Measuring Livelihood Impacts:

A Review of Livelihoods Indicators

Livelihood Monitoring Unit (LMU)
Rural Livelihoods Program
CARE Bangladesh

March 2004

Research Tool

Prepared by TANGO International, Inc.

 PREFACE

The Livelihoods Monitoring Unit (LMU) of the CARE Bangladesh Rural Livelihoods
Programme was designed as a lesson-learning intervention to develop a system for
monitoring change in the livelihoods and entitlements of the rural poor in the Northwest and
Southeast regions of Bangladesh.

In order to monitor changes in livelihoods it is important to develop approaches to identify

and monitor key livelihood indicators that enable us to understand how our programs impact
upon the lives of the rural poor. These indicators need to represent the key components of
livelihoods for the poor, measurement of which will allow organizations to effectively
ascertain the program impact.

A workshop was organized in March 2004 by CARE-Rural Livelihood Program to review and
assess the "competing Livelihood indicators" already in use by CARE, other donors and
local NGOs to monitor changes in livelihoods. This report is the outcome of review and
assessment of the current livelihood indicators used with in CARE and also by other external
organizations. The report gives details of the process involved in building a consensus and
generating a set of best proxies as " Livelihood indicators". It also describes a finite and
comprehensive set of indicators (26 livelihood indicators, organized around 9 livelihood
outcome themes) that meet a number of standard criteria, such as, validity, measurability,
relevance and flexibility. Please note that these set of indicators are a refined versions and
the LMU is going to test these during the Northwest Livelihoods Survey 2004.

Sarah Gillingham Md. Mehrul Islam

Livelihoods Adviser, RLP Livelihood Coordinator, RLP

N.B.: Comments are welcome and should be sent to Sarah Gillingham

sarah@carebangladesh.org or Md. Mehrul Islam mehrul@carebangladesh.org
 Table of Contents

1.0 Introduction.................................................................................................................... 1
2.0 The Household Livelihoods Approach........................................................................... 1
3.0 Pre-Workshop Preparation ............................................................................................ 2
4.0 Workshop Organization and Process ............................................................................ 2
5.0 Workshop Outcomes ..................................................................................................... 3
5.1. Food Security ................................................................................................... 3
5.2. Nutritional Security ........................................................................................... 4
5.3. Economic Security............................................................................................ 6
5.4. Shelter/Water and Sanitation Security.............................................................. 7
5.5. Health Security ................................................................................................. 8
5.6. Educational Security......................................................................................... 9
5.7. Gender Status .................................................................................................. 9
5.8 Community Participation................................................................................. 11
5.9 Access to Institutions...................................................................................... 11
6.0 Concluding Observations ............................................................................................ 12

List of Tables/ Figures

Table 1. Food Security Indicators........................................................................................ 5

Table 2. Indicators of Nutritional Security............................................................................ 5
Table 3. Indicators of Economic Security ............................................................................ 7
Table 4. Shelter and WATSAN Security.............................................................................. 8
Table 5. Health Security Indicators ..................................................................................... 8
Table 6. Education Security Indicators................................................................................ 9
Table 7. Gender Status Indicators..................................................................................... 10
Table 8. Community Participation Indicators..................................................................... 11
Table 9. Access to Institutions Indicators .......................................................................... 12
Figure 1. Household Livelihood Security: A Framework for Analysis ..….………………..…13
Annex-A: Comparison of Indicators Used by CARE-Bangladesh and Other Organisation..14
Annex-B: Participant’s List .................... ..…...…………………………………………...……..18

Measuring Livelihood Impacts: A Review of Livelihoods Indicators ii

 Summary Report

1.0 Introduction

The Livelihoods Monitoring Project (LMP) is a component part of the larger

CARE/Bangladesh Rural Livelihood Program (RLP) which also includes the GO-INTERFISH
project in northwest Bangladesh and the SHABGE project in both the northwest and
southeast regions of the country. The purpose of LMP is to assess the impacts on rural
livelihoods of these two development projects and more broadly to monitor the well-being of
rural livelihoods over time in these two targeted regions. Rather than the project-specific
monitoring and evaluation of outputs (which the RLP projects have individually installed),
LMP has the broader mission of measuring impact-level changes in livelihoods. In effect, it
provides systematic documentation of livelihood development in the targeted regions and
offers important insights on the determinants of livelihood change. As its central
methodological framework, LMP has designed for each region a sequence of two baseline
surveys spaced over time in order to capture the impacts of RLP project interventions on
local livelihoods. The first baselines were completed in 2002-03, and the second set is under
preparation. In addition, LMP has also carried out in-depth thematic studies on debt and
migration—seen as livelihood strategies particularly relevant to development programming—
and other thematic studies will be considered.

A major challenge to LMP effort is to identify and operationalize the appropriate set of robust
indicators that can capture changes in livelihood well-being, especially those changes
associated with project interventions. While the household livelihood system approach has
been widely embraced by the donor community in Bangladesh, there remains a lack of
general consensus over which specific livelihood indicators—consistent with the livelihoods
approach—should be effectively and efficiently monitored through time. To move toward
that consensus, the LMP project staff organized an livelihood indicators workshop which
convened representatives of different CARE/Bangladesh programs as well as
representatives of other donor and GOB agencies. The intent of the workshop was to
review and assess the “competing” indicators already in use, then agree upon a
comprehensive but finite set of indicators that would effectively measure change in livelihood
well-being, and finally devise a strategy for measuring the selected indicators. This set of
indicators are thus designed to serve as a systematic set of sensors that would periodically
assess the progress of livelihood systems in the regions where CARE/Bangladesh has
targeted its project interventions. At the same time, it was intended that these indicators
would serve the entire development community in Bangladesh and provide a consistent
basis for livelihood assessment across the country.

