWMI symposium and exhibition held at
[3] Meskerem Ali and Degefa Tolossa (2015). Household Food
Ghion Hotel, Addis Ababa, Ethiopia 7-9 March 2006. Colombo,
security status and its determinants in Girar Jarso Woreda, North
Sri Lanka; International Water Management Institute, .190 pp.
Shewa Zone of Oromia Region, Ethiopia. Journal of Sustainable
Development in Africa, 17 (7), 118-137, Clarion University of [26] KGVDP (2015). Modern small scale Irrigation project beneficiary
Pennsylvania, Clarion, Pennsylvania. households up to 2015, unpublished report, Kobo, Ethiopia.
272 Journal of Food Security

[27] Goitom Sisay (2009). Rural-urban marketing linkages and its Northwest Ethiopia, feed the future- Research award, Bahir Dar,
implication for integrated development, the case of Kobo town Ethiopia.
and its surrounding Kebeles, unpublished M.A thesis, Department [47] Getnet Mekuria, Yalewsew Wubneh and Tilahun Tewabe (2017).
of Geography and Environmental Studies, Addis Ababa Household Dietary Diversity and Associated Factors among
University, Ethiopia residents of Finote Selam town, North West Ethiopia: a cross
[28] CSA (2010). The 2007 population and Housing census of Ethiopia: sectional study, BMC Nutrition (2017) 3; 28.
results from country level statistical report. FDRE, Population [48] WFP (2008). Food consumption analysis Calculation and use of
Census Commission the food consumption score in food security analysis,
[29] Kothari C.R. (2004). Research methodology; methods and Vulnerability Analysis and Mapping Branch, Rome, Italy.
techniques, Second revised edition, new age international (P) [49] Ephrem Tegegne (2015). Livelihoods and Food Security In The
limited publishers. Small Urban Centres Of Ethiopia; The Case Of Durame,
[30] Creswell J. (2012). Educational research; planning, conducting, Wolenchiti And Debre Sina Towns, a PhD unpublished thesis,
and evaluating quantitative and qualitative research, fourth edition, Department of Geography and Environmental Studies, Addis
Pearson. Ababa University, Addis Ababa, Ethiopia.
[31] Babbie E. (1990). Survey Research Methods. Belmont; Wadsworth [50] Messay Mulugeta (2010). Food Security Attainment Role of
Publishing. Urban Agriculture; a Case Study from Adama Town, Central
[32] Creative Research Systems (CRS) (2012). Sample Size Formulas Ethiopia. Journal of Sustainable Development in Africa Vol 12,
for our Sample Size Calculator No.3, pp. 223-249, Clarion University of Pennsylvania, Clarion,
http;//www.surveysystem.com/sample-size-formula.htm accessed Pennsylvania.
on January 25, 2015. [51] Getinet Kebede (2011). Linking Small-Scale Irrigation and
[33] Hoddinott J. (ed) (2001). Methods for Rural Development Projects. Household Food Security in Drought Prone Area of North East
Food security in practice, Vol 1, IFRI, Washington D.C. Ethiopia; A Case Study of Alawuha Irrigation Scheme in Gubalafto
[34] Wiesmann D., J. Hoddinott, N.-L. Aberman and M. Ruel. (2006). Wereda, North Wollo Zone, unpublished M.Sc thesis, Institute Of
Review and Validation of Dietary Diversity, food frequency and Development Studies, Addis Ababa University, Ethiopia.
other proxy indicators of household Food security, report [52] Gutu Tesso (2015). Why the urban households remain Food
submitted to World Food Programme, Rome, by IFPRI. insecure in Developing countries? Empirical Evidence from
[35] Maxwell D. (2008). The coping strategy Index: a tool for rapid Nekemte Town of Ethiopia, International Journal of Managerial
Measurement of Household Foods and the impact of food aid Studies and Research (IJMSR), 3(10), 108-117.
programs in Humanitarian Agencies, Field method manual, 2nd , [53] Foster J., J. Greer and E. Thorbecke (1984). A class of
TANGO, USAID, WFP and CARE. Decomposable Poverty Measures. Econometrica, Vol.52 ,No.3
[36] CSA (2005). The 2005 National statistics, Addis Ababa, Ethiopia. (May,1984) pp 761-766.