2.0 The Household Livelihoods Approach

The household livelihoods approach is a conceptual model of household decision making

within wider contexts of constraints and opportunities. The core analytical unit is the
household, membership in which is defined in terms of regular roles, rights, and
responsibilities distributed across gender and age. As demonstrated in Figure 1, households
mobilize access to packages of assets, often depicted as different types of capital (human,
economic, physical, etc.). How these assets are acquired is highly variable ranging from
such mechanisms as biological reproduction, participation in markets, inheritance practices,
and in other public and private institutions. Since households exist within physical and
institutional environments (or Context in the Figure), which are themselves subject to change
(e.g., the occurrence of a stress event such as a flood or conflict), household access to
assets can vary in time. Households also make decisions about the allocation of assets
between consumption, production and income generation, and exchange activities. In the

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 1

case of rural households, decisions are made with regard to crop choice and technology, the
division of household labor between production and income generation (including out-
migration), the enhancement of human capital through education, and the distribution of both
product and income. In its entirety, the set of mobilization and allocation decisions result in
outcomes, which constitute categories of well-being, such as food security, nutritional
security, economic security, and so on. The strength of the livelihood approach lies in its
comprehensiveness. Well-being, as determined by livelihood outcomes, encompasses not
only the standard food and income indicators, but also those related to health, education,
shelter, access to water and hygiene, institutional support, participation, justice, and gender
equity. In this holistic perspective, the process of evaluating and monitoring household
livelihoods requires an integrated focus on all these well-being dimensions and the complex
relationships among them.

Consistent with this approach, then, the strategy of the workshop was to identify those
livelihood indicators which, as a group, would allow development agencies and other
stakeholders to monitor complex changes in livelihoods and to understand the processes
(i.e., the decisions and their motives) that have produced such change.

3.0 Pre-Workshop Preparation

In preparation for the indicator workshop, the LMP team examined several sources to
compile a large set of outcome indicators currently in use in Bangladesh. These indicators
were drawn from those that are employed within the various development programs of
CARE/Bangladesh, from those used in Bangladesh by other donors, including national
NGOs, and from the set or recommended indicators generated by past consultancies with
the RLP program. Examples of these indicators are provided in Annex A. The indicators
from each source were organized into the livelihood outcome categories presented in Figure
1, thus providing a framework for comparison. The team looked for commonalities among
these indicators and assessed them in terms of their “robustness” (i.e., how adequately they
reflected actual household well-being) and their feasibility with regard to the constraints of
data collection. This process identified a set of “candidate” indicators that were presented to
the workshop participants.

4.0 Workshop Organization and Process

The workshop was held on March 21, 2004 and was attended by CARE/Bangladesh
program staff, other NGO staff and management, and GOB representatives (see Annex B
for a list of participants). The stated goal of the workshop was to achieve a group consensus
in identifying a finite, manageable list of robust livelihood indicators and to develop for each
indicator a strategy for measurement. At the beginning of the workshop, the group discussed
the eligibility requirements that would define a viable indicator. It was agreed that the final
set of indicators must meet the following set of standards:

1. Validity: The indicator must be a valid representation of a change in household well-

being, as determined by the categories of livelihood outcomes (i.e., food security,
economic security, shelter, etc.).
2. Measurability: The indicator must be amenable to measurement within the
constraints of time and resources.
3. Relevance: The indicator must be relevant to program objectives.
4. Flexibility: The indicator must be flexible enough to be applied to different contexts
and livelihood systems.

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 2

It was further agreed that the set of indicators—taken as a whole—should be comprehensive
in the sense that they holistically address all the different dimensions (i.e., outcome
categories) of the household livelihood approach.

During the initial sessions of the workshop, the different CARE/Bangladesh indicators,
organized by outcome category, were presented to the participants, then the indicators used
by other donor organizations were presented. Following the presentation, the participants
were invited to brainstorm other indicators that would best capture change in the livelihood
well-being. All suggested indicators were recorded according to their respective livelihood
outcome category. In a following session the participants split into break-out groups
representing each of the livelihood categories in order to review and evaluate the potential
indicators in light of the standards established above. Each group presented their
recommendations in a plenary session, including the preferred set of indicators and the
respective strategies for measurement. A final plenary session sought to reduce the total
number of candidate indicators and to achieve the desired consensus.

5.0 Workshop Outcomes

The consensus-building process engaged during the workshop identified a set of 26

livelihood indicators organized around 9 livelihood outcome “themes.” The workshop
participants recognized and focused on the fact that the objectives of a given development
program and the indicators of successful performance of the program are often indirectly
related. For example, a desirable food security impact of “adequate quantity of food
consumed” is not directly measurable. Rather it must be approached indirectly, or
“operationalized” by selecting components of the reality (i.e. food security) that can in fact be
measured and which do accurately depict food security. Since, of course, food security or
even “adequate amount of food” cannot be directly measured, it is necessary to identify the
operational variables that can measure it. In some cases, it is necessary to measure a
behavior that is thought to strongly correlate with the specific component, in other words, a
proxy variable. The use of vegetable oil, for example, is considered to be a strong proxy for
protein consumption, since the oil is mostly used to fry fish or meat. Consequently, the
following workshop outcomes are presented according to each livelihood theme (or outcome
in Figure 1), with a discussion of the several components of each theme and the operational
and/or proxy variables that have been chosen to measure each component.

5.1. Food Security

One widely-accepted conceptual framework defines food security in terms of food

availability, access, and utilization (Riley et al. 1999). From a household livelihood
perspective, food security is thus a function of whether food is available either on-farm or in
the market, whether households have access to the food, and whether patterns of food
utilization, including intra-household distribution, are such that the nutritional needs of all
household members are met. In essence, a livelihood analysis of food security at the impact
level assesses the quantity and quality of food available to households throughout the year
and the distribution food among all household members. Often, it is the case that the food
security is effectively measured by a household’s capacity to cope with stress periods—
either seasonal or interannual. Thus, the indicators that capture these components of food
security (summarized in Table 1) are as follows:

1) Duration of the lean period: In Bangladesh, as elsewhere, many rural households are
confronted with a regular seasonal period of stress. The occurrence of seasonal stress
is usually related to the cropping cycles (food stocks from previous harvest are low, but
current crop is not yet ready) or to seasonal fluctuations in the rural employment market.
In Bangladesh the lean period is a commonly-recognized component phenomenon of

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 3

 food insecurity, and it is measured in terms of the specific months of the year that a
household is forced to alter the quantity and quality of food that it consumes. More food
secure households have a reduced lean period or none at all, while severely food
insecure households may constrict consumption all year long. This indicator is
measured directly in a questionnaire format, and assumes that the information reflects a
normal year.

2) Quantity of food consumed per day during the lean period: This indicator reinforces
the first one and provides a more in-depth assessment of household food security. It
measures the quantity of meals consumed per day during the lean period and identifies
the months in which the amount of food prepared per meal is reduced. While the first
indicator measures the number of “problem-months”, this indicator provides a more
refined measure of severity of food insecurity. It is measured directly in a questionnaire
format.

3) Share of household budget spent on food items: It has been well-documented in the
development literature that as a households increases its income, a smaller share of the
household budget is spent on food items. Following this logic, the household budget
share allocated to food becomes a proxy variable of food security. In other words, the
household that spends less of total annual income on food is considered to be more food
secure relative to one that spends more. This indicator is measured in a standard
questionnaire format that asks for estimated percentage allocations of household
expenditures within a small set of budget categories (food, health, education, production
inputs, shelter, etc.). It is assumed that this information represents a normal year.

4) Quality of the diet: Frequently, food security does not result from inadequate quantities
of food, but from a lack of protein, vitamins, and minerals. This indicator is a proxy
variable of diet quality in two ways. First of all, vegetable oil is a major source of calorie
intake in the diet; second, in Bangladeshi homes, oil is used to fry protein and vitamin-
based foods. Since soybean oil is more expensive and nutritious than mustard oil, the
consumption of both is measured, under the assumption that a family consuming larger
amounts of soybean oil (per capita) has a higher quality diet—thus more food secure—
relative to households with a lesser per capita oil consumption.

5) Diet diversity by type of household member: This indicator records the number of
days in which meat, fish, eggs, and dhal were consumed during the past month or week
by four categories of households members—male adults, female adults, male children,
and female children. This indicator is, in effect, a proxy variable that measures both food
utilization, particularly the intra-household distribution of food, and quality of diet. The
underlying assumption is that the household with the greater diversity of diet and more
equal distribution of food among individual members can be judged more food security
relative to a household with less diet diversity or with certain categories of members
regularly excluded from the consumption of quality foods. Information for this indicator,
as in the case of oil consumption, is measured in a standard questionnaire format.

5.2. Nutritional Security

Nutritional security is a livelihood outcome closely related to food security, particularly the
food utilization component. The conventional components of nutritional security are child
and maternal nutritional status, since these are two of the most vulnerable groups in
Bangladesh (and elsewhere in the world) and because of the long-lasting damage that even
temporary malnutrition can cause in child-bearing women and children. The indicators that
measure nutritional security in women and children are well-known and widely-accepted
(Table 2).

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 4

6) Stunting and wasting among children 6-59 months of age: This indicator is the
commonly used measure of moderate and severe malnutrition in a population. Children
whose ages fall within the range of 6-59 months are weighed and measured for height.
The three variables—weight, height, and age—are then used to create ratios of height to
age (stunting) and weight to height (wasting), which are compared with a reference
population to establish the incidence of severe or moderate malnutrition. Weight and
height are recorded using specialized equipment, and field staff require substantial
training. In some parts of Bangladesh, the age of children may not be readily known.

7) Body mass index for women of reproductive age: This indictor is equally robust and
widely used in measure maternal nutritional status for women between the ages of 14
and 50 (estimated reproductive age). Height and body weight are used to calculate a
body mass index that identifies severely underweight (i.e., malnourished) mothers. This
indicator also requires specialized equipment and trained field staff.