[37] Hoddinott J. and Yohannes Y. (2002). Dietary diversity as a food [54] FAO (2003). Trade reforms and food security – conceptualizing
security indicator. Food consumption and Nutrition division the linkages. FAO, Rome.
discussion paper. Washington D.C IFPR. [55] FAO (2013). The state of food insecurity in the world: the multiple
[38] Arimond M. and M. T. Ruel (2004). Dietary Diversity, dietary dimensions of food security. FAO, Rome
quality and child Nutritional status: Evidence from eleven http://www.fao.org/docrep/016/i3027e/i3027e.pdf accessed on
demographic and health surveys. Washington D.C. Food and 24/01/2015.
Nutrition Technical Assistance (FANTA) project, Academy for [56] Ruel M., James L. Garrett, Saul S.Morris, Daniel Maxwell, Arne
Educational Development (AED). Oshaug, Patrice Engle, Purnima Menon, Alison Slack, and
[39] Hatloy A., Halland J., Diarra M., and Oshaug A. (2000). Food Lawrence Haddad. (1998). urban challenges to Food and
Variety, Socio-Economic Status and Nutritional Status in Urban Nutrition security: a review of Food security, Health and
and Rural Areas In Koutiala (Mali). Public. Health Nutr.3:57-65. Caregiving in the cities. FCND Discussion paper No.51, Food
[40] Degeye Goshu, Belay Kassa and Mengistu Ketema (2013). Consumption and Nutrition Division, International Food Policy
Measuring Diet Quantity and Quality Dimensions of Food research institute (IFPRI), Washington D.C.
Security in Rural Ethiopia, Journal of Development and [57] CSA and World Bank (2017). LSMS-Integrated Survey on
Agricultural Economics, 5 (5), 174-185. Agriculture Ethiopia Socio Economic Survey (ESS) 2015/16,
[41] WHO/FAO (1998). Preparation and use of Food based dietary Addis Ababa, Ethiopia.
Guidelines. Report of a joint WHO/FAO consultation. WHO/FAO [58] CSA and WFP (2014). Comprehensive Food security and
technical report series No.880 Geneva. Vulnerability Analysis (CFSVA), Ethiopia.
[42] Ruel M. (2006). Operationalizing Dietary Diversity: A review of [59] Haile Tesfay (2008). Impact of Irrigation Development on Poverty
Measurement Issues and Research Priorities. Food Consumption Reduction in Northern Ethiopia, a PhD unpublished Thesis,
and Nutrition Division, International Food Policy research Department of Food Business and Development, National
institute (IFPRI), Washington D.C. University of Ireland, Cork.
[43] FAO (2011). Women in Agriculture; Closing the gender gap for [60] World Bank (2006). Ethiopia: Managing Water Resources to
development. Rome, Italy; FAO and Agriculture Organization of maximize Sustainable growth, A World Bank Water Resource
the United Nations. Assistance Strategy for Ethiopia, Washington, D.C.
[44] Kassie G.T., Asfaw W., Zeleke G., and Scott D. (2008). The [61] Rose D., S. Chotard, L. Oliveira, N. Mock and M. Libombo
Nexus of Migration, HIV/AIDS and Food Security in Ethiopia, (2008). Comparative evaluation of dietary indicators used in food
Global Mountain Programme of the CGIAR, Addis Ababa, Ethiopia. consumption assessment of at risk populations, Food and
[45] Arega Bazezew (2012). Determining Food security Indicators at Nutrition Bulletin, vol, 29, No 2, pp 113-122.
Household level in Drought prone areas of the Amhara Region of [62] Kennedy G., Berardo A., Papavero C., Horjus P., Ballard T., Dop
Ethiopia: the case of Lay Gaint District, Ethiopian Journal of MC., Delbaere J. and Brouwer ID. (2010). Proxy measures of
Environmental Studies and Management (EJESM), Vol.5 No.4. household food consumption for food security assessment and
[46] Girma Nega, Melkie Endris, Degnet Teferi, Amanuel Nana and surveillance: comparison of the household dietary diversity and
Rigbe W/Michael (2015). Dietary Diversity and Associated food consumption scores, Public Health Nutrition: 13(12), pp.
factors among Rural Households in South Gonder Zone, 2010-2018.