Table 1. Food Security Indicators

Indicator Justification Means of Measurement Assessment

1. Duration of the Food security inversely Table within a standard Need to specify
lean period (in related to length of questionnaire format in the situation of a
months/year) lean period a household survey. “normal year”
2. Meals/day and Measures severity of HH survey: table above Need to specify
quantity/meal the lean period has column for # meals the situation of a
during the lean per day “normal year”
period
3. Share Share of expenditure HH survey: Inquire into Proxy variable;
expenditure on on food less in food shares of major need to specify
food items security household expenditure items normal year
4. Oil Per capita oil Per capita consumption Proxy variable
Consumption consumption closely for soybean and
related to quality of mustard oil
diet
5. Diet diversity Diversity a measure of Per month consumption Proxy variable;
diet quality and intra- of meat, fish, eggs, and need to adjust to
household utilization dhal per adult and recall capacity
children, male and
female

Table 2. Indicators of Nutritional Security

Indicator Justification Means of Measurement Assessment

6. Incidence of Effect measure of Nutritional component in Robust, well-
stunting and severe and moderate HH survey: accepted
wasting among malnutrition in children measurement of height, indicator;
children 6-59 weight, age for all requires
months children in age bracket specialized
equipment and
trained field staff
7. Body mass Effective measure of Measurement of weight Widely-used;
index for women maternal malnutrition and height for all women requires
of reproductive between 14 and 50 specialized
age years of age equipment and
trained field staff

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 5

5.3. Economic Security

Although economic security is intimately related to household livelihood security, the

economic status of poor households is notoriously difficult to measure directly. Household
income among poor families is often derived from multiple informal sources, and labor is
sometimes compensated in non-monetary units (such as food). Wage work itself is irregular
and sporadic, making wage income difficult to remember and measure. Moreover, income
from agricultural sales is equally resistant to regular accounting. To evaluate economic
security, three components of economic security are addressed.

First, annual household income, if available, is a valuable measure of economic security. In

most cases, however, the sales of agricultural and animal products occur over multiple
events; wage labor is similarly irregular; and because households must struggle to obtain
cash in many different ways, diversity of income sources is an uncertain measure of security.
Second, the value of key household assets can serve an effect proxy variable. In rural
Bangladesh, the ownership of land and livestock tend to be effective measures of the
comparative economic status of households. But even if land and livestock are not
considered, there are key durable consumer goods, such as radio, jewelry, and bicycles
which reflect economic differences in a robust way. Third, levels of savings and debt can
effectively establish the economic status of households. Within these components of
economic security, the following indicators were chosen (Table 3).

8) Annual household income stream: This indicator assumes that the more economic
secure household will have a greater annual cash income stream. It is acknowledged
that the actual overall income stream will tend to be underestimated due to several
reasons, including recall, but that the error will be randomly distributed across the
sample. Thus, small changes in this indicator through time cannot be interpreted as
significant change in the economic status. This indicator is measured as part of a
household survey, and the data should include the estimated cash income earned by
every member of the household. When one member has engaged in several wage-
earning episodes, the annual value of each should be computed. Only the value of
agricultural and livestock sales, not consumed products, should enter into the
calculation. Questionnaire tables organized by income-earning episode throughout the
year have proven to be an effective measurement strategy.

9) Household asset index: The assumption underlying this indicator is that households
with a greater investment in key consumer durables are more economically secure, i.e.,
they have access to more income. The set of key assets can change from one rural
context to another, but generally it includes means of transportation, agricultural
equipment, fishing equipment, televisions, radios, sewing machines, jewelry, etc. The
final composition of the asset list should reflect distinct consumer preferences for items
that are expensive enough that not all households can obtain them. Once the list is
compiled, a monetary unit values is attributed to each of the assets, then the index is
calculated as the total value of all assets owned by the household. It has been argued
that the value of land and livestock should be included in this calculation; however, in a
society where many households are landless, the high value of land (and livestock) may
overwhelm the overall index, creating a bimodal distribution of index values between
landed and landless. Household asset lists can be gathered as part of a household
survey.1

1
This component of economic security has its inherent dangers, since the actual value of any asset
cannot always be taken into consideration without burdensome data collection (e.g., new bicycle vs.
used bicycle; color TV vs. black-and-white TV).

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 6

10) Household debt levels: Household debt is an important indicator of economic security
only when compared to other household characteristics. The level of debt for one
household may represent an important investment in rural infrastructure, such as the
purchase of land or equipment; the same level of debt for another household may
indicate an excessive financial burden and a situation of dependency. Debt is an
indicator that is easily measured in a survey format, but it should be calculated against a
comparative base, such as a debt/asset ratio or a debt/annual income level.

11) Household savings levels: The level of savings of a household is an effective

indicator of economic resiliency to contextual shocks and stresses. The households with
higher levels of savings demonstrate a higher level of economic security. Savings
information is normally gathered in a household survey.

Table 3. Indicators of Economic Security

Indicator Justification Means of Measurement Assessment

8. Annual Cash income levels In a household survey, Results tend to be
household income differentiate economic use a table of earning indicate true
stream status episodes by member relative income
positions of
households
9. HH Presence of key Compile assets lists in Land and
assets*index durable consumer survey, attribute a unit livestock values
goods reflects value to each asset can also be
economic status included
10. HH debt Can reflect access to Debt levels can be Debt levels must
levels financial capital or gathered in household have a base of
excessive burden surveys comparison; debt
to income; debt to
assets
11. HH savings Savings reflects Household survey: Robust measure
levels resiliency current savings of livelihood
vulnerability

5.4. Shelter/Water and Sanitation Security

This livelihood category is considered critical in the context of Bangladesh due to the high
population density, the lack of sanitation infrastructure, particularly in rural areas, and the
high levels of arsenic intrusion in wells. The components of this category include the quality
of housing, access to latrines, and access to drinking water that is safe from both bio-
contamination and arsenic (Table 4).

12) Housing condition: This indicator is measured by a proxy variable: the type of roofing
material used in the residence. More permanent roofing, such as zinc sheeting, reflects
a better standard of housing relative to grass roofing materials. Another aspect of this
indicator is the presence of electricity in the house.

13) WATSAN infrastructure: This is a key indicator that includes the type of latrine used
by the household, ranging from no latrine (open defecation) and hanging latrines
(unsanitary) to covered pit latrines; and the type of drinking water system, ranging from
an open water body (e.g. river or pond) to a community well or an individual deep
tubewell. Other aspects of indicator quality include the distance to the water source and
the number of families that share the source. Finally, the indicator captures whether the
water source is arsenic-free.

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 7

Table 4. Shelter and WATSAN Security

Indicator Justification Means of Measurement Assessment

12. Housing Investment in housing Data from household Proxy variables
condition us a robust indicator of survey on type of
well-being roofing and access to
electricity
13. WATSAN Water and sanitation Household survey data Some of these
infrastructure are critical livelihood measures type of latrine indicator
components in use, type of water characteristics
Bangladesh: source and whether it is available from
population density; safe from arsenic community-level
arsenic contamination analysis

5.5. Health Security

Several components of health security are considered to be critical in livelihood security

assessment. The first is the frequency of illness among all household members. In highly
vulnerable households, illness episodes can severely compromise the productiveness of
family members, reducing already-low levels of incomes and production, thereby affecting
food and nutritional security. The second component is access to primary health care. The
health security of rural families is directly related to their level of access to appropriate
medical care. In Bangladesh, a third component of health security is the incidence
specifically of diarrhea episodes, particularly among children. Diarrhea is, in fact, a proxy
variable of the quality of the health and sanitation environment in which the family resides,
and it is highly responsive to development interventions (e.g., ORS treatments), thus
eminently treatable. The indicators of health security are summarized in Table

14) Family illness episodes over last month: This indicator measures the number of
illness episodes over the last month, recording the type of illness, days sick, days of
productivity lost, and type of treatment sought—for every member of the household.
This information is gathered during the household survey. It is a strong indicator of
health security, if it can be assumed that illness does not have a strong seasonal pattern.
If so, then the period of recall will have to be expanded.

15) Incidence of diarrhea episodes over last month: this indicator is also gathered during
the household survey, and it includes detailed information on the length of sickness, the
days lost to work (if applicable), and the form of treatment.

Table 5. Health Security Indicators

Indicator Justification Means of Measurement Assessment

14. Family illness Detailed information on Household survey Captures age and
episodes over the health situation of the gathers data on each gender
last year household by member episode during the last differences; time
month, including work frame must be
days lost, treatment altered if there is
source, cost a seasonal illness
pattern
15. Incidence of Diarrhea a “signal” Household survey Captures age and
diarrhea condition of poor provides information on gender
health situation; each episode by differences; time
responsive to member frame must be
treatment altered if there is
a seasonal illness
pattern

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 8

5.6. Educational Security

This livelihood category is comprised of several components, including the overall level of
education of the household, gender differences in educational access, and the overall
literacy rates of adults in the household. Workshop participants acknowledged that
educational quality, while important, is a component that is generally overlooked in livelihood
assessments, usually due to the difficulty of measurement. The education security
indicators are summarized in Table 6.

16) Family members with completed primary: This indicator, combined with the
following one, is a measure of the amount of education in the household. The highest
educational level for this indicator is the situation in which all members within the
appropriate age category have completed primary school. This information is
gathered in the demographic section of the household survey questionnaire.

17) Family members with completed secondary: This indicator, when taken with the
first, provides a measure of the level of education within the household. Again, the
highest educational level for this indicator is all members of the appropriate age
category having completed secondary school.

18) Adult literacy rates: This indicator is often used to measure the lowest level of
educational achievement—adult illiteracy. This information is gathered in the
demographic section of a household survey questionnaire.

Table 6. Education Security Indicators

Indicator Justification Means of Measurement Assessment

16. Children Measure of the level of Survey of household Assumes no 15
completed primary education within the members over 15 years year olds in
level of school household of age with primary primary school,
complete does not measure
drop-outs;
captures gender
differences
17. Children Measure of the level of Survey of household Assumes no 20
completed education within the members over 20 years year olds in
secondary level of household of age with secondary secondary school,
school complete does not measure
drop-outs;
captures gender
differnces
18. Adult literacy Illiteracy often a Survey of all adults Strong negative
measure of the lowest indicator
level of security

5.7. Gender Status

One of the critical categories of household livelihood security in Bangladesh is the status of
women. Gender status in Bangladesh is part of the livelihood focus on basic rights and
justice, since women have been traditionally cloistered and their movement in society highly
restricted. Moreover, women have been the victims of structural oppression, including
violence. The major components that reflect gender status include the incidence of violence
against women, the participation of women in household decision-making, marriage age for
girls, and dowry levels. The relevant indicators are presented in Table 7.

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 9

19) Violence against women reduced: Structural violence against women, especially
husbands against wives, is a critical problem in Bangladeshi rural society. The livelihood
approach places great emphasis on justice and physical security, and reduction in
violence is a strong livelihood goal. Measurement of this highly sensitive indicator is
very difficult, just as the interventions designed to reduce violence toward females
produce only gradual change. Household surveys are not the appropriate strategy of
investigating violence against women, and it is suggested that tools of community
analysis be employed. These include focus group discussions with men and women,
key informant interviews with salish members, local political leaders, and local NGOs.

20) Female participation in household decision-making: It is assumed that as women

obtain more voice with households (and within village society) that their status will
improve. This indicator must be measured through a series of proxy variables, such as
questions regarding participation in specific decision contexts (e.g., land purchase,
migration, marriage of children, dowry, etc.) or questions regarding freedom of
movement to particular places (market, relative’s home, etc.).

21) Age at marriage: Traditionally women are obliged to marry very young, often forcing
them to abandon their studies. This indicator compares the ages at marriage of the adult
females in the household with the ages at marriage of their sons and daughters.
Indications of a pattern toward marriage at later age are assumed to represent an
improved status for women.

22) Dowry: Dowry is considered to be a symbol of the oppression of women and is often
associated with violence against young wives. The absence of dowry at marriage and
symbolically low values of dowry are considered indicators of improved status for
women. Again, the dowries of the adult females in the household are compared with
those of their sons and daughters.

Table 7. Gender Status Indicators

Indicator Justification Means of Measurement Assessment

19. VAW Violence against Community analysis: Necessary to do
decreased women widespread, FGDs, KIIs, salish some background
directly related to interviews—attempt to work on conflict
status measure quantities of resolution
violence episodes institutions:
salish, mosque
committee, etc.
20. Participation Status related to Household survey Interview must be
in household participation household (female respondents) on done in non-
decision-making decisions, village mobility of women, threatening
society participation in decisions context
21. Marriage age Status related to early Household survey of Adult women
marriage marriage ages compared to ages
of married
children
22. Dowry Dowry a critical social Household survey of Compares adult
incidence issue, associated with dowry incidence and women with
female violence levels married children

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 10

5.8 Community Participation

This livelihood outcome category focuses on the level of participation of households in wider
village society. In essence, this category attempts to assess the flows of social capital within
a village and how individual households are able to mobilize and access these networks. In
Bangladesh, village society is organized around traditional and formal social groups. In the
former category are such institutions as local samities, mosque committees, and other
informal associations. The more formal groups include the union parishad, NGOs, and
CBOs. It is assumed here that household livelihood security is enhanced by the density of
social relations, i.e., the amount of social capital available to households, especially the
vulnerable ones. While fairly subtle and elusive, the following indicators (in Table 8) are
designed to capture this livelihood dimension.

23) Effective presence of village groups: This indicator seeks to measure the amount of
nature of social relations in the village as determined by the number and type of social
groupings. The roles of the respective groups are also assessed. It is assumed that a
village with a more numerous and diverse set of social groups will enjoy higher levels of
social capital.

24) Participation of vulnerable households in community activities: As a complement

to the previous indicator, this one focuses on the inclusiveness of the village groups. It
seeks to assess if vulnerable households (such as female-headed or lower-caste
households) actively participate and benefit from participation. Both these indicators are
measured through a combination of community-level inquiry (FGDs and other qualitative
tools) and household-level inquiry. Furthermore, in the household surveys, men and
women are interviewed separately in order to capture gender differences in participation.

Table 8. Community Participation Indicators

Indicator Justification Means of Measurement Assessment

23. Effective Active local groups Household survey Community
presence of increase level of determines household analysis can be
community groups community solidarity knowledge of social done on pre-
groups. FGDs with survey visit
community leaders
another important
source of data

24. Vulnerable Indicator of Household survey asks Open-ended

households inclusiveness in village about level of questions in
participate in society participation, benefits of survey form
social activities the different groups

5.9 Access to Institutions

Household livelihood well-being is influenced by extent to which a household is integrated

into a wider socio-political system. In Bangladesh, most rural villages are serviced by
various types of external agents, who represent wider public or private institutions. For
example, there are government representatives that serve local populations, including health
agents, rural extensionists, local NGO staff, school teachers and so forth. The access of
individual households (and within households of men and women) to such external services
have an impact on livelihood security, and it is assumed that livelihood well-being is
enhanced by more effective access patterns. Also, where a larger number of institutions are
present, the resource flow to the village is assumed to be greater. The relevant indicators
are summarized in Table 9.

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 11

25) Uses of external services: This indicator measures the number of external agents in
the village and the use of their services by individual households. It includes government
offices, NGOs, and, in some cases, representatives of private institutions. This
information is gathered through the household survey.

26) Evaluation of external services: This indicator elicits the household’s evaluation of the
value of the external service. If the previous indicator seeks to evaluate the intensity of
the external presence in the village, this indicator seeks to assess the impact of this
presence, from the household perspective. Men and women in the households are
interviewed separately for these indicators.

Table 9. Access to Institutions Indicators

Indicator Justification Means of Measurement Assessment

25. Use of Level of integration in Household survey elicits Open-ended
external services wider system a source from men and women questions
of resources their use of these
services
26. Impacts of Level of integration in Household survey elicits Open-ended
external services wider system a source evaluations of impacts questions
of resources from both men and
women

6.0 Concluding Observations

The result of the livelihood indictors workshop demonstrated that a consensus around a core
set of livelihood indicators is possible. It further reinforced the argument that the holistic
household livelihood framework is a useful, integrative one for assessing change among the
target populations where development programs are operative. The list of 26 indicators that
was generated during the workshop is comprehensive and realistic, but it reveals many the
gaps that still challenge our understanding of household dynamics and well-being. For
example, the dynamic discussions that characterized the workshop underlined how difficult it
is to measure some of the more subtle forms of livelihood change, such as the status of
women or the participation of vulnerable households. One might wish that these concepts
could be measured as easily as a baby is weighed, but it would be wrong to assume that
difficulty of measurement makes the indicator less important or less insightful. To measure
the process of change is a daunting order, especially the type of change that can slip by
unperceived. The success of this workshop notwithstanding, there is still much work to be
done in refining the indicators and in adapting appropriate and effective methodologies.

A second gap that the workshop could not address was that of the relationship between
different livelihood outcome categories and indicators. The group did not have the time to
ask if economic security always meant food security, or health security. How these
component parts interrelate is ultimately a critical piece of the livelihood puzzle. At some
point in time, the LMP team will need to address how the livelihood components reinforce
one another. Does increasing women’s economic security change their status?

Finally, this final set of indicators, while robust and refined, should be considered as
prototypes, waiting to be tested and ready to undergo another round of modifications. As
livelihoods change, so will the set of indicators have to adjust. For now, they are an
improved version, not yet perfect, but one that can be distributed and set to the test.

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 12

Figure 1. HOUSEHOLD LIVELIHOOD SECURITY: A FRAMEWORK FOR ANALYSIS

Natural ASSETS
Resources Rights to:
Natural Capital Human Capital Social Capital Economic Capital
Institutions Political Capital Food Security

Infrastructure Nutritional
Security
History Production
& Health Security
Economic, Income Consumption
Cultural, and Activities activities Shelter Security
Political HOUSEHOLD
Environment WatSan Security

Demography Education
Security

Community
Processing, Participation
SHOCKS Exchange,
& Marketing Gender equity
STRESSES Activities
Access to
Institutions

CONTEXT LIVELIHOOD STRATEGY LIVELIHOOD OUTCOMES

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 13

Annex A: Comparison of Indicators Used by CARE-Bangladesh and Other Organisation

Indicators CARE indicators Indicators used by other organization

A. Nutritional Security • % children of 6-59 months old < -2 standard • % children with low birth rate (UNICEF)
deviation from the mean (stunting, wasting) • % HHs consuming Vitamin A, iodized salt
• % women have BMI of at least 18.5 (UNICEF)
• Appropriate breast feeding and weaning • Per capita food intake (Kcal) (PROSHIKA)
practices for new-borns. • % cereal, roots, tubers as part of dietary energy
supply (CFS)

B. Food Security • # meals per day during lean period • HHs with homestead gardens (HKI)
• Duration of lean period (# months) • Dependency ratio (HKI)
• Consumption of cooking oil; consumption of • Cropping intensity (FAO)
staple • Wage rate variability (FAO)
• % HHs increased food intake (# meals with • Gross crop and livestock income (HKI)
dietary diversity) • Net value of crop and livestock production
• Intra-HH allocation improved to meet specific • Crop handling and storage losses
needs of women and children • Income diversity
• % income used to procure food decreases • Household asset index
• Debt-to-asset ratio
• Household expenditures on specific items
• Dietary diversity
• Number of daily eating occasions
• Farming system diversity (HKI)
• Months of adequate food stock
• Perceived food security
• Coping strategies index (HKI)

C. Economic Security • Average #/% of women income earners per • % HHs with manual labor as principal
household occupation (HKI)
• % HHs using loans for business enterprise • Wage levels for day labor (ag and non-ag)
activities (CEGIS)

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 14

 Indicators CARE indicators Indicators used by other organization
• % women with adequate income equal to wage • Amount cultivated land/land ownership (HKI)
or ag labor index • Level of outmigration to bosti (HKI)
• % women with accumulated savings/ investment • Male out-migration (CEGIS)
above mandatory savings • %HHs with less than $1/day income (UNICEF)
• % HHs with increasing assets • HHs dependent on manual labor (HKI)
• % HHs (at least 10% of those headed by • Income from agricultural sales
women) with increased monthly cash income • Less dependence on mohajon loans
• Increased level of savings used for productive • #HH working days per year
activities rather than for crisis management (FAO/CEGIS/SIFAR)
• % increase in the value of key household assets
and services purchased
• # vulnerable HHs identified several alternative
livelihood options (income diversity)
• Source(s) of cooking fuel

D. Health Security • Annual average HH expenditures on health • Level of outmigration to bosti (HKI)
decreased • Male out-migration (CEGIS)
• # work days lost to illness among HHs • %HHs with less than $1/day income (UNICEF)
• % mothers with increase in antenatal nutrition
• #/% TB cases with access to facilities
• % decrease in general diseases
• % decrease in neonatal mortality, morbidity
rates
• % decrease in maternal mortality, morbidity
rates
• # children with access to immunisation

E. WATSAN Secuirty • % HHs using functional latrines • % HHs affected by salinity (CEGIS)
• % children under five with at least one episode • % HHs affected by arsenic (CEGIS)
of diarrhoea in last two weeks
• % incidence of diarrhoea among workers
• % HHs with access to safe water (drinking,

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 15

 Indicators CARE indicators Indicators used by other organization
cooking, bathing and washing)

F. Education Security • % women with basic numeracy skill • % children enrolled in primary
• % adults literate • % children enrolled in secondary
• Increased access to improved learning • Adult literacy rates (CEGIS)
environment for students
• # children graduated basic education (girls and
boys)
• # SMC functioning and playing active role to
ensure quality learning

G. Shelter • Housing: roof materials, wall materials, floor

materials, number of living rooms
• Access to electricity -------------------
• % HHs with shelter compatible to climate
change

H. Community participation • % vulnerable households participating in • NGO membership

Community Resource Management Committee • Membership in local groups (savings, gardening,
(CRMC) activitie etc.) (HKI)
• % women engaged in community organizations
• % women invited to more social (community)
events and gatherings
• # community groups managing development
projects
• # mother groups with greater role for community
development beyond the school

I. Access to Institution/ • # extreme poor with increased access to

service savings services
• # women with complications received EmOC

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 16

 Indicators CARE indicators Indicators used by other organization
services at available health facilities
• # of community groups that dialogue with
Upazila, UP and District level GoB, and other -------------------------
public institutions
• Access to and use of government and other
services
• # NGOs active in the community

J. Justice • Increased access to local services (law • Improved bargaining for wage rates
enforcement, psychological, health and legal) for • Access to khas and other natural resources
VAW victims

K. Gender • Women’s access to institutions and services • Marriage and dowry practices
Increased • Women participate in decisions over HH
• Women involved in decision making regarding investments
marriage, education, mobility, and control
related to their children and assets at HH level
• % decrease in violence against women (VAW)
• Mobility of women expanded
• Increase in female leadership in VAW forum,
shalish UP in upholding women rights

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 17

Annex B: Participant’s List

Livelihood Indicator Workshop

Rural Livelihood Program
CARE-Bangladesh

Serial Organization Name & address

Number
1 HKI Ms Gudrun Stallkamp
Nutrition program officer, Hellen Keller International
House #340 Road 5 DOHS
Baridhara , Dhaka
2 DFID Gerard Hendriksen
DFID, Bangladesh
3 PDO-ICZM Md. Sayed Iftekhar
Program Development Office
Integrated Coastal zone management program
Saimon Centre House # 4/A Road # 22,
Gulshan 1 Dhaka 1212

4 CEGIS Atiq Kainan Ahmed, Anthropologist

House 6, Road 23/C
Gulshan 1, Dhaka 1212
5 FAO Julia Wolf
Rural Institution and Participation service
FAO, Rome Italy
6 RDRS Sattya Roy
PM (Fish)
Rangpur Dinajpur Rural Service (RDRS),
Dhap, Rangpur
7 CBSG M. Obaidul Islam
3/3 Block –B, Lalmatia
Dhaka 1207
8 DFID Fourth Niaz Ahmed Apu
Fish Social-Economist, Fourth Fisheries Project
Fisheries Department of Fisheries Matshya Bhaban
Ramna, Dhaka-1000

9 RLEP Esha Husain

Rural Livelihoods Evaluation Partnership
House # 10, Road # 135
Gulshan –1
Dhaka 1212
10 World Fish A.K.M Firoz Khan
Centre Research Associate World Fish Centre
Road # 22 B House # 7, Block- F, Banani,
Dhaka 1213
11 Action Aid Partha Hefaz Sheikh
Action Aid Bangladesh

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 18

 Serial Organization Name & address
Number
12 CARE Dr. Ziya Uddin Ahmed
Bangladesh Health Sector
13 Dr. Syed Asif Altaf
HIV Coordinator, NSDP
14 Jagannath K Dutta
Senior Advisor D M& E, CBHQ
15 M. Rafiqul Islam (TC_ M&E)
16 Gender Coordinator, CBHQ
17 Loretta Payne,
RLP Coordinator
18 Sarah Gillingham,
Livelihood Advisor , RLP
19 Claudia Schaerer
PC-RVCC
20 H J M Kamal
PC-RLP
21 Sajeda Begum
PC-LIFT
22 Abdus Sobhan
PC-RLP
23 Abdul Awal
PC, RLP
24 Abdul Wadud
25 M. Zakaria
TC-LMU
26 Mehrul Islam
PC, RLP
27 ABDUL Wadud
PM-RLP
28 Faruk Ahmed
PDO-INCOME
29 TANGO Tim Finan
TANGO Consultant

Measuring Livelihood Impacts: A Review of Livelihoods Indicators 